summary refs log tree commit diff
path: root/www
diff options
context:
space:
mode:
Diffstat (limited to 'www')
-rw-r--r--www/causal.agency/.gitignore4
-rw-r--r--www/causal.agency/Makefile23
-rw-r--r--www/causal.agency/alpha.html92
-rw-r--r--www/causal.agency/index.774
-rw-r--r--www/causal.agency/lands.html176
-rw-r--r--www/causal.agency/style.css28
-rw-r--r--www/git.causal.agency/.gitignore13
-rw-r--r--www/git.causal.agency/Makefile53
-rw-r--r--www/git.causal.agency/cgitrc30
-rw-r--r--www/git.causal.agency/custom.css86
-rw-r--r--www/git.causal.agency/filter.c158
-rw-r--r--www/git.causal.agency/index.781
-rw-r--r--www/photo.causal.agency/.gitignore2
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0832.txt6
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0850.txt6
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0852.txt4
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0858.txt6
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0859.txt6
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0865.txt2
-rw-r--r--www/photo.causal.agency/2024-04-10/IMG_0890.txt9
-rw-r--r--www/photo.causal.agency/2024-04-14/IMG_1054.txt5
-rw-r--r--www/photo.causal.agency/2024-04-14/IMG_1058.txt6
-rw-r--r--www/photo.causal.agency/2024-04-14/IMG_1066.txt10
-rw-r--r--www/photo.causal.agency/generate.sh211
-rw-r--r--www/photo.causal.agency/rsync.sh5
-rw-r--r--www/temp.causal.agency/.gitignore1
-rw-r--r--www/temp.causal.agency/Makefile15
-rw-r--r--www/temp.causal.agency/up.c193
-rw-r--r--www/text.causal.agency/.gitignore4
-rw-r--r--www/text.causal.agency/001-make.7159
-rw-r--r--www/text.causal.agency/002-writing-mdoc.7138
-rw-r--r--www/text.causal.agency/003-pleasant-c.7120
-rw-r--r--www/text.causal.agency/004-uloc.764
-rw-r--r--www/text.causal.agency/005-testing-c.773
-rw-r--r--www/text.causal.agency/006-some-libs.796
-rw-r--r--www/text.causal.agency/007-cgit-setup.7271
-rw-r--r--www/text.causal.agency/008-how-irc.7193
-rw-r--r--www/text.causal.agency/009-casual-update.7127
-rw-r--r--www/text.causal.agency/010-irc-suite.7409
-rw-r--r--www/text.causal.agency/011-libretls.7220
-rw-r--r--www/text.causal.agency/012-inability.739
-rw-r--r--www/text.causal.agency/013-hot-tips.7156
-rw-r--r--www/text.causal.agency/014-using-vi.7135
-rw-r--r--www/text.causal.agency/015-reusing-tags.7155
-rw-r--r--www/text.causal.agency/016-using-openbsd.7505
-rw-r--r--www/text.causal.agency/017-unpasswords.7153
-rw-r--r--www/text.causal.agency/018-operating-systems.786
-rw-r--r--www/text.causal.agency/019-mailing-list.7286
-rw-r--r--www/text.causal.agency/020-c-style.7172
-rw-r--r--www/text.causal.agency/021-time-machine.7144
-rw-r--r--www/text.causal.agency/022-swans-are-dead.7164
-rw-r--r--www/text.causal.agency/023-sparse-checkout.7144
-rw-r--r--www/text.causal.agency/024-seprintf.7137
-rw-r--r--www/text.causal.agency/025-v6-pwd.7330
-rw-r--r--www/text.causal.agency/026-git-comment.7190
-rw-r--r--www/text.causal.agency/027-openbsd-linode.7202
-rw-r--r--www/text.causal.agency/028-names.781
-rw-r--r--www/text.causal.agency/029-topics.7116
-rw-r--r--www/text.causal.agency/030-discs.7114
-rw-r--r--www/text.causal.agency/031-books-2021.7127
-rw-r--r--www/text.causal.agency/032-albums-2021.7173
-rw-r--r--www/text.causal.agency/033-jorts.7485
-rw-r--r--www/text.causal.agency/034-voices.756
-rw-r--r--www/text.causal.agency/035-addendum-2021.7111
-rw-r--r--www/text.causal.agency/036-compassion.7105
-rw-r--r--www/text.causal.agency/037-care.7167
-rw-r--r--www/text.causal.agency/038-agency.785
-rw-r--r--www/text.causal.agency/039-apologies.781
-rw-r--r--www/text.causal.agency/040-sound-memory.7165
-rw-r--r--www/text.causal.agency/041-albums-2022.7185
-rw-r--r--www/text.causal.agency/042-comfort-music.762
-rw-r--r--www/text.causal.agency/043-little-blessings.778
-rw-r--r--www/text.causal.agency/Makefile66
-rw-r--r--www/text.causal.agency/colb.c16
-rw-r--r--www/text.causal.agency/feed.sh58
75 files changed, 8508 insertions, 0 deletions
diff --git a/www/causal.agency/.gitignore b/www/causal.agency/.gitignore
new file mode 100644
index 00000000..b00b1c3c
--- /dev/null
+++ b/www/causal.agency/.gitignore
@@ -0,0 +1,4 @@
+index.html
+leveler.html
+scheme.css
+scheme.png
diff --git a/www/causal.agency/Makefile b/www/causal.agency/Makefile
new file mode 100644
index 00000000..8c74f8f1
--- /dev/null
+++ b/www/causal.agency/Makefile
@@ -0,0 +1,23 @@
+WEBROOT = /var/www/causal.agency
+
+GEN = index.html scheme.css scheme.png
+FILES = ${GEN} style.css alpha.html lands.html
+
+all: ${FILES}
+
+.SUFFIXES: .7 .html
+
+.7.html:
+	mandoc -T html -O style=style.css $< > $@
+
+scheme.css:
+	scheme -st > scheme.css
+
+scheme.png:
+	scheme -g > scheme.png
+
+install: ${FILES}
+	install -C -m 644 ${FILES} ${WEBROOT}
+
+clean:
+	rm -f ${GEN}
diff --git a/www/causal.agency/alpha.html b/www/causal.agency/alpha.html
new file mode 100644
index 00000000..0d83f530
--- /dev/null
+++ b/www/causal.agency/alpha.html
@@ -0,0 +1,92 @@
+<!DOCTYPE html>
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>all 26 letters of the alphabet RANKED</title>
+<style>
+body, button { font-size: 200%; text-align: center; }
+button { margin: 1em; padding: 1ch; }
+button#shuffle { font-size: 100%; }
+</style>
+
+which letter do you like more?
+<p>
+<button id="a">A</button>
+<button id="b">B</button>
+<p>
+<details>
+<summary>current ranking</summary>
+<p>
+<span id="ranking">ABCDEFGHIJKLMNOPQRSTUVWXYZ</span>
+<p>
+<button id="shuffle">reshuffle</button>
+</details>
+
+<script>
+let buttonA = document.getElementById("a");
+let buttonB = document.getElementById("b");
+let ranking = document.getElementById("ranking");
+
+let alpha = "ABCDEFGHIJKLMNOPQRSTUVWXYZ".split("");
+let rand = (bound) => Math.floor(Math.random() * bound);
+function shuffle() {
+	for (let i = alpha.length - 1; i > 0; --i) {
+		let j = rand(i + 1);
+		let x = alpha[i];
+		alpha[i] = alpha[j];
+		alpha[j] = x;
+	}
+}
+if (localStorage.getItem("alpha")) {
+	alpha = localStorage.getItem("alpha").split("");
+} else {
+	shuffle();
+}
+
+let index = 0;
+let even = true;
+function choose(o) {
+	if (o == "b") {
+		let x = alpha[index];
+		alpha[index] = alpha[index + 1];
+		alpha[index + 1] = x;
+	}
+	index += 2;
+	if (index > alpha.length - 2) {
+		even = !even;
+		index = (even ? 0 : 1);
+	}
+	update();
+}
+
+document.onkeydown = function(event) {
+	if (event.key.toUpperCase() == alpha[index]) {
+		choose("a");
+	} else if (event.key.toUpperCase() == alpha[index + 1]) {
+		choose("b");
+	}
+}
+
+function update() {
+	localStorage.setItem("alpha", alpha.join(""));
+	ranking.innerText = alpha.join("");
+	let a = buttonA;
+	let b = buttonB;
+	if (rand(2)) {
+		a = buttonB;
+		b = buttonA;
+	}
+	let lc = (c) => c;
+	if (rand(2)) lc = (c) => c.toLowerCase();
+	a.innerText = lc(alpha[index]);
+	b.innerText = lc(alpha[index + 1]);
+	a.onclick = () => choose("a");
+	b.onclick = () => choose("b");
+}
+update();
+
+document.getElementById("shuffle").onclick = function() {
+	if (confirm("Are you SURE you want to throw away all your hard work?")) {
+		shuffle();
+		update();
+	}
+}
+</script>
diff --git a/www/causal.agency/index.7 b/www/causal.agency/index.7
new file mode 100644
index 00000000..d29c1a6b
--- /dev/null
+++ b/www/causal.agency/index.7
@@ -0,0 +1,74 @@
+.Dd November 28, 2023
+.Dt CAUSAL.AGENCY 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm june
+.Nd computer enthusiast (she/her)
+.
+.Sh SYNOPSIS
+.Nm mail
+.Mt june@causal.agency
+.Nm
+in
+.Li #ascii.town
+on tilde.chat
+.
+.Sh DESCRIPTION
+I make mostly IRC software in C.
+I like
+.Ox
+but also the GPL.
+I just want to read books
+and try to learn to be kinder.
+When I can I'd like to talk to strangers
+and experience more magic.
+.
+.Pp
+.Lk https://git.causal.agency code
+\(em
+.Lk https://text.causal.agency words
+\(em
+.Lk /list/ mailist
+.
+.Pp
+These are some things I've done:
+.Bl -tag -width Ds
+.It Lk https://git.causal.agency/pounce/about pounce
+a multi-client-first IRC bouncer
+.It Lk https://git.causal.agency/catgirl/about catgirl
+a cosy IRC client
+.It Lk https://git.causal.agency/litterbox/about litterbox
+a full-text search IRC logger
+.It Lk https://git.causal.agency/scooper/about scooper
+a web interface for litterbox
+.It Lk https://git.causal.agency/kitd/about kitd
+a process supervisor
+.It Lk https://git.causal.agency/imbox/about "imbox & git-fetch-email"
+a tool to pull patches out of IMAP
+.It Lk https://git.causal.agency/bubger/about bubger
+a mailing list archive generator for IMAP
+.It Lk https://git.causal.agency/notemap/about notemap
+a tool to mirror text files to IMAP notes
+.It Lk https://ascii.town/explore.html torus@ascii.town
+a collaborative ASCII art project
+.It Lk ssh://play@ascii.town play@ascii.town
+some games to play over
+.Xr ssh 1
+.It Lk https://git.causal.agency/cards/about cards
+a
+.Pa CARDS.DLL
+loader for SDL
+.It Lk scheme.png scheme
+an earthy terminal colour scheme
+.El
+.
+.Sh SEE ALSO
+.Bl -bullet
+.It
+.Lk /bin/ bin
+.It
+.Lk lands.html "Magic lands quiz"
+.It
+.Lk alpha.html "alphabet ranking game"
+.El
diff --git a/www/causal.agency/lands.html b/www/causal.agency/lands.html
new file mode 100644
index 00000000..7aaadd80
--- /dev/null
+++ b/www/causal.agency/lands.html
@@ -0,0 +1,176 @@
+<!DOCTYPE html>
+<title>Lands Quiz</title>
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<style>
+html { font: 14pt sans-serif; line-height: 1.5em; }
+body { padding: 1em 1ch; max-width: 78ch; margin: auto; }
+h1 { text-align: center; }
+h2 { margin-top: 0; }
+button { font-size: 100%; padding: 0.5em 1ch; }
+img { max-width: 100%; }
+div.cols { display: grid; grid-template-columns: 1fr 1fr; gap: 2ch; }
+</style>
+
+<h1 id="loading">Loading...</h1>
+<h1 id="error" hidden>Failed to load cards :(</h1>
+
+<div id="game" hidden>
+<h1>Magic Lands Quiz</h1>
+<p>Try to guess the colours of mana each land produces!</p>
+<div class="cols">
+	<div>
+		<img id="back" src="https://backs.scryfall.io/normal/0/a/0aeebaf5-8c7d-4636-9e82-8c27447861f7.jpg">
+		<a id="link" target="_blank">
+			<img id="image1" hidden>
+			<img id="image2" hidden>
+		</a>
+	</div>
+	<div>
+		<h2 id="name"></h2>
+		<input type="checkbox" id="w"> <label for="w">White</label><br>
+		<input type="checkbox" id="u"> <label for="u">Blue</label><br>
+		<input type="checkbox" id="b"> <label for="b">Black</label><br>
+		<input type="checkbox" id="r"> <label for="r">Red</label><br>
+		<input type="checkbox" id="g"> <label for="g">Green</label><br>
+		<p><button id="submit">Submit</button></p>
+		<h3>Score: <span id="score">0</span>/<span id="total">0</span></h3>
+	</div>
+</div>
+</div>
+
+<script>
+function shuffle(arr) {
+	let rand = (bound) => Math.floor(Math.random() * bound);
+	for (let i = arr.length-1; i > 0; --i) {
+		let j = rand(i+1);
+		let x = arr[i];
+		arr[i] = arr[j];
+		arr[j] = x;
+	}
+}
+
+const CardBack =
+"https://backs.scryfall.io/normal/0/a/0aeebaf5-8c7d-4636-9e82-8c27447861f7.jpg";
+
+function hideCard() {
+	document.getElementById("back").hidden = false;
+	document.getElementById("image1").hidden = true;
+	document.getElementById("image2").hidden = true;
+}
+
+function showCard(card) {
+	document.getElementById("back").hidden = true;
+	document.getElementById("link").href = card.scryfall_uri;
+	let image1 = document.getElementById("image1");
+	let image2 = document.getElementById("image2");
+	if (card.card_faces) {
+		image1.src = card.card_faces[0].image_uris.normal;
+		image2.src = card.card_faces[1].image_uris.normal;
+		image1.hidden = false;
+		image2.hidden = false;
+	} else {
+		image1.src = card.image_uris.normal;
+		image1.hidden = false;
+	}
+}
+
+function resetChecks() {
+	for (let c of "wubrg") {
+		let input = document.getElementById(c);
+		input.checked = false;
+		input.disabled = false;
+		input.labels[0].style.fontWeight = "normal";
+	}
+}
+
+function checkChecks(card) {
+	let score = 0;
+	let total = 0;
+	let checked = 0;
+	for (let c of "wubrg") {
+		let input = document.getElementById(c);
+		let produced = card.produced_mana.includes(c.toUpperCase());
+		if (produced) {
+			total++;
+			input.labels[0].style.fontWeight = "bold";
+			if (input.checked) score++;
+		}
+		if (input.checked) checked++;
+		input.disabled = true;
+	}
+	if (checked > total) score -= (checked - total);
+	if (score < 0) score = 0;
+	return { score: score, total: total };
+}
+
+document.onkeydown = function(event) {
+	for (let c of "wubrg") {
+		if (event.key == c) {
+			let input = document.getElementById(c);
+			if (!input.disabled) input.checked ^= true;
+		}
+	}
+	if (event.key == "Enter") {
+		document.getElementById("submit").click();
+	}
+}
+
+let score = 0;
+let total = 0;
+let cards = [];
+let card = null;
+
+function nextCard() {
+	hideCard();
+	resetChecks();
+	card = cards.shift();
+	document.getElementById("name").innerText = card.name;
+}
+
+document.getElementById("submit").onclick = function() {
+	if (card) {
+		let { score: cardScore, total: cardTotal } = checkChecks(card);
+		total += cardTotal;
+		score += cardScore;
+		document.getElementById("score").innerText = score;
+		document.getElementById("total").innerText = total;
+		showCard(card);
+		card = null;
+		if (cards.length) {
+			this.innerText = "Next card";
+		} else {
+			this.disabled = true;
+			this.innerText = "No more cards";
+		}
+	} else {
+		nextCard();
+		this.innerText = "Submit";
+	}
+}
+
+function loadCards(resp) {
+	let loading = document.getElementById("loading");
+	let error = document.getElementById("error");
+	let game = document.getElementById("game");
+	if (resp.status != 200) {
+		loading.hidden = true;
+		error.hidden = false;
+	}
+	resp.json().then((json) => {
+		cards.push(...json.data);
+		if (json.has_more) {
+			setTimeout(() => fetch(json.next_page).then(loadCards), 50);
+		} else {
+			loading.hidden = true;
+			game.hidden = false;
+			shuffle(cards);
+			nextCard();
+		}
+	});
+}
+
+const Search =
+"https://api.scryfall.com/cards/search?q=t:land+id>=2+produces>=2+produces!=wubrg";
+fetch(Search).then(loadCards);
+
+</script>
diff --git a/www/causal.agency/style.css b/www/causal.agency/style.css
new file mode 100644
index 00000000..ee218533
--- /dev/null
+++ b/www/causal.agency/style.css
@@ -0,0 +1,28 @@
+@import url("scheme.css");
+
+table.head, table.foot { width: 100%; }
+td.head-rtitle, td.foot-os { text-align: right; }
+td.head-vol { text-align: center; }
+div.Pp { margin: 1ex 0ex; }
+div.Nd, div.Bf, div.Op { display: inline; }
+span.Pa, span.Ad { font-style: italic; }
+span.Ms { font-weight: bold; }
+dl.Bl-diag > dt { font-weight: bold; }
+code.Nm, code.Fl, code.Cm, code.Ic, code.In, code.Fd, code.Fn,
+code.Cd { font-weight: bold; font-family: inherit; }
+
+div.head, div.foot { display: flex; justify-content: space-between; }
+.head-ltitle, .foot-left { flex: 1; }
+.head-vol, .foot-date { flex: 0 1 auto; text-align: center; }
+.head-rtitle, .foot-os { flex: 1; text-align: right; }
+
+html { font-family: monospace; line-height: 1.25em; }
+body { max-width: 80ch; margin: 1em auto; padding: 0 1ch; }
+table { border-collapse: collapse; }
+table.Nm code.Nm { padding-right: 1ch; }
+table.foot { margin-top: 1em; }
+
+html { background-color: var(--ansi16); color: var(--ansi17); }
+a { color: var(--ansi4); }
+a:visited { color: var(--ansi5); }
+a.permalink { color: var(--ansi3); text-decoration: none; }
diff --git a/www/git.causal.agency/.gitignore b/www/git.causal.agency/.gitignore
new file mode 100644
index 00000000..eaed8039
--- /dev/null
+++ b/www/git.causal.agency/.gitignore
@@ -0,0 +1,13 @@
+*.html
+about-filter
+compress
+ctags
+email-filter
+filter
+gzip
+hilex
+htagml
+mandoc
+mtags
+owner-filter
+source-filter
diff --git a/www/git.causal.agency/Makefile b/www/git.causal.agency/Makefile
new file mode 100644
index 00000000..86b9f3eb
--- /dev/null
+++ b/www/git.causal.agency/Makefile
@@ -0,0 +1,53 @@
+PREFIX = /var/www
+CONFDIR = ${PREFIX}/conf
+DATADIR = ${PREFIX}/cgit
+BINDIR = ${PREFIX}/bin
+WEBROOT = ${PREIFX}/git.causal.agency
+
+CFLAGS += -Wall -Wextra
+LDFLAGS = -static -pie
+
+BINS += about-filter
+BINS += ctags
+BINS += email-filter
+BINS += gzip
+BINS += hilex
+BINS += htagml
+BINS += mandoc
+BINS += mtags
+BINS += owner-filter
+BINS += source-filter
+
+HTMLS = index.html
+
+all: ${BINS} ${HTMLS}
+
+compress ctags mandoc:
+	${MAKE} -C /usr/src/usr.bin/$@ LDFLAGS='${LDFLAGS}'
+	mv /usr/src/usr.bin/$@/$@ $@
+	${MAKE} -C /usr/src/usr.bin/$@ clean
+
+gzip: compress
+	ln -f compress $@
+
+hilex htagml mtags:
+	rm -f ../../bin/$@
+	${MAKE} -C ../../bin $@ LDFLAGS='${LDFLAGS}'
+	mv ../../bin/$@ $@
+
+about-filter email-filter owner-filter source-filter: filter
+	ln -f filter $@
+
+index.html: index.7
+	mandoc -Thtml -Ostyle=https://causal.agency/style.css index.7 >index.html
+
+install: cgitrc custom.css ${BINS}
+	install -m 644 cgitrc ${CONFDIR}
+	install -m 644 custom.css ${DATADIR}
+	install -d -o www -g daemon ${PREFIX}/cache/cgit
+	install -d -m 1700 -o www -g daemon ${PREFIX}/tmp
+	install -s ${BINS} ${BINDIR}
+	install -m 644 ${HTMLS} ${WEBROOT}
+
+clean:
+	rm -f compress filter ${BINS} ${HTMLS}
diff --git a/www/git.causal.agency/cgitrc b/www/git.causal.agency/cgitrc
new file mode 100644
index 00000000..0666fd28
--- /dev/null
+++ b/www/git.causal.agency/cgitrc
@@ -0,0 +1,30 @@
+root-title=causal agency
+root-desc=“I think some people from the Gentoo project are behind this.”
+logo=
+
+clone-url=https://$HTTP_HOST/$CGIT_REPO_URL
+snapshots=tar.gz zip
+
+enable-blame=1
+enable-commit-graph=1
+enable-subject-links=1
+enable-follow-links=1
+enable-index-owner=0
+repository-sort=age
+branch-sort=age
+
+css=/custom.css
+about-filter=/bin/about-filter
+source-filter=/bin/source-filter
+#owner-filter=/bin/owner-filter
+email-filter=/bin/email-filter
+
+readme=:README.7
+readme=:README
+
+remove-suffix=1
+enable-git-config=1
+scan-path=/git.causal.agency
+
+cache-root=/cache/cgit
+cache-size=1024
diff --git a/www/git.causal.agency/custom.css b/www/git.causal.agency/custom.css
new file mode 100644
index 00000000..b3f4f425
--- /dev/null
+++ b/www/git.causal.agency/custom.css
@@ -0,0 +1,86 @@
+/* This Source Code Form is subject to the terms of the Mozilla Public
+ * License, v. 2.0. If a copy of the MPL was not distributed with this
+ * file, You can obtain one at https://mozilla.org/MPL/2.0/. */
+
+@import url("cgit.css");
+
+* { line-height: 1.25em; }
+
+div#cgit {
+	max-width: 117ch;
+	margin: auto;
+	font-family: monospace;
+	-moz-tab-size: 4;
+	tab-size: 4;
+}
+
+div#cgit table#header td.sub {
+	border-top: none;
+}
+div#cgit table#header td.sub.right {
+	padding-right: 1em;
+}
+div#cgit table.tabs {
+	border-bottom: none;
+}
+div#cgit div.content {
+	border-bottom: none;
+}
+div#cgit table.list th a {
+	color: inherit;
+}
+div#cgit table.list tr:nth-child(even) {
+	background: inherit;
+}
+div#cgit table.list tr:hover {
+	background: inherit;
+}
+div#cgit table.list tr.nohover-highlight:hover:nth-child(even) {
+	background: inherit;
+}
+
+div#cgit table.blob td.linenumbers a:target {
+	color: goldenrod;
+	text-decoration: underline;
+	outline: none;
+}
+
+div#cgit div#summary {
+	max-width: 80ch;
+}
+
+/* for hilex(1) */
+div#cgit pre .Ke { color: dimgray; }
+div#cgit pre .Ma { color: green; }
+div#cgit pre .Co { color: navy; }
+div#cgit pre .St { color: teal; }
+div#cgit pre .Fo { color: teal; font-weight: bold; }
+div#cgit pre .Su { color: olive; }
+
+/* for htagml(1) */
+div#cgit pre a.tag { color: inherit; text-decoration: underline; }
+div#cgit pre a.tag:target { color: goldenrod; outline: none; }
+
+/* for mandoc(1) */
+table.head, table.foot { width: 100%; }
+td.head-rtitle, td.foot-os { text-align: right; }
+td.head-vol { text-align: center; }
+div.Pp { margin: 1ex 0ex; }
+div.Nd, div.Bf, div.Op { display: inline; }
+span.Pa, span.Ad { font-style: italic; }
+span.Ms { font-weight: bold; }
+dl.Bl-diag > dt { font-weight: bold; }
+code.Nm, code.Fl, code.Cm, code.Ic, code.In, code.Fd, code.Fn,
+code.Cd { font-weight: bold; font-family: inherit; }
+
+h1.Sh { font-size: 1.5em; }
+table.Nm td:first-child { padding-right: 1ch; }
+code.Fl { white-space: nowrap; }
+span.RsT { font-style: italic; }
+dl.Bl-tag:not(.Bl-compact) > dt { margin-top: 1em; }
+ul.Bl-bullet:not(.Bl-compact) > li { margin-top: 1em; }
+div.Bd-indent { margin-left: 4ch; }
+table.Bl-column { width: 100%; }
+table.foot { margin-top: 1em; }
+
+div#cgit a.permalink { color: inherit; }
diff --git a/www/git.causal.agency/filter.c b/www/git.causal.agency/filter.c
new file mode 100644
index 00000000..7c7e9320
--- /dev/null
+++ b/www/git.causal.agency/filter.c
@@ -0,0 +1,158 @@
+#include <err.h>
+#include <fcntl.h>
+#include <fnmatch.h>
+#include <limits.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/wait.h>
+#include <unistd.h>
+
+#define Q(...) #__VA_ARGS__
+
+#define MANDOC_OPTIONS "fragment,man=%N.%S,includes=../tree/%I"
+
+static int about(int argc, char *argv[]) {
+	if (argc < 2) return 1;
+	if (!fnmatch("README.[1-9]", argv[1], 0)) {
+		execlp("mandoc", "mandoc", "-T", "html", "-O", MANDOC_OPTIONS, NULL);
+		err(127, "mandoc");
+	} else if (!fnmatch("*.[1-9]", argv[1], 0)) {
+		execlp(
+			"mandoc", "mandoc", "-T", "html", "-O", "toc," MANDOC_OPTIONS, NULL
+		);
+		err(127, "mandoc");
+	} else {
+		execlp("hilex", "hilex", "-l", "text", "-f", "html", "-o", "pre", NULL);
+		err(127, "hilex");
+	}
+}
+
+static int email(void) {
+	size_t cap = 0;
+	char *buf = NULL;
+	if (getline(&buf, &cap, stdin) < 0) err(1, "getline");
+	if (buf[0] == 'C' && !strncmp(&buf[strcspn(buf, " ")], " McEnroe", 8)) {
+		printf("June%s", &buf[strcspn(buf, " ")]);
+	} else {
+		printf("%s", buf);
+	}
+	return 0;
+}
+
+static int owner(void) {
+	printf(Q(<a href="https://liberapay.com/june/donate"><img alt="Donate using Liberapay" src="https://liberapay.com/assets/widgets/donate.svg"></a>));
+	return 0;
+}
+
+#define CTAGS_PATTERN "*.[chlmy]"
+#define TEMPLATE "/tmp/filter.XXXXXXXXXX"
+
+static char tmp[PATH_MAX];
+static char tags[] = TEMPLATE;
+static void cleanup(void) {
+	unlink(tmp);
+	unlink(tags);
+}
+
+static int source(int argc, char *argv[]) {
+	if (argc < 2) return 1;
+	if (
+		strcmp("Makefile", argv[1]) &&
+		strcmp(".profile", argv[1]) &&
+		strcmp(".shrc", argv[1]) &&
+		fnmatch(CTAGS_PATTERN, argv[1], 0) &&
+		fnmatch("*.mk", argv[1], 0) &&
+		fnmatch("*.[1-9]", argv[1], 0) &&
+		fnmatch("*.sh", argv[1], 0)
+	) {
+		execlp("hilex", "hilex", "-t", "-n", argv[1], "-f", "html", NULL);
+		err(127, "hilex");
+	}
+
+	const char *ext = strrchr(argv[1], '.');
+	if (!strcmp(argv[1], ".profile") || !strcmp(argv[1], ".shrc")) {
+		ext = ".sh";
+	} else if (!strcmp(argv[1], "Makefile")) {
+		ext = ".mk";
+	} else if (!ext) {
+		ext = "";
+	}
+
+	snprintf(tmp, sizeof(tmp), TEMPLATE "%s", ext);
+	int fd = mkstemps(tmp, strlen(ext));
+	if (fd < 0) err(1, "%s", tmp);
+	atexit(cleanup);
+
+	char buf[4096];
+	for (ssize_t len; 0 < (len = read(STDIN_FILENO, buf, sizeof(buf)));) {
+		if (write(fd, buf, len) < 0) err(1, "%s", tmp);
+	}
+	if (close(fd) < 0) err(1, "%s", tmp);
+
+	fd = mkstemp(tags);
+	if (fd < 0) err(1, "%s", tags);
+	close(fd);
+	pid_t pid = fork();
+	if (pid < 0) err(1, "fork");
+	if (!pid) {
+		if (!fnmatch(CTAGS_PATTERN, argv[1], 0)) {
+			execlp("ctags", "ctags", "-w", "-f", tags, tmp, NULL);
+			warn("ctags");
+		} else {
+			execlp("mtags", "mtags", "-f", tags, tmp, NULL);
+			warn("mtags");
+		}
+		_exit(127);
+	}
+	int status;
+	if (wait(&status) < 0) err(1, "wait");
+
+	int rw[2];
+	if (pipe(rw) < 0) err(1, "pipe");
+	pid = fork();
+	if (pid < 0) err(1, "fork");
+	if (!pid) {
+		dup2(rw[1], STDOUT_FILENO);
+		close(rw[0]);
+		close(rw[1]);
+		execlp("hilex", "hilex", "-f", "html", tmp, NULL);
+		warn("hilex");
+		_exit(127);
+	}
+	pid = fork();
+	if (pid < 0) err(1, "fork");
+	if (!pid) {
+		dup2(rw[0], STDIN_FILENO);
+		close(rw[0]);
+		close(rw[1]);
+		execlp("htagml", "htagml", "-im", "-f", tags, tmp, NULL);
+		warn("htagml");
+		_exit(127);
+	}
+	close(rw[0]);
+	close(rw[1]);
+
+	if (wait(&status) < 0) err(1, "wait");
+	if (wait(&status) < 0) err(1, "wait");
+	return status;
+}
+
+int main(int argc, char *argv[]) {
+#ifdef __OpenBSD__
+	int error;
+	switch (getprogname()[0]) {
+		break; case 'a': error = pledge("stdio exec", NULL);
+		break; case 's': error = pledge("stdio tmppath proc exec", NULL);
+		break; default:  error = pledge("stdio", NULL);
+	}
+	if (error) err(1, "pledge");
+#endif
+	switch (getprogname()[0]) {
+		case 'a': return about(argc, argv);
+		case 'e': return email();
+		case 'o': return owner();
+		case 's': return source(argc, argv);
+		default: return 1;
+	}
+}
diff --git a/www/git.causal.agency/index.7 b/www/git.causal.agency/index.7
new file mode 100644
index 00000000..58a40dfe
--- /dev/null
+++ b/www/git.causal.agency/index.7
@@ -0,0 +1,81 @@
+.Dd January 12, 2024
+.Dt GIT.CAUSAL.AGENCY 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm causal agency
+.Nd \(dqI think some people from the Gentoo project are behind this.\(dq
+.
+.Sh DESCRIPTION
+basically cgit (awful software)
+getting hammered by web crawlers
+keeps making my machine crash.
+this static page will be here
+until I can find a better solution.
+clone urls and tarball urls are still functional.
+.
+.Bl -tag
+.It src \(em dontfiles
+.Dl git clone https://git.causal.agency/src
+.It ascii.town
+.Bl -tag
+.It torus \(em collaborative ASCII art
+.Dl git clone https://git.causal.agency/torus
+.It play \(em some games for SSH
+.Dl git clone https://git.causal.agency/play
+.El
+.It email
+.Bl -tag
+.It imbox \(em IMAP to mbox
+.Dl git clone https://git.causal.agency/imbox
+.It bubger \(em IMAP archive generator
+.Dl git clone https://git.causal.agency/bubger
+.It notemap \(em notemap
+.Dl git clone https://git.causal.agency/notemap
+.El
+.It forks
+.Bl -tag
+.It shulker \(em Discord to vanilla Minecraft bridge
+.Dl git clone https://git.causal.agency/shulker
+.It cgit-pink \(em web frontend for git
+.Dl git clone https://git.causal.agency/cgit-pink
+.It dash \(em patched shell with cmake build
+.Dl git clone https://git.causal.agency/dash
+.El
+.It games
+.Bl -tag
+.It wep \(em Windows Entertainment Pack recreations
+.Dl git clone https://git.causal.agency/wep
+.It cards \(em CARDS.DLL loader for SDL
+.Dl git clone https://git.causal.agency/cards
+.El
+.It irc
+.Bl -tag
+.It scooper \(em web interface for litterbox
+.Dl git clone https://git.causal.agency/scooper
+.It litterbox \(em IRC logger
+.Dl git clone https://git.causal.agency/litterbox
+.It pounce \(em IRC bouncer
+.Dl git clone https://git.causal.agency/pounce
+.It catgirl \(em IRC client
+.Dl git clone https://git.causal.agency/catgirl
+.El
+.It ports
+.Bl -tag
+.It jorts \(em my own ports tree for macOS
+.Dl git clone https://git.causal.agency/jorts
+.It exman \(em manuals for other systems
+.Dl git clone https://git.causal.agency/exman
+.It libretls \(em libtls for OpenSSL
+.Dl git clone https://git.causal.agency/libretls
+.It ports \(em Fx and Ox ports for this software
+.Dl git clone https://git.causal.agency/ports
+.El
+.It system
+.Bl -tag
+.It kitd \(em process supervisor for OpenBSD
+.Dl git clone https://git.causal.agency/kitd
+.It catsit \(em (deprecated) process supervisor
+.Dl git clone https://git.causal.agency/catsit
+.El
+.El
diff --git a/www/photo.causal.agency/.gitignore b/www/photo.causal.agency/.gitignore
new file mode 100644
index 00000000..a5f66a9d
--- /dev/null
+++ b/www/photo.causal.agency/.gitignore
@@ -0,0 +1,2 @@
+static/
+*.JPG
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0832.txt b/www/photo.causal.agency/2024-04-10/IMG_0832.txt
new file mode 100644
index 00000000..65724024
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0832.txt
@@ -0,0 +1,6 @@
+a red brick wall with some faded black graffiti.
+in the lower third, some bricks are missing
+from the outer layer in an arc shape.
+along the bottom is a ledge of conrete
+lightly covered in brick dust and chunks
+below the missing areas above.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0850.txt b/www/photo.causal.agency/2024-04-10/IMG_0850.txt
new file mode 100644
index 00000000..4cbb3def
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0850.txt
@@ -0,0 +1,6 @@
+grey steel beams of a building in early construction
+on a background of blue sky with some light clouds.
+the beams are intersecting at odd points,
+implying the final building will not be a simple box.
+the sun casts dark shadows into the interiors
+of the I-shaped metal.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0852.txt b/www/photo.causal.agency/2024-04-10/IMG_0852.txt
new file mode 100644
index 00000000..707d7cd6
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0852.txt
@@ -0,0 +1,4 @@
+in the foreground, a metal construction fence.
+behind that, the bright red arm of a sort of small crane.
+the arm is horizontal and crushing a perpendicular piece of fence,
+which has deformed smoothly under it.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0858.txt b/www/photo.causal.agency/2024-04-10/IMG_0858.txt
new file mode 100644
index 00000000..42f243e4
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0858.txt
@@ -0,0 +1,6 @@
+an uneven grid of old wooden-framed windows in an alley.
+the red paint on the frames is peeling badly,
+completely stripped in some spots.
+in the reflections of the lower windows
+we see the roofs of the opposite buildings
+and hints of clouds in the sky.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0859.txt b/www/photo.causal.agency/2024-04-10/IMG_0859.txt
new file mode 100644
index 00000000..ca33d7e0
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0859.txt
@@ -0,0 +1,6 @@
+an old backetball hoop mounted in an alley.
+the backboard has been graffitied
+and vines have invaded.
+a few red strands of net are left hanging from the hoop.
+the fence behind is painted with a design of yellow, purple, white and blue.
+it's the kind of hoop airbud might be hanging around.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0865.txt b/www/photo.causal.agency/2024-04-10/IMG_0865.txt
new file mode 100644
index 00000000..7a955fc2
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0865.txt
@@ -0,0 +1,2 @@
+deep tire tread pressed into mud in the center of an alley.
+a small branch of evergreen lies to one side.
diff --git a/www/photo.causal.agency/2024-04-10/IMG_0890.txt b/www/photo.causal.agency/2024-04-10/IMG_0890.txt
new file mode 100644
index 00000000..9d2cdc43
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-10/IMG_0890.txt
@@ -0,0 +1,9 @@
+a pipe coming out of a light brown brick wall.
+the pipe comes out of a metal square in the centre of the wall,
+travels up and left for a bit,
+before continuing straight up out of frame.
+opposite, in the bottom right,
+is the top of a red metal grate in front
+of a ground-level window.
+the brick below where the pipe enters the wall
+is stained dark.
diff --git a/www/photo.causal.agency/2024-04-14/IMG_1054.txt b/www/photo.causal.agency/2024-04-14/IMG_1054.txt
new file mode 100644
index 00000000..f4803ee2
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-14/IMG_1054.txt
@@ -0,0 +1,5 @@
+a short wall of natural rock,
+all broken up somewhat neatly
+along horizontal and vertical lines.
+most of the rock is cool grey,
+while some parts are warm brown.
diff --git a/www/photo.causal.agency/2024-04-14/IMG_1058.txt b/www/photo.causal.agency/2024-04-14/IMG_1058.txt
new file mode 100644
index 00000000..21aeb189
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-14/IMG_1058.txt
@@ -0,0 +1,6 @@
+moss on a bit of exposed natural rock
+surrounded by mostly brown grass.
+there is shorter, darker green and brown moss,
+as well as longer lighter green moss.
+some small pieces of the rock are broken off
+and lay in little piles.
diff --git a/www/photo.causal.agency/2024-04-14/IMG_1066.txt b/www/photo.causal.agency/2024-04-14/IMG_1066.txt
new file mode 100644
index 00000000..81747287
--- /dev/null
+++ b/www/photo.causal.agency/2024-04-14/IMG_1066.txt
@@ -0,0 +1,10 @@
+two green buds on the end of a thin branch
+on a blurry brown backdrop.
+the branch enters the frame
+from the bottom left corner,
+and there are three other pairs of buds
+along it,
+out of focus.
+there is a hint of another bebudded branch
+in the background,
+but there is otherwise very little green.
diff --git a/www/photo.causal.agency/generate.sh b/www/photo.causal.agency/generate.sh
new file mode 100644
index 00000000..4b30db92
--- /dev/null
+++ b/www/photo.causal.agency/generate.sh
@@ -0,0 +1,211 @@
+#!/bin/sh
+set -eu
+
+mkdir -p static/preview static/thumbnail
+
+resize() {
+	local photo=$1 size=$2 output=$3
+	if ! test -f $output; then
+		# FIXME: convert complains about not understanding XML
+		echo $output >&2
+		convert $photo -auto-orient -thumbnail $size $output 2>/dev/null ||:
+	fi
+}
+
+preview() {
+	local photo=$1
+	local preview=preview/${photo##*/}
+	resize $photo 25% static/$preview
+	echo $preview
+}
+
+thumbnail() {
+	local photo=$1
+	local thumbnail=thumbnail/${photo##*/}
+	resize $photo 5% static/$thumbnail
+	echo $thumbnail
+}
+
+encode() {
+	sed '
+		s/&/\&amp;/g
+		s/</\&lt;/g
+		s/"/\&quot;/g
+	' "$@"
+}
+
+page_title() {
+	date -j -f '%F' $1 '+%B %e, %Y'
+}
+
+page_head() {
+	local date=$1
+	local title=$(page_title $date)
+	cat <<-EOF
+	<!DOCTYPE html>
+	<meta charset="utf-8">
+	<meta name="viewport" content="width=device-width, initial-scale=1.0">
+	<link rel="alternate" type="application/atom+xml" href="../feed.atom">
+	<title>${title}</title>
+	<style>
+	html { color: #bbb; background-color: black; font-family: sans-serif; }
+	figure { margin: 1em; padding-top: 0.5em; text-align: center; }
+	img { max-width: calc(100vw - 2.5em); max-height: calc(100vh - 2.5em); }
+	details { max-width: 78ch; margin: 0.5em auto; }
+	</style>
+	<h1>${title}</h1>
+	EOF
+}
+
+photo_info() {
+	local photo=$1
+	ExposureTime=
+	FNumber=
+	FocalLength=
+	PhotographicSensitivity=
+	eval $(
+		identify -format '%[EXIF:*]' $photo 2>/dev/null |
+		grep -E 'ExposureTime|FNumber|FocalLength|PhotographicSensitivity' |
+		sed 's/^exif://'
+	)
+}
+
+photo_id() {
+	local photo=$1
+	photo=${photo##*/}
+	photo=${photo%%.*}
+	echo $photo
+}
+
+page_photo() {
+	local photo=$1 preview=$2 description=$3
+	if ! test -f $description; then
+		description=/dev/null
+	fi
+	photo_info $photo
+	cat <<-EOF
+	<figure id="$(photo_id $photo)">
+		<a href="${photo##*/}">
+			<img src="../${preview}" alt="$(encode $description)">
+		</a>
+		<figcaption>
+			${ExposureTime} ·
+			ƒ/$(bc -S 1 -e ${FNumber}) ·
+			$(bc -e ${FocalLength}) mm ·
+			${PhotographicSensitivity} ISO
+			<details>
+				<summary>description</summary>
+				$(encode $description)
+			</details>
+		</figcaption>
+	</figure>
+	EOF
+}
+
+index_head() {
+	cat <<-EOF
+	<!DOCTYPE html>
+	<meta charset="utf-8">
+	<meta name="viewport" content="width=device-width, initial-scale=1.0">
+	<link rel="alternate" type="application/atom+xml" href="feed.atom">
+	<title>Photos</title>
+	<style>
+	html { color: #bbb; background-color: black; font-family: sans-serif; }
+	a { text-decoration: none; color: inherit; }
+	</style>
+	EOF
+}
+
+index_page() {
+	local date=$1 root=${2:-}
+	cat <<-EOF
+	<h1><a href="${root}${root:+/}${date}/">$(page_title $date)</a></h1>
+	EOF
+}
+
+index_photo() {
+	local date=$1 photo=$2 thumbnail=$3 root=${4:-}
+	cat <<-EOF
+	<a href="${root}${root:+/}${date}/#$(photo_id $photo)">
+		<img src="${root}${root:+/}${thumbnail}">
+	</a>
+	EOF
+}
+
+Root=https://photo.causal.agency
+
+atom_head() {
+	local updated=$(date -u '+%FT%TZ')
+	cat <<-EOF
+	<?xml version="1.0" encoding="utf-8"?>
+	<feed xmlns="http://www.w3.org/2005/Atom">
+	<title>Photos</title>
+	<author><name>june</name><email>june@causal.agency</email></author>
+	<link href="${Root}"/>
+	<link rel="self" href="${Root}/feed.atom"/>
+	<id>${Root}/</id>
+	<updated>${updated}</updated>
+	EOF
+}
+
+atom_entry_head() {
+	local date=$1
+	local updated=$(
+		date -ju -f '%s' $(stat -f '%m' static/${date}/index.html) '+%FT%TZ'
+	)
+	cat <<-EOF
+	<entry>
+	<title>$(page_title $date)</title>
+	<link href="${Root}/${date}/"/>
+	<id>${Root}/${date}/</id>
+	<updated>${updated}</updated>
+	<content type="html">
+	EOF
+}
+
+atom_entry_tail() {
+	cat <<-EOF
+	</content>
+	</entry>
+	EOF
+}
+
+atom_tail() {
+	cat <<-EOF
+	</feed>
+	EOF
+}
+
+set --
+for date in 20*; do
+	mkdir -p static/${date}
+	page=static/${date}/index.html
+	if ! test -f $page; then
+		echo $page >&2
+		page_head $date >$page
+		for photo in ${date}/*.JPG; do
+			preview=$(preview $photo)
+			if ! test -f static/${photo}; then
+				ln $photo static/${photo}
+			fi
+			page_photo $photo $preview ${photo%.JPG}.txt >>$page
+		done
+	fi
+	set -- $date "$@"
+done
+
+echo static/index.html >&2
+index_head >static/index.html
+echo static/feed.atom >&2
+atom_head >static/feed.atom
+for date; do
+	index_page $date >>static/index.html
+	atom_entry_head $date >>static/feed.atom
+	for photo in ${date}/*.JPG; do
+		thumbnail=$(thumbnail $photo)
+		index_photo $date $photo $thumbnail >>static/index.html
+		index_photo $date $photo $thumbnail $Root | encode >>static/feed.atom
+	done
+	atom_entry_tail >>static/feed.atom
+done
+atom_tail >>static/feed.atom
diff --git a/www/photo.causal.agency/rsync.sh b/www/photo.causal.agency/rsync.sh
new file mode 100644
index 00000000..957911d2
--- /dev/null
+++ b/www/photo.causal.agency/rsync.sh
@@ -0,0 +1,5 @@
+#!/bin/sh
+set -eu
+
+sh generate.sh
+rsync -av static/ scout:/var/www/photo.causal.agency
diff --git a/www/temp.causal.agency/.gitignore b/www/temp.causal.agency/.gitignore
new file mode 100644
index 00000000..e31ee94e
--- /dev/null
+++ b/www/temp.causal.agency/.gitignore
@@ -0,0 +1 @@
+up
diff --git a/www/temp.causal.agency/Makefile b/www/temp.causal.agency/Makefile
new file mode 100644
index 00000000..a69a2b48
--- /dev/null
+++ b/www/temp.causal.agency/Makefile
@@ -0,0 +1,15 @@
+CGI_BIN = /var/www/cgi-bin
+
+CFLAGS += -std=c11 -Wall -Wextra -Wpedantic $$(pkg-config --cflags kcgi)
+LDLIBS = -static $$(pkg-config --static --libs kcgi-html)
+
+up:
+
+clean:
+	rm -f up
+
+install: up
+	install up ${CGI_BIN}/up
+
+uninstall:
+	rm -f ${CGI_BIN}/up
diff --git a/www/temp.causal.agency/up.c b/www/temp.causal.agency/up.c
new file mode 100644
index 00000000..561a8901
--- /dev/null
+++ b/www/temp.causal.agency/up.c
@@ -0,0 +1,193 @@
+/* Copyright (C) 2020  June McEnroe <june@causal.agency>
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as published by
+ * the Free Software Foundation, either version 3 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+#include <err.h>
+#include <fcntl.h>
+#include <stdarg.h>
+#include <stdint.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/types.h>
+#include <sysexits.h>
+#include <time.h>
+#include <unistd.h>
+
+#include <kcgi.h>
+#include <kcgihtml.h>
+
+static const char *Page = "up";
+static const struct kvalid Key = { NULL, "file" };
+
+static enum kcgi_err head(struct kreq *req, enum khttp http, enum kmime mime) {
+	return khttp_head(req, kresps[KRESP_STATUS], "%s", khttps[http])
+		|| khttp_head(req, kresps[KRESP_CONTENT_TYPE], "%s", kmimetypes[mime]);
+}
+
+static enum kcgi_err fail(struct kreq *req, enum khttp http) {
+	return head(req, http, KMIME_TEXT_PLAIN)
+		|| khttp_body(req)
+		|| khttp_printf(req, "%s\n", khttps[http]);
+}
+
+static int dir = -1;
+static const char *upload(const char *ext, void *ptr, size_t len) {
+	static char name[256];
+	snprintf(
+		name, sizeof(name), "%jx%08x%s%s",
+		(intmax_t)time(NULL), arc4random(),
+		(ext && ext[0] != '.' ? "." : ""), (ext ? ext : "")
+	);
+	int fd = openat(dir, name, O_CREAT | O_EXCL | O_WRONLY, 0644);
+	if (fd < 0) {
+		warn("%s", name);
+		return NULL;
+	}
+	ssize_t n = write(fd, ptr, len);
+	int error = close(fd);
+	if (n < 0 || error) {
+		warn("%s", name);
+		return NULL;
+	}
+	return name;
+}
+
+static enum kcgi_err handle(struct kreq *req) {
+	if (req->page) return fail(req, KHTTP_404);
+
+	if (req->method == KMETHOD_GET) {
+		struct khtmlreq html;
+		struct khtmlreq *h = &html;
+		return head(req, KHTTP_200, KMIME_TEXT_HTML)
+			|| khttp_body(req)
+			|| khtml_open(h, req, 0)
+			|| khtml_elem(h, KELEM_DOCTYPE)
+			|| khtml_elem(h, KELEM_TITLE)
+			|| khtml_puts(h, "Upload")
+			|| khtml_closeelem(h, 1)
+			|| khtml_attr(
+				h, KELEM_FORM,
+				KATTR_METHOD, "post",
+				KATTR_ACTION, "",
+				KATTR_ENCTYPE, "multipart/form-data",
+				KATTR__MAX
+			)
+			|| khtml_attr(
+				h, KELEM_INPUT,
+				KATTR_TYPE, "file",
+				KATTR_NAME, Key.name,
+				KATTR__MAX
+			)
+			|| khtml_attr(
+				h, KELEM_INPUT,
+				KATTR_TYPE, "submit",
+				KATTR_VALUE, "Upload",
+				KATTR__MAX
+			)
+			|| khtml_close(h);
+
+	} else if (req->method == KMETHOD_POST) {
+		struct kpair *field = req->fieldmap[0];
+		if (!field || !field->valsz) return fail(req, KHTTP_400);
+
+		const char *ext = strrchr(field->file, '.');
+		const char *name = upload(ext, field->val, field->valsz);
+		if (!name) return fail(req, KHTTP_507);
+
+		return head(req, KHTTP_303, KMIME_TEXT_PLAIN)
+			|| khttp_head(req, kresps[KRESP_LOCATION], "/%s", name)
+			|| khttp_body(req)
+			|| khttp_puts(req, name);
+
+	} else if (req->method == KMETHOD_PUT) {
+		struct kpair *field = req->fields;
+		if (!field || !field->valsz) return fail(req, KHTTP_400);
+
+		const char *ext = req->suffix;
+		if (!ext[0]) ext = strrchr(field->file, '.');
+		const char *name = upload(ext, field->val, field->valsz);
+		if (!name) return fail(req, KHTTP_507);
+
+		return head(req, KHTTP_200, KMIME_TEXT_PLAIN)
+			|| khttp_body(req)
+			|| khttp_printf(
+				req, "%s://%s/%s\n", kschemes[req->scheme], req->host, name
+			);
+
+	} else {
+		return fail(req, KHTTP_405);
+	}
+}
+
+int main(int argc, char *argv[]) {
+	int error;
+	const char *path = (argc > 1 ? argv[1] : ".");
+	dir = open(path, O_DIRECTORY);
+	if (dir < 0) err(EX_NOINPUT, "%s", path);
+
+#ifdef __OpenBSD__
+	error = unveil(path, "wc");
+	if (error) err(EX_OSERR, "unveil");
+#endif
+
+	if (!khttp_fcgi_test()) {
+#ifdef __OpenBSD__
+		error = pledge("stdio wpath cpath proc", NULL);
+		if (error) err(EX_OSERR, "pledge");
+#endif
+
+		struct kreq req;
+		error = khttp_parse(&req, &Key, 1, &Page, 1, 0);
+		if (error) errx(EX_PROTOCOL, "khttp_parse: %s", kcgi_strerror(error));
+
+#ifdef __OpenBSD__
+		error = pledge("stdio wpath cpath", NULL);
+		if (error) err(EX_OSERR, "pledge");
+#endif
+
+		error = handle(&req);
+		if (error) errx(EX_PROTOCOL, "%s", kcgi_strerror(error));
+		khttp_free(&req);
+		return EX_OK;
+	}
+
+#ifdef __OpenBSD__
+	error = pledge("stdio wpath cpath unix sendfd recvfd proc", NULL);
+	if (error) err(EX_OSERR, "pledge");
+#endif
+
+	struct kfcgi *fcgi;
+	error = khttp_fcgi_init(&fcgi, &Key, 1, &Page, 1, 0);
+	if (error) errx(EX_CONFIG, "khttp_fcgi_init: %s", kcgi_strerror(error));
+
+#ifdef __OpenBSD__
+	error = pledge("stdio wpath cpath recvfd", NULL);
+	if (error) err(EX_OSERR, "pledge");
+#endif
+
+	for (
+		struct kreq req;
+		!(error = khttp_fcgi_parse(fcgi, &req));
+		khttp_free(&req)
+	) {
+		error = handle(&req);
+		if (error && error != KCGI_HUP) break;
+	}
+	if (error != KCGI_EXIT) {
+		errx(EX_PROTOCOL, "khttp_fcgi_parse: %s", kcgi_strerror(error));
+	}
+	khttp_fcgi_free(fcgi);
+}
diff --git a/www/text.causal.agency/.gitignore b/www/text.causal.agency/.gitignore
new file mode 100644
index 00000000..66b3e637
--- /dev/null
+++ b/www/text.causal.agency/.gitignore
@@ -0,0 +1,4 @@
+*.txt
+colb
+feed.atom
+igp
diff --git a/www/text.causal.agency/001-make.7 b/www/text.causal.agency/001-make.7
new file mode 100644
index 00000000..b4805729
--- /dev/null
+++ b/www/text.causal.agency/001-make.7
@@ -0,0 +1,159 @@
+.Dd September 17, 2018
+.Dt MAKE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Using Make
+.Nd writing less Makefile
+.
+.Sh DESCRIPTION
+Let's talk about
+.Xr make 1 .
+I think an important thing to know about
+.Xr make 1
+is that you don't need to write a
+.Pa Makefile
+to use it.
+There are default rules
+for C, C++ and probably Fortran.
+To build
+.Pa foo
+from
+.Pa foo.c ,
+just run:
+.
+.Pp
+.Dl make foo
+.
+.Pp
+The default rule for C files uses the
+.Ev CFLAGS
+variable,
+so you can set that in the environment
+to pass flags to the C compiler:
+.
+.Pp
+.Dl CFLAGS=-Wall make foo
+.
+.Pp
+It also uses
+.Ev LDLIBS
+for linking,
+so you can add libraries with:
+.
+.Pp
+.Dl LDLIBS=-lcurses make foo
+.
+.Pp
+Obviously writing this every time
+would become tedious,
+so it might be time to write a
+.Pa Makefile .
+But it really doesn't need much:
+.
+.Bd -literal -offset indent
+CFLAGS += -Wall -Wextra
+LDLIBS = -lcurses
+
+foo:
+.Ed
+.
+.Pp
+Assigning
+.Ev CFLAGS
+with
+.Ql +=
+preserves the system default
+or anything passed in the environment.
+Declaring
+.Pa foo
+as the first rule
+makes it the default when
+.Ql make
+is run without a target.
+Note that the rule doesn't need a definition;
+the default will still be used.
+.
+.Pp
+If
+.Pa foo
+is built from serveral source files,
+unfortunately a rule definition is required:
+.
+.Bd -literal -offset indent
+OBJS = foo.o bar.o baz.o
+
+foo: $(OBJS)
+	$(CC) $(LDFLAGS) $(OBJS) $(LDLIBS) -o $@
+.Ed
+.
+.Pp
+This rule uses
+.Ev LDFLAGS
+for passing linker flags,
+which is what the default rule does.
+The
+.Ql $@
+variable here expands to
+.Ql foo ,
+so this rule can be copied easily
+for other binary targets.
+.
+.Pp
+If some sources depend on a header file,
+they can be automatically rebuilt
+when the header changes
+by declaring a dependency rule:
+.
+.Pp
+.Dl foo.o bar.o: foo.h
+.
+.Pp
+Note that several files can appear
+either side of the
+.Ql ":" .
+.
+.Pp
+Lastly,
+it's always nice to add a
+.Cm clean
+target:
+.
+.Bd -literal -offset indent
+clean:
+	rm -f $(OBJS) foo
+.Ed
+.
+.Pp
+I hope this helps getting started with
+.Xr make 1
+without writing too much
+.Pa Makefile !
+.
+.Sh EXAMPLES
+The example
+.Pa Makefile
+in its entirety:
+.
+.Bd -literal -offset indent
+CFLAGS += -Wall -Wextra
+LDLIBS = -lcurses
+OBJS = foo.o bar.o baz.o
+
+foo: $(OBJS)
+	$(CC) $(LDFLAGS) $(OBJS) $(LDLIBS) -o $@
+
+foo.o bar.o: foo.h
+
+clean:
+	rm -f $(OBJS) foo
+.Ed
+.
+.Sh AUTHORS
+.An Mt june@causal.agency
+.
+.Pp
+This document is produced from
+.Xr mdoc 7
+source available from
+.Lk https://git.causal.agency/src/tree/www/text.causal.agency
diff --git a/www/text.causal.agency/002-writing-mdoc.7 b/www/text.causal.agency/002-writing-mdoc.7
new file mode 100644
index 00000000..b377d364
--- /dev/null
+++ b/www/text.causal.agency/002-writing-mdoc.7
@@ -0,0 +1,138 @@
+.Dd September 27, 2018
+.Dt WRITING-MDOC 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Writing mdoc
+.Nd semantic markup
+.
+.Sh DESCRIPTION
+I recently learned how to write man pages
+so that I could document
+a bunch of little programs I've written.
+Modern man pages are written in
+.Xr mdoc 7 ,
+whose documentation is also available from
+.Lk http://mandoc.bsd.lv .
+.
+.Pp
+.Xr mdoc 7
+differs from many other markup languages
+by providing
+.Dq semantic markup
+rather than just
+.Dq physical markup.
+What this means is that
+the markup indicates what something is,
+not how to format it.
+For example,
+the
+.Ql \&Ar
+macro is used to indicate
+command-line arguments
+rather than one of the macros
+for bold, italic or underline.
+This frees each author of having to choose
+and enables consistent presentation
+across different man pages.
+.
+.Pp
+Another advantage of semantic markup
+is that information can be extracted from it.
+For example,
+.Xr makewhatis 8
+can easily extract the name and short description
+from each man page
+thanks to the
+.Ql \&Nm
+and
+.Ql \&Nd
+macros.
+I use the same information
+to generate an Atom feed for these documents,
+though in admittedly a much less robust way than
+.Xr mandoc 1 .
+.
+.Pp
+When it comes to actually writing
+.Xr mdoc 7 ,
+it can take some getting used to.
+The language is of
+.Xr roff 7
+lineage
+so its syntax is very particular.
+Macros cannot appear inline,
+but must start on new lines
+beginning with
+.Ql \&. .
+Sentences should likewise
+always start on a new line.
+Since I'm in the habit of writing with
+semantic line breaks,
+I actually find these requirements
+fit in well.
+.
+.Pp
+The more frustrating syntax limitation to me
+is the rule against empty lines.
+Without them,
+it can be quite difficult to edit a lengthy document.
+Thankfully,
+lines with only a
+.Ql \&.
+on them are allowed,
+but this still causes visual noise.
+To alleviate that,
+I have a
+.Xr vim 1
+syntax file for
+.Xr mdoc 7
+which conceals the lone dots:
+.
+.Bd -literal -offset indent
+if exists("b:current_syntax")
+	finish
+endif
+
+runtime! syntax/nroff.vim
+unlet! b:current_syntax
+
+setlocal sections+=ShSs
+syntax match mdocBlank /^\\.$/ conceal
+setlocal conceallevel=2
+
+let b:current_syntax = "mdoc"
+.Ed
+.
+.Pp
+It also adds the
+.Xr mdoc 7
+section header and subsection header macros to the
+.Cm sections
+option to make
+.Xr vim 1 Ap s
+.Ic {
+and
+.Ic }
+motions
+aware of them.
+.
+.Pp
+With that,
+I've found writing man pages pleasant and rewarding.
+I've started writing other documents with
+.Xr mdoc 7
+as well,
+as you can see here.
+.
+.Sh SEE ALSO
+.Lk http://rhodesmill.org/brandon/2012/one-sentence-per-line/ "Semantic Linefeeds"
+.
+.Sh AUTHORS
+.An Mt june@causal.agency
+.
+.Pp
+This document is produced from
+.Xr mdoc 7
+source available from
+.Lk https://git.causal.agency/src/tree/www/text.causal.agency
diff --git a/www/text.causal.agency/003-pleasant-c.7 b/www/text.causal.agency/003-pleasant-c.7
new file mode 100644
index 00000000..16030b7e
--- /dev/null
+++ b/www/text.causal.agency/003-pleasant-c.7
@@ -0,0 +1,120 @@
+.Dd September 30, 2018
+.Dt PLEASANT-C 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Pleasant C
+.Nd it's good, actually
+.
+.Sh DESCRIPTION
+I've been writing a lot of C lately
+and actually find it very pleasant.
+I want to talk about some of its ergonomic features.
+These are C99 features unless otherwise noted.
+.
+.Ss Initializer syntax
+Struct and union initializer syntax
+is well generalized.
+Designators can be chained,
+making initializing nested structs easy,
+and all uninitialized fields are zeroed.
+.
+.Bd -literal -offset indent
+struct {
+	struct pollfd fds[2];
+} loop = {
+	.fds[0].fd = STDIN_FILENO,
+	.fds[1].fd = STDOUT_FILENO,
+	.fds[0].events = POLLIN,
+	.fds[1].events = POLLOUT,
+};
+.Ed
+.
+.Ss Variable-length arrays
+VLAs can be multi-dimensional,
+which can avoid manual stride multiplications
+needed to index a flat
+.Xr malloc 3 Ap d
+array.
+.
+.Bd -literal -offset indent
+uint8_t glyphs[len][height][width];
+fread(glyphs, height * width, len, stdin);
+.Ed
+.
+.Ss Incomplete array types
+The last field of a struct can be an
+.Dq incomplete
+array type,
+which means it doesn't have a length.
+A variable amount of space for the struct can be
+.Xr malloc 3 Ap d ,
+or the struct can be used as
+a sort of pointer with fields.
+.
+.Bd -literal -offset indent
+struct Line {
+	enum Filter type;
+	uint8_t data[];
+} *line = &png.data[1 + lineSize()];
+.Ed
+.
+.Ss Anonymous struct and union fields (C11)
+Members of structs or unions
+which are themselves structs or unions
+can be unnamed.
+In that case,
+each of the inner fields
+is treated as a member of the outer struct or union.
+This makes working with tagged unions nicer.
+.
+.Bd -literal -offset indent
+struct Message {
+	enum { Foo, Bar } type;
+	union {
+		uint8_t foo;
+		uint32_t bar;
+	};
+} msg = { .type = Foo, .foo = 0xFF };
+.Ed
+.
+.Ss Static assert (C11)
+Assertions can be made at compile time.
+Most useful for checking sizes of structs.
+.
+.Bd -literal -offset indent
+static_assert(13 == sizeof(struct PNGHeader), "PNG IHDR size");
+.Ed
+.
+.Ss Leading-break switch
+This one is just an odd style choice
+I came across that C happens to allow.
+To prevent accidental fall-through
+in switch statements,
+you can put breaks before the case labels.
+.
+.Bd -literal -offset indent
+while (0 < (opt = getopt(argc, argv, "h:w:"))) {
+	switch (opt) {
+		break; case 'h': height = optarg;
+		break; case 'w': width = optarg;
+		break; default:  return EX_USAGE;
+	}
+}
+.Ed
+.
+.Sh AUTHORS
+.An Mt june@causal.agency
+.
+.Pp
+This document is produced from
+.Xr mdoc 7
+source available from
+.Lk https://git.causal.agency/src/tree/www/text.causal.agency
+.
+.Sh CAVEATS
+This isn't meant to be advice.
+It's just how I like to write C,
+and I don't
+.Dq ship
+software in C.
diff --git a/www/text.causal.agency/004-uloc.7 b/www/text.causal.agency/004-uloc.7
new file mode 100644
index 00000000..edd78d80
--- /dev/null
+++ b/www/text.causal.agency/004-uloc.7
@@ -0,0 +1,64 @@
+.Dd December 14, 2018
+.Dt ULOC 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm ULOC
+.Nd unique lines of code
+.
+.Sh DESCRIPTION
+There are many tools available
+which measure SLOC: source lines of code.
+These tools are strangely complex
+for what they intend to do,
+which is to estimate the relative sizes of projects.
+They perform some amount of parsing
+in order to discount comments in various languages,
+and for reasons unknown each format their ouput
+in some oddly encumbered way.
+.
+.Pp
+I propose a much simpler method
+of estimating relative sizes of projects:
+unique lines of code.
+ULOC can be calculated with standard tools as follows:
+.
+.Bd -literal -offset indent
+sort -u *.h *.c | wc -l
+.Ed
+.
+.Pp
+In my opinion,
+the number this produces
+should be a better estimate of
+the complexity of a project.
+Compared to SLOC,
+not only are blank lines discounted,
+but so are close-brace lines
+and other repetitive code
+such as common includes.
+On the other hand,
+ULOC counts comments,
+which require just as much maintenance
+as the code around them does,
+while avoiding inflating the result
+with license headers which appear in every file,
+for example.
+.
+.Pp
+It can also be amusing
+to read all of your code sorted alphabetically.
+.
+.Sh AUTHORS
+.An Mt june@causal.agency
+.
+.Pp
+This document is produced from
+.Xr mdoc 7
+source available from
+.Lk https://git.causal.agency/src/tree/www/text.causal.agency
+.
+.Sh CAVEATS
+Estimates such as these
+should not be used for decision making
+as if they were data.
diff --git a/www/text.causal.agency/005-testing-c.7 b/www/text.causal.agency/005-testing-c.7
new file mode 100644
index 00000000..d0c636ff
--- /dev/null
+++ b/www/text.causal.agency/005-testing-c.7
@@ -0,0 +1,73 @@
+.Dd December 21, 2018
+.Dt TESTING-C 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Testing C
+.Nd a simple unit testing setup
+.
+.Sh DESCRIPTION
+This is a simple approach
+to unit testing in C
+that I've used in a couple projects.
+At the bottom of a C file
+with some code I want to test,
+I add:
+.
+.Bd -literal -offset indent
+#ifdef TEST
+#include <assert.h>
+
+int main(void) {
+	assert(...);
+	assert(...);
+}
+
+#endif
+.Ed
+.
+.Pp
+This file normally produces a
+.Pa .o
+to be linked into the main binary.
+For testing,
+I produce separate binaries
+and run them with
+.Xr make 1 :
+.
+.Bd -literal -offset indent
+TESTS = foo.t bar.t
+
+\&.SUFFIXES: .t
+
+\&.c.t:
+	$(CC) $(CFLAGS) -DTEST $(LDFLAGS) $< $(LDLIBS) -o $@
+
+test: $(TESTS)
+	set -e; $(TESTS:%=./%;)
+.Ed
+.
+.Pp
+Note that the test binaries
+aren't linked with the rest of the code,
+so there is potential for simple stubbing or mocking.
+.
+.Pp
+To get the best output
+from C's simple
+.Xr assert 3 ,
+it's best to assert the result
+of a helper function
+which takes the expected output
+and the test input,
+rather than calling
+.Xr assert 3
+inside the helper function.
+This way,
+the message printed by the assert failure
+contains a useful line number
+and the expected output
+rather than just variable names.
+.
+.Sh AUTHORS
+.An Mt june@causal.agency
diff --git a/www/text.causal.agency/006-some-libs.7 b/www/text.causal.agency/006-some-libs.7
new file mode 100644
index 00000000..5af65404
--- /dev/null
+++ b/www/text.causal.agency/006-some-libs.7
@@ -0,0 +1,96 @@
+.Dd December 11, 2019
+.Dt SOME-LIBS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Some Libraries
+.Nd good ones
+.
+.Sh DESCRIPTION
+This is a little list of C libraries
+I've had good experiences using.
+.
+.Bl -tag -width Ds
+.It Fl lcurl
+The library behind the
+.Xr curl 1
+command.
+It downloads or uploads things on the internet
+through a number of protocols,
+not just HTTP.
+It has an easy-to-use library API,
+appropriately named
+.Xr libcurl-easy 3 .
+I've used it to implement a
+.Lk https://causal.agency/bin/title.html "page title fetcher" .
+.
+.It Fl lcurses
+Okay so this one really isn't great.
+Its interfaces can seem archaic
+and its documentation is often poor.
+However, it gets the job done
+and is commonly available pretty much everywhere.
+Interesting to note that
+.Nx
+uses its own implementation of curses
+that is not GNU ncurses,
+unlike
+.Fx .
+.
+.It Fl ledit
+This is a BSD line editing library,
+similar to GNU readline.
+It supports right-aligned prompts,
+which I prefer for variable-length
+information in shells.
+.
+.It Fl lkcgi
+A CGI and FastCGI library
+for web applications in C.
+Don't worry,
+it isolates HTTP parsing and input validation
+from application logic
+in sandboxed processes.
+I think it's an excellent example
+of how to design an API for C.
+I used it to implement the
+.Lk https://ascii.town/explore.html "torus web viewer" .
+.
+.It Fl lsqlite3
+An embedded relational database engine.
+It's amazing what you can do with this,
+and it's super easy to use!
+My one gripe with it is that the library and SQL documentation
+are not available as
+.Xr man 1
+pages.
+I'm currently working on a project using SQLite,
+but it hasn't gotten very far yet.
+.
+.It Fl ltls
+This is a new library in LibreSSL
+which provides a much simpler interface for TLS sockets
+compared to
+.Fl lssl .
+It's much more like what you'd expect
+from other TLS socket wrappers,
+with calls like
+.Xr tls_connect 3 ,
+.Xr tls_read 3
+and
+.Xr tls_write 3 .
+I've used this for IRC clients, bouncers and bots.
+.
+.It Fl lz
+An implementation of the DEFLATE compression algorithm
+and gzip format.
+It's all documented in comments in
+.In zlib.h ,
+which isn't bad,
+but for my own use I copied the docs into
+.Lk https://code.causal.agency/june/zlib-man-pages "manual pages" .
+I've used this for decoding and encoding PNG images.
+.El
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/007-cgit-setup.7 b/www/text.causal.agency/007-cgit-setup.7
new file mode 100644
index 00000000..44fb436a
--- /dev/null
+++ b/www/text.causal.agency/007-cgit-setup.7
@@ -0,0 +1,271 @@
+.Dd December 15, 2019
+.Dt CGIT-SETUP 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm cgit setup
+.Nd configuration notes
+.
+.Sh DESCRIPTION
+I just set up cgit on
+.Lk https://git.causal.agency
+to replace an instance of gitea.
+After 30 days of uptime,
+gitea had accumulated over 11 hours of CPU time
+and was using hundreds of megabytes of memory.
+cgit is much more lightweight
+and much more in line with my aesthetic.
+I'm documenting how I set it up here
+mostly to remind myself in the future.
+.
+.Ss slowcgi
+cgit is CGI software,
+but
+.Xr nginx 8
+only supports FastCGI.
+I used
+.Xr slowcgi 8
+as a compatibility layer
+by adding the following to
+.Pa /etc/rc.conf :
+.Bd -literal -offset indent
+slowcgi_enable="YES"
+slowcgi_flags="-p / -s /var/run/slowcgi.sock"
+.Ed
+.
+.Ss nginx
+I added the following in a new
+.Cm server
+block to
+.Pa /usr/local/etc/nginx/nginx.conf :
+.Bd -literal -offset indent
+root /usr/local/www/cgit;
+location / {
+	try_files $uri @cgit;
+}
+location @cgit {
+	fastcgi_pass unix:/var/run/slowcgi.sock;
+	fastcgi_param SCRIPT_FILENAME $document_root/cgit.cgi;
+	fastcgi_param SCRIPT_NAME /;
+	fastcgi_param PATH_INFO $uri;
+	fastcgi_param QUERY_STRING $query_string;
+	fastcgi_param REQUEST_METHOD $request_method;
+	fastcgi_param CONTENT_TYPE $content_type;
+	fastcgi_param CONTENT_LENGTH $content_length;
+	fastcgi_param HTTPS $https if_not_empty;
+	fastcgi_param SERVER_PORT $server_port;
+	fastcgi_param SERVER_NAME $server_name;
+}
+.Ed
+.
+.Pp
+The
+.Cm try_files
+directive causes
+.Xr nginx 8
+to first try to serve static files from
+.Pa /usr/local/www/cgit
+before passing anything else on to FastCGI.
+.
+.Pp
+The
+.Va SCRIPT_FILENAME
+parameter tells
+.Xr slowcgi 8
+the path of the CGI binary to run.
+Setting
+.Va SCRIPT_NAME
+to
+.Pa /
+tells cgit its root URL
+and avoids it using query strings for everything.
+.
+.Ss cgit
+cgit doesn't provide any configuration to start from,
+so you have to just read
+.Xr cgitrc 5 .
+I added the following to
+.Pa /usr/local/etc/cgitrc :
+.Bd -literal -offset indent
+cache-size=1024
+clone-url=https://$HTTP_HOST/$CGIT_REPO_URL
+snapshots=tar.gz zip
+remove-suffix=1
+enable-git-config=1
+scan-path=/home/june/pub
+.Ed
+.
+.Pp
+The
+.Cm cache-size
+option enables caching,
+which by default is stored in
+.Pa /var/cache/cgit ,
+so I made sure that directory exists
+and is writable by the
+.Sy www
+user.
+The
+.Cm clone-url
+option sets the clone URL to advertise.
+cgit will automatically serve git over HTTP.
+The
+.Cm snapshots
+option makes tarballs available for tags and commits.
+.
+.Pp
+The
+.Cm scan-path
+option causes cgit to scan the given path
+for git repositories.
+I'm putting mine in
+.Pa ~/pub .
+The
+.Cm remove-suffix
+option causes cgit to remove the
+.Pa .git
+suffix from the URLs it uses
+for the repositories it finds,
+so that
+.Pa ~/pub/pounce.git
+is served at
+.Pa /pounce .
+The
+.Cm enable-git-config
+option allows controlling some cgit options
+from the
+.Xr git-config 1
+of each repository.
+See
+.Sx git
+below.
+.
+.Pp
+I also set up a filter to render
+.Xr mdoc 7
+files
+and do syntax highlighting
+by adding the following to
+.Pa cgitrc :
+.Bd -literal -offset indent
+readme=:README.7
+readme=:README
+about-filter=/usr/local/libexec/cgit-filter
+source-filter=/usr/local/libexec/cgit-filter
+.Ed
+.
+.Pp
+The
+.Cm readme
+options tell cgit which files to look for
+to render the
+.Dq about
+page.
+The colon prefix causes it to look for them
+in the git tree.
+The
+.Pa /usr/local/libexec/cgit-filter
+script contains the following:
+.Bd -literal -offset indent
+#!/bin/sh
+case "$1" in
+	(*.[1-9])
+		/usr/bin/mandoc -T utf8 | /usr/local/libexec/ttpre
+		;;
+	(*)
+		exec /usr/local/libexec/hi -t -n "$1" -f html -o anchor
+		;;
+esac
+.Ed
+.
+.Pp
+Filter scripts are run with the filename as their first argument
+and the contents of the file on standard input.
+The
+.Xr ttpre 1
+command is my own utility to convert
+.Xr man 1
+output to HTML.
+The
+.Xr hi 1
+command is my own
+.Lk https://causal.agency/bin/hi.html "syntax highlighter" .
+.
+.Ss git
+I create my repositories in
+.Pa ~/pub
+with
+.Ql git init --bare
+and use
+.Pa git.causal.agency:pub/example.git
+locally as the remote.
+Descriptions are set by editing the
+.Pa description
+file in each repository.
+The section and homepage can be set with
+.Xr git-config 1
+through the keys
+.Cm cgit.section
+and
+.Cm cgit.homepage ,
+respectively,
+thanks to the
+.Cm enable-git-config
+option above.
+.
+.Ss Redirects
+I added the following to the
+.Cm server
+block that used to serve gitea in
+.Pa nginx.conf :
+.Bd -literal -offset indent
+location ~* /june/([^.]+)[.]git(.*) {
+	return 301 https://git.causal.agency/$1$2?$query_string;
+}
+location ~* /june/([^/]+) {
+	return 301 https://git.causal.agency/$1;
+}
+location / {
+	return 301 https://git.causal.agency;
+}
+.Ed
+.
+.Pp
+This redirects any links to my gitea repos
+to the corresponding repo in cgit.
+The first
+.Sy location
+block also redirects gitea HTTP clone URLs to cgit
+so that
+.Xr git-pull 1
+continues to work on existing clones.
+.
+.Ss Update: fast HTTPS clones
+Someone pointed out that cloning my repos
+over HTTPS was incredibly slow,
+and this is because cgit only implements the
+.Dq dumb
+HTTP git transport.
+To speed up cloning,
+I send the URLs used by the
+.Dq smart
+HTTP transport to
+.Xr git-http-backend 1
+instead:
+.Bd -literal -offset indent
+location ~ /.+/(info/refs|git-upload-pack) {
+	fastcgi_pass unix:/var/run/slowcgi.sock;
+	fastcgi_param SCRIPT_NAME /usr/local/libexec/git-core/git-http-backend;
+	fastcgi_param GIT_HTTP_EXPORT_ALL 1;
+	fastcgi_param GIT_PROJECT_ROOT /home/june/pub;
+	include fastcgi_params;
+}
+.Ed
+.
+.Pp
+I factored out the FastCGI parameters
+I'm using with cgit
+to be included here as well.
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/008-how-irc.7 b/www/text.causal.agency/008-how-irc.7
new file mode 100644
index 00000000..aba1bbf9
--- /dev/null
+++ b/www/text.causal.agency/008-how-irc.7
@@ -0,0 +1,193 @@
+.Dd March  8, 2020
+.Dt HOW-IRC 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm How I Relay Chat
+.Nd in code
+.
+.Sh DESCRIPTION
+I've been writing a lot of IRC software lately
+.Pq Sx SEE ALSO ,
+and developed some nice code patterns
+that I've been reusing.
+Here they are.
+.
+.Ss Parsing
+I use fixed size buffers almost everywhere,
+so it's necessary to know IRC's size limits.
+A traditional IRC message is a maximum of 512 bytes,
+but the IRCv3 message-tags spec adds
+(unreasonably, in my opinion)
+8191 bytes for tags.
+IRC messages also have a maximum of 15 command parameters.
+.Bd -literal -offset indent
+enum { MessageCap = 8191 + 512 };
+enum { ParamCap = 15 };
+.Ed
+.
+.Pp
+If I'm using tags,
+I'll use X macros
+to declare the set I care about.
+X macros are a way of maintaining parallel arrays,
+or in this case an enum and an array.
+.Bd -literal -offset indent
+#define ENUM_TAG \e
+	X("msgid", TagMsgid) \e
+	X("time", TagTime)
+
+enum Tag {
+#define X(name, id) id,
+	ENUM_TAG
+#undef X
+	TagCap,
+};
+
+static const char *TagNames[TagCap] = {
+#define X(name, id) [id] = name,
+	ENUM_TAG
+#undef X
+};
+.Ed
+.
+.Pp
+The TagNames array is used by the parsing function
+to assign tag values into the message structure,
+which looks like this:
+.Bd -literal -offset indent
+struct Message {
+	char *tags[TagCap];
+	char *nick;
+	char *user;
+	char *host;
+	char *cmd;
+	char *params[ParamCap];
+};
+.Ed
+.
+.Pp
+I'm a fan of using
+.Xr strsep 3
+for simple parsing.
+Although it modifies its input
+(replacing delimiters with NUL terminators),
+since the raw message is in a static buffer,
+it is ideal for so-called zero-copy parsing.
+I'm not going to include the whole parsing function here,
+but I will at least include the part that many get wrong,
+which is dealing with the colon-prefixed trailing parameter:
+.Bd -literal -offset indent
+msg.cmd = strsep(&line, " ");
+for (int i = 0; line && i < ParamCap; ++i) {
+	if (line[0] == ':') {
+		msg.params[i] = &line[1];
+		break;
+	}
+	msg.params[i] = strsep(&line, " ");
+}
+.Ed
+.
+.Ss Handling
+To handle IRC commands and replies
+I add handler functions to a big array.
+I usually have some form of helper as well
+to check the number of expected parameters.
+.Bd -literal -offset indent
+typedef void HandlerFn(struct Message *msg);
+
+static const struct Handler {
+	const char *cmd;
+	HandlerFn *fn;
+} Handlers[] = {
+	{ "001", handleReplyWelcome },
+	{ "PING", handlePing },
+	{ "PRIVMSG", handlePrivmsg },
+};
+.Ed
+.
+.Pp
+Since I keep these arrays sorted anyway,
+I started using the standard
+.Xr bsearch 3
+function,
+but a basic for loop probably works just as well.
+I do wish I could compile-time assert
+that the array really is sorted, though.
+.Bd -literal -offset indent
+static int compar(const void *cmd, const void *_handler) {
+	const struct Handler *handler = _handler;
+	return strcmp(cmd, handler->cmd);
+}
+
+void handle(struct Message msg) {
+	if (!msg.cmd) return;
+	const struct Handler *handler = bsearch(
+		msg.cmd,
+		Handlers, ARRAY_LEN(Handlers),
+		sizeof(*handler), compar
+	);
+	if (handler) handler->fn(&msg);
+}
+.Ed
+.
+.Ss Capabilities
+For IRCv3 capabilties
+I use X macros again,
+this time with another handy macro
+for declaring bit flag enums.
+.Bd -literal -offset indent
+#define BIT(x) x##Bit, x = 1 << x##Bit, x##Bit_ = x##Bit
+
+#define ENUM_CAP \e
+	X("message-tags", CapMessageTags) \e
+	X("sasl", CapSASL) \e
+	X("server-time", CapServerTime)
+
+enum Cap {
+#define X(name, id) BIT(id),
+	ENUM_CAP
+#undef X
+};
+
+static const char *CapNames[] = {
+#define X(name, id) [id##Bit] = name,
+	ENUM_CAP
+#undef X
+};
+.Ed
+.
+.Pp
+The
+.Fn BIT
+macro declares, for example,
+.Dv CapSASL
+as the bit flag and
+.Dv CapSASLBit
+as the corresponding index.
+The
+.Vt "enum Cap"
+is used as a set,
+for example checking if SASL is enabled with
+.Ql caps & CapSASL .
+.
+.Pp
+These patterns are serving my IRC software well,
+and my IRC projects are serving me well.
+It is immensely satisfying
+to be (near) constantly using software
+that I wrote myself and am happy with,
+regardless of how niche it may be.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://git.causal.agency/pounce/about "IRC bouncer"
+.It
+.Lk https://git.causal.agency/litterbox/about "IRC logger"
+.It
+.Lk https://git.causal.agency/catgirl/about "IRC client"
+.El
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/009-casual-update.7 b/www/text.causal.agency/009-casual-update.7
new file mode 100644
index 00000000..0548436a
--- /dev/null
+++ b/www/text.causal.agency/009-casual-update.7
@@ -0,0 +1,127 @@
+.Dd May  6, 2020
+.Dt CASUAL-UPDATE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm casual update
+.Nd software developments
+.
+.Sh DESCRIPTION
+I've been figuring out more of IMAP
+and Internet Messages in general
+while working on a new project
+so I've revisited some older ones.
+I've copied my somewhat more proper
+IMAP parsing code into them,
+so they should be more robust.
+.
+.Pp
+.Xr imbox 1
+is my tool to export messages
+in mboxrd format directly from IMAP.
+It's mostly for applying patches sent by email
+without having any kind of local mail setup.
+For that,
+it includes the
+.Xr git-fetch-email 1
+wrapper which works very similarly to
+.Xr git-send-email 1 .
+I learned by reading the source of
+.Xr git-subtree 1
+that
+.Xr git-rev-parse 1
+can be used by shell scripts
+to parse long options,
+so I added those.
+I also added the
+.Fl Fl apply
+flag to automatically pipe to
+.Xr git-am 1
+with the right flags for mboxrd.
+.
+.Pp
+.Xr notemap 1
+is a tool for mirroring text files
+to an IMAP Notes mailbox,
+which is used by FastMail's web UI
+and the macOS/iOS Notes app.
+Its original parsing code
+was particularly ad-hoc.
+Since I've now learned
+how UTF-8 headers are encoded,
+I updated it to properly encode
+the file name in the Subject line.
+.
+.Pp
+I also got distracted by
+a conversation about UNIX-domain sockets
+where I was comparing the macOS and FreeBSD
+.Xr unix 4
+pages and the Linux
+.Xr unix 7
+page.
+This lead me to make
+.Xr exman 1 ,
+a tool to locally install and read
+manual pages for Linux, POSIX,
+.Fx ,
+.Nx
+and
+.Ox .
+I've already gotten quite a bit of use out of it.
+.
+.Pp
+In yet another IRC distraction,
+I was talking about some further plans for my IRC software,
+and realized it might be time to write
+my future projects list down.
+I opened a
+.Pa .plan
+file,
+immediately wondered how anyone can write plain text,
+then switched to a
+.Pa plan.7
+file.
+There's nothing I won't use
+.Xr mdoc 7
+for.
+After a little setup,
+I can now be fingered,
+and make jokes about this silly little protocol
+from the days of old.
+.Xr finger 1 Ap s
+default output fills me with joy:
+.Bd -unfilled -offset indent
+No Mail.
+No Plan.
+.Ed
+.
+.Pp
+And speaking of IRC and plans,
+I've been meaning to tag
+.Xr catgirl 1
+version 1.0 for a while now.
+I've been using it as my main client
+and my commits to it have really slowed down.
+When I do tag it,
+I'm planning on writing another post
+about my whole
+.Dq suite
+of IRC software
+and how the parts work together.
+Watch this space.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://git.causal.agency/imbox "imbox"
+.It
+.Lk https://git.causal.agency/notemap "notemap"
+.It
+.Lk https://git.causal.agency/exman "exman"
+.It
+.Lk https://git.causal.agency/catgirl "catgirl"
+.El
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/010-irc-suite.7 b/www/text.causal.agency/010-irc-suite.7
new file mode 100644
index 00000000..515a30ab
--- /dev/null
+++ b/www/text.causal.agency/010-irc-suite.7
@@ -0,0 +1,409 @@
+.Dd June 19, 2020
+.Dt IRC-SUITE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm IRC suite
+.Nd my own IRC software
+.
+.Sh DESCRIPTION
+Over the past months
+.Po
+eight of them, according to
+.Xr git-log 1
+.Pc
+I developed a new
+.Dq suite
+of IRC software
+that I now use full-time,
+consisting of a bouncer,
+a new logging and search solution,
+and a terminal client.
+These new programs share some characteristics:
+they are all TLS-only
+and use the libtls API from LibreSSL,
+they can all be entirely configured from the command line
+or with equivalent configuration files,
+they are all designed as
+a one process to one IRC connection mapping,
+and they all take advantage of IRCv3 features.
+.
+.Pp
+For context,
+I was previously running
+the znc IRC bouncer
+and using the Textual IRC client
+with its plain text logs.
+I also continue to use
+the Palaver IRC client for iOS.
+.
+.Ss Background
+A bouncer is a piece of server software
+that stays connected to IRC at all times
+and acts as a relay
+between your client and the IRC server.
+When the client is disconnected,
+the bouncer buffers incoming messages
+to send to the client when it reconnects.
+.
+.Pp
+Aside from this,
+bouncers have another advantage:
+client multiplexing.
+Several clients,
+for instance on different computers
+or a phone,
+should be able to connect to the same bouncer,
+and send and receive messages under the same nick.
+Unfortunately,
+znc does not handle this use-case well at all.
+Out of the box it offers two options:
+either any client connection totally clears the buffer,
+causing other clients to miss chat history;
+or the buffer is never cleared,
+causing every client connection
+to be repeatedly spammed with redundant history.
+There is also a znc wiki page
+that suggests one way to solve this issue
+is to connect znc to itself multiple times over.
+Yikes.
+.
+.Ss pounce
+My dissatisfaction with
+connecting multiple clients to znc
+directly motivated me to start working
+on a new multi-client-focused IRC bouncer.
+The result is
+.Xr pounce 1 ,
+based on a rather straightforward
+single-producer (the IRC server)
+multiple-consumer (each IRC client)
+ring buffer.
+Each client has its own
+independent position in the buffer
+so can properly be brought up to date
+whenever it connects.
+.
+.Pp
+Additionally,
+by assuming support for the IRCv3
+.Sy server-time
+extension,
+all IRC events can be accurately
+relayed to clients at any time,
+and the internals of
+.Xr pounce 1
+can be kept very simple.
+In fact,
+it completely avoids parsing most IRC messages,
+simply pushing them into the buffer
+with an associated timestamp.
+.
+.Pp
+The usernames sent by clients during registration
+are used as opaque identifiers for buffer consumers.
+This was chosen since most clients
+can be configured to send an arbitrary username,
+and those that don't often default
+to the name of the client itself,
+making it an appropriate identifier.
+.
+.Pp
+Later,
+I added a way for clients
+to be informed of their own buffer positions
+using a vendor-specific IRCv3 capability.
+This means a client
+can save the position
+of the last message it actually received,
+and request to set its position
+when it reconnects,
+ensuring no messages are lost
+to network issues
+or software crashes.
+.
+.Ss calico
+Due to the simple design of mapping
+one process to one IRC (server) connection,
+it is necessary to run several instances of
+.Xr pounce 1 .
+Initially I simply used different ports for each,
+but as I connected to more networks
+and even ran some instances for friends,
+it became less feasible.
+.
+.Pp
+The solution I came up with
+was to dispatch incoming connections
+using Server Name Indication, or SNI.
+This way,
+multiple domains pointing to the same host
+could be used with only one port
+to connect to different instances of
+.Xr pounce 1 .
+For example,
+I use a
+.Li *.irc.causal.agency
+wildcard DNS entry
+and a subdomain for each IRC network,
+all on port 6697.
+.
+.Pp
+The
+.Xr calico 1
+daemon included with pounce
+accomplishes this dispatch
+using the
+.Dv MSG_PEEK
+flag of
+.Xr recvmsg(2)
+on incoming connections.
+Since SNI is immediately sent by TLS clients
+as part of the ClientHello message in clear-text,
+it can be processed
+without doing any actual TLS.
+The connection itself is then
+sent to the corresponding
+.Xr pounce 1
+instance
+over UNIX-domain socket,
+which handles TLS as normal.
+This means that
+.Xr calico 1
+and
+.Xr pounce 1
+operate entirely independently of each other.
+.
+.Ss litterbox
+Based on the multiple-consumer ring buffer design,
+I realized it would be easy
+to implement additional functionality
+as independent purpose-built clients
+which connect to
+.Xr pounce 1
+alongside regular clients.
+This could allow dedicated OTR or DCC software
+to operate in parallel with a basic client,
+or for more passive software
+to provide notifications
+or dedicated logging.
+.
+.Pp
+For the latter,
+I wanted to do better than
+plain text log files.
+.Xr grep 1
+over files works fine,
+but search could be faster and smarter,
+and the text format is
+more lossy and less structured
+than I'd like it to be.
+Conveniently,
+SQLite provides an extension
+(actually two)
+for full-text search.
+.
+.Pp
+The litterbox project
+is my dedicated logging solution
+using SQLite FTS5.
+It consists of three tools:
+the
+.Xr litterbox 1
+daemon itself which connects to pounce
+and logs messages to SQLite,
+the
+.Xr scoop 1
+command line query tool,
+and the
+.Xr unscoop 1
+plain text import tool.
+The
+.Xr scoop 1
+tool constructs SQL queries
+and formats the results for viewing,
+with coloured nicks
+and piped to a pager
+by default.
+.
+.Pp
+The
+.Xr litterbox 1
+daemon
+can also provide a simple
+.Dq online
+.Pq over IRC
+search query interface
+to other connected clients.
+The simplest way to allow different
+.Xr pounce 1
+clients to talk to each other
+was to route private messages to self
+internally without sending them to the IRC server.
+So from any client
+I can simply message myself
+a full-text search query
+and
+.Xr litterbox 1
+responds with the results.
+.
+.Pp
+Along with routing self-messages,
+.Xr pounce 1
+also provides a vendor-specific IRCv3 capability
+for passive clients such as
+.Xr litterbox 1
+to indicate that they should not influence
+the automatic away status,
+which is normally only set
+when no clients are connected.
+.
+.Pp
+An advantage of this architecture
+of dedicated clients
+rather than bouncer modules
+is that they need not run
+on the same host.
+I run my bouncers on a VPS,
+but I'd rather not store my private logs there,
+so
+.Xr litterbox 1
+runs instead on a Raspberry Pi
+in my apartment.
+Also,
+since it is essentially
+just a regular IRC bot,
+it could be used independently
+for keeping public logs for a channel.
+.
+.Ss catgirl
+There's not really that much to say
+about the client,
+.Xr catgirl 1 .
+Of the three projects
+it contains the most code
+but is also the least interesting,
+in my opinion.
+It just does what I want a client to do,
+and gets the details right.
+.
+.Pp
+Tab complete is ordered by most recently seen or used,
+and completing several nicks
+inserts commas between them
+as well as the colon following the final nick.
+In the input line,
+the prompt is updated
+to reflect whether the input
+will be interpreted as a command or as a message.
+Messages are automatically scanned for URLs,
+which can be opened or copied with commands
+specifying the nick or a substring of the URL.
+.
+.Pp
+Scrolling in a window creates a split view,
+keeping the latest messages visible.
+Nick colours are based instead on usernames,
+keeping them more stable across renames,
+and mentions in messages are coloured
+to make the conversation easier to follow.
+The visibility of ignored messages
+can be toggled at any time.
+Channels can be muted
+so their activity is hidden
+from the status line
+unless you are pinged.
+.
+.Pp
+.Xr catgirl 1
+is configured entirely on the command line
+or in equivalent simple configuration files.
+There's no dynamic manipulation of state
+using complex
+.Ql /
+commands like in some other clients.
+.
+.Pp
+The major caveat is that
+.Xr catgirl 1
+connects to only one network at a time.
+This keeps the configuration, the interface
+and the code much simpler.
+.Xr tmux 1 ,
+.Xr screen 1
+or a tabbed terminal emulator
+are good options to run several instances.
+.
+.Pp
+If you're interested in giving
+.Xr catgirl 1
+a quick (and necessarily limited) try,
+you can
+.Li ssh chat@ascii.town .
+.
+.Ss Future
+I think I'm done with IRC software for now.
+As mentioned above,
+there are a few more pieces
+that could fit in to this setup,
+but I don't really want or need them right now.
+One thing I definitely want to try
+at some point
+is adding a litterbox component
+to index the contents of URLs
+to make finding previously shared links easier.
+.
+.Pp
+If you try any of this software
+and have feedback,
+let me know in
+.Li #ascii.town
+on tilde.chat
+or by email.
+And of course,
+patches are always welcome.
+.
+.Ss Update: scooper
+Somehow I had the motivation
+to create a web interface for litterbox:
+.Xr scooper 1 .
+It can be used either as CGI
+or as a FastCGI worker,
+and I used the excellent
+.Xr kcgi 3
+library for it.
+.
+.Pp
+The main advantage of this interface
+is that you can click on a search result
+to be brought to its context in the log viewer.
+I also added an option to
+.Xr litterbox 1
+to provide a corresponding scooper link
+in response to its query interface.
+.
+.Pp
+A small demo of scooper is hosted at
+.Aq Lk "https://causal.agency/scooper/" .
+It publicly logs the
+.Li #litterbox
+channel on tilde.chat.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk "https://git.causal.agency/pounce" pounce
+.It
+.Lk "https://git.causal.agency/litterbox" litterbox
+.It
+.Lk "https://git.causal.agency/catgirl" catgirl
+.It
+.Lk "https://www.sqlite.org/fts5.html" "SQLite FTS5 Extension"
+.It
+.Lk "https://git.causal.agency/scooper" scooper
+.It
+.Lk "https://kristaps.bsd.lv/kcgi/" kcgi
+.El
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/011-libretls.7 b/www/text.causal.agency/011-libretls.7
new file mode 100644
index 00000000..c29c325e
--- /dev/null
+++ b/www/text.causal.agency/011-libretls.7
@@ -0,0 +1,220 @@
+.Dd August  9, 2020
+.Dt LIBRETLS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm LibreTLS
+.Nd libtls for OpenSSL
+.
+.Sh DESCRIPTION
+This is a sort of announcement post about LibreTLS,
+my port of libtls from LibreSSL to OpenSSL.
+If you've wanted to try any of my software
+but have been unable to because of LibreSSL,
+LibreTLS is an option that will likely work for you.
+I'm including instructions
+for building it and my IRC software
+on Debian as an example,
+since manually installing libraries
+is less straightforward than it could be.
+.
+.Pp
+libtls is
+.Do
+a new TLS library,
+designed to make it easier to write foolproof applications
+.Dc .
+It was developed as part of LibreSSL,
+.Ox Ap s
+fork of OpenSSL,
+and is implemented against their version of libssl.
+It provides a nice high-level API
+for TLS sockets,
+with functions like
+.Xr tls_connect 3 ,
+.Xr tls_read 3
+and
+.Xr tls_write 3 .
+This is a vast improvement over libssl's
+confusing mess of an API!
+Its relative obscurity is a real shame
+for C programmers.
+.
+.Pp
+An obvious cause of its obscurity
+is that it is tied to LibreSSL.
+Although LibreSSL is available
+for platforms other than
+.Ox ,
+it conflicts with OpenSSL
+so is difficult to install alongside it
+and is often not packaged at all.
+Additionally,
+even if a user manually installs LibreSSL,
+libtls is likely not to work on some distros
+due to its hardcoded CA bundle file path.
+.
+.Pp
+Since libtls is implemented against libssl,
+which originates in OpenSSL,
+it should be possible to use libtls with it.
+This is what I set out to do in LibreTLS.
+I started by importing the sources
+from a LibreSSL-portable release,
+then worked on porting the portions
+that were incompatible with OpenSSL.
+.
+.Pp
+The simpler changes just involved
+replacing internal struct field accesses
+with public APIs.
+libtls accesses libssl internals
+using a hack to get the header files
+to declare private struct fields,
+and for basically no reason.
+The bigger changes involved
+reimplementing some functions
+which only exist in LibreSSL,
+but these were still quite small.
+I also imported the necessary compatibility functions
+from LibreSSL's libcrypto
+and adapated the autotools build files
+to produce only a libtls
+which depends on OpenSSL.
+.
+.Pp
+Along the way
+I decided to make one small behavioural change
+in order for LibreTLS to be more likely
+to work for everyone.
+I removed the hardcoded CA file path
+and changed the default configuration
+to use OpenSSL's default CA paths,
+which include a CA directory.
+This seems to be the preferred CA source
+on systems such as Debian,
+where the default CA file path doesn't exist.
+.
+.Pp
+I think the reason LibreSSL
+wants to avoid using a CA directory
+is so that it can fully load the CA file
+once before being sandboxed.
+However,
+using OpenSSL's default configuration,
+the CA file will still be loaded immediately
+if it exists.
+If it doesn't exist,
+sandboxed applications
+will fail when trying to
+load certificates from the directory,
+but unsandboxed applications
+will work just fine.
+Since LibreSSL's libtls
+would fail either way,
+I think the new behaviour
+is an improvement.
+.
+.Pp
+Another advantage of separating libtls from LibreSSL
+is that it is unencumbered by OpenSSL's
+awkward double-license,
+both of which are incompatible with the GPL.
+libtls is all new ISC-licensed code,
+and future versions of OpenSSL (3.0)
+will be released under the Apache 2.0 license,
+which is compatible with GPLv3.
+In the future,
+GPL software will be able to link with
+libtls and OpenSSL without additional permissions.
+.
+.Pp
+It's also worth noting that LibreSSL
+likely will not be able to import any code
+from future versions of OpenSSL,
+since Apache 2.0 is on
+.Ox Ap s
+license shitlist.
+LLVM is also slowly changing their license
+to Apache 2.0,
+so it'll be interesting to see what
+.Ox
+does.
+.
+.Ss Installing Manually
+To install LibreTLS on Debian,
+for example,
+fetch a release tarball from
+.Lk https://causal.agency/libretls/
+and install the build dependencies:
+.Bd -literal -offset indent
+sudo apt-get install build-essential libssl-dev pkgconf
+.Ed
+.
+.Pp
+.Xr pkgconf 1
+isn't a dependency of LibreTLS itself,
+but it's how my software
+configures its build
+for a dependency on libtls.
+The usual build steps
+will install the library:
+.Bd -literal -offset indent
+\&./configure
+make all
+sudo make install
+.Ed
+.
+.Pp
+The library will be installed in
+.Pa /usr/local/lib
+by default,
+and you need to make sure
+the dynamic linker
+will be able to find it there.
+On Debian,
+.Pa /usr/local/lib
+already appears in
+.Pa /etc/ld.so.conf.d/libc.conf ,
+but on other systems
+you'll probably need to add it to either
+.Pa /etc/ld.so.conf
+or a new file such as
+.Pa /etc/ld.so.conf.d/local.conf .
+Once the library is installed
+and the path is configured,
+the linker cache needs to be refreshed:
+.Bd -literal -offset indent
+sudo ldconfig
+.Ed
+.
+.Pp
+You'll probably also need to set
+.Ev PKG_CONFIG_PATH
+for the configure scripts
+of my software:
+.Bd -literal -offset indent
+PKG_CONFIG_PATH=/usr/local/lib/pkgconfig ./configure
+.Ed
+.
+.Pp
+On
+.Fx ,
+LibreTLS and some of my IRC software
+can be installed from my own
+.Lk https://git.causal.agency/ports/ "ports tree"
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://git.causal.agency/libretls/about LibreTLS
+.It
+.Lk https://man.openbsd.org/tls_init.3 "libtls API documentation"
+.El
+.
+.Pp
+Another alternative libtls implementation,
+.Lk https://sr.ht/~mcf/libtls-bearssl/ "libtls-bearssl"
+.
+.Sh AUTHORS
+.An June Bug Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/012-inability.7 b/www/text.causal.agency/012-inability.7
new file mode 100644
index 00000000..d352143b
--- /dev/null
+++ b/www/text.causal.agency/012-inability.7
@@ -0,0 +1,39 @@
+.Dd November 26, 2020
+.Dt INABILITY 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Inability
+.Nd losing the ability to create
+.
+.Sh DESCRIPTION
+For often weeks, sometimes months at a time,
+I lose the ability to write new code.
+I can still make fixes
+and little cleanups
+in my existing projects,
+but if I try to work on something new,
+nothing happens.
+I can't get anything done.
+.
+.Pp
+I think it's now been
+over 3 months
+since I've created anything.
+I don't know what to do about it.
+In the past I've eventually
+regained the ability to code,
+but it's unclear to me how or why.
+I also don't know what
+I should be doing instead.
+Writing code is the only hobby
+I've ever really developed,
+so without it I basically
+don't do anything.
+.
+.Pp
+Does this happen to anyone else?
+How do you cope?
+.
+.Sh AUTHORS
+.Mt june@causal.agency
diff --git a/www/text.causal.agency/013-hot-tips.7 b/www/text.causal.agency/013-hot-tips.7
new file mode 100644
index 00000000..63b6e353
--- /dev/null
+++ b/www/text.causal.agency/013-hot-tips.7
@@ -0,0 +1,156 @@
+.Dd December  2, 2020
+.Dt HOT-TIPS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm hot tips
+.Nd from my files
+.
+.Sh DESCRIPTION
+This is a short list of tips
+from my configuration files and code
+that might be useful.
+.
+.Ss Shell
+.Bl -tag -width Ds
+.It CDPATH=:~
+This is useful if you sometimes type,
+for example,
+.Ql cd src/bin
+wanting to go to
+.Pa ~/src/bin
+but you aren't in
+.Pa ~ .
+If the path doesn't exist
+in the current directory,
+.Ic cd
+will try it in
+.Pa ~
+as well.
+.
+.It alias ls='LC_COLLATE=C ls'
+This makes it so that
+.Xr ls 1
+lists files in ASCIIbetical order,
+which puts capitalized names like
+.Pa README
+and
+.Pa Makefile
+first.
+.
+.It git config --global commit.verbose true
+Not shell but close enough.
+This makes it so the entire diff is shown
+below the usual summary
+in the editor for a
+.Xr git-commit(1)
+message.
+Useful for doing a quick review
+of what you're committing.
+.El
+.
+.Ss (neo)vim
+.Bl -tag -width Ds
+.It set inccommand=nosplit
+This is the only
+.Xr nvim 1
+feature I really care about
+aside from the improved defaults.
+This provides a live preview of what a
+.Ic :s
+substitution command will do.
+It makes it much easier to
+write complex substitutions.
+.
+.It nmap <leader>s vip:sort<CR>
+This mapping sorts the lines of a paragraph,
+or block of text separated by blank lines.
+I use this a lot to sort
+#include directives.
+.
+.It nmap <leader>S $vi{:sort<CR>
+Similar to the last mapping,
+this one sorts lines inside braces.
+I use this to sort
+switch statement cases
+or array initializers.
+.
+.It nmap <leader>a m':0/^#include <<CR>:nohlsearch<CR>O#include <
+I use this mapping to add new
+#include directives,
+usually followed by
+.Ic <leader>s
+and
+.Ic ''
+to sort them
+and return to where I was.
+.
+.It nmap <leader>d :0delete<CR>:0read !date +'.Dd \e%B \e%e, \e%Y'<CR>
+I use this to replace the first line of
+.Xr mdoc 7
+files with the current date.
+.El
+.
+.Ss C
+.Bl -tag -width Ds
+.It #define Q(...) #__VA_ARGS__
+This is what I've started using
+to quote things like SQL statements
+or HTML fragments in C.
+Anything that happens to be valid C tokens,
+which is most code,
+can be quoted this way.
+Macros are not expanded
+inside the quoted part.
+You can embed (matched) quotes
+without having to escape them.
+Whitespace gets collapsed,
+so you can write nicely formatted multi-line SQL
+that doesn't mess up your debug logging,
+for example.
+.Bd -literal -offset indent
+const char *sql = Q(
+	INSERT OR IGNORE INTO names (nick, user, host)
+	VALUES (:nick, :user, :host);
+);
+.Ed
+.
+.It #define BIT(x) x##Bit, x = 1 << x##Bit, x##Bit_ = x##Bit
+I use this macro to declare bitflag enums.
+It takes advantage of
+auto-incrementing enum items
+so you don't need to set the values manually.
+You also get constants
+for both the bit index
+and the flag value
+for each item.
+.Bd -literal -offset indent
+enum Attr {
+	BIT(Bold),
+	BIT(Reverse),
+	BIT(Italic),
+	BIT(Underline),
+};
+.Ed
+.Pp
+For example,
+defines
+.Sy ItalicBit = 2
+and
+.Sy Italic = 1 << 2 .
+Ignore the extraneous constants.
+.
+.It typedef int FnType(const char *str, size_t len);
+You can just typedef function types!
+It annoys me more than it probably should
+that everyone writes ugly
+function pointer typedefs.
+Just stick
+.Sy typedef
+on the front of a function declaration
+and use
+.Vt FnType * .
+.El
+.
+.Sh AUTHORS
+.Mt june@causal.agency
diff --git a/www/text.causal.agency/014-using-vi.7 b/www/text.causal.agency/014-using-vi.7
new file mode 100644
index 00000000..e6a6a7a0
--- /dev/null
+++ b/www/text.causal.agency/014-using-vi.7
@@ -0,0 +1,135 @@
+.Dd January 11, 2021
+.Dt USING-VI 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Using vi
+.Nd simpler tools
+.
+.Sh DESCRIPTION
+Happy new year
+and hello from
+.Xr vi 1 !
+I'm in the mood to post something
+but not in the mood for
+.Dq social
+media.
+This one will probably be short.
+.
+.Pp
+Yesterday I was trying to work on sandboxing
+.Xr catgirl 1
+(that's the IRC client I work on)
+with
+.Xr pledge 2
+and
+.Xr unveil 2
+on
+.Ox ,
+as suggested by the maintainer of its port.
+I've done similar things before,
+but only on server software
+rather than user software.
+.
+.Pp
+Anyway I was in
+.Xr ssh 1
+to my
+.Ox
+VM
+.Po
+sadly I don't currently have any hardware to run
+.Ox
+on
+.Pc
+using my usual editor,
+which is
+.Xr nvim 1 .
+I'm honestly not very thrilled
+with what neovim is doing lately,
+but the cleaned up defaults
+make my configuration files happier.
+.
+.Pp
+The real problem with
+.Xr nvim 1 ,
+though,
+is that it's laggy as hell on
+.Ox .
+There is significant delay
+on every single keystroke,
+as if I'm typing remotely to a server
+on the other side of the world,
+but this is on a local VM!
+.
+.Pp
+So I did the only reasonable thing:
+I typed
+.Sy :qa
+followed by
+.Sy vi .
+The difference was astonishing.
+Typing and editing suddenly felt
+.Em physical
+again.
+(I put that in italics even though I know it won't render.)
+Not only was it a vast improvement over
+.Xr nvim 1
+in
+.Xr ssh 1
+in a VM,
+it was a marked improvement over
+.Xr nvim 1
+running locally and natively.
+.
+.Pp
+Now obviously
+.Xr vi 1
+doesn't have all the bells and whistles
+of newer editors,
+but of course the core editing model
+that makes
+.Xr vim 1
+and
+.Xr nvim 1
+so good is there,
+and in purer form,
+I think.
+The
+.Xr vi 1
+manual page
+is feasible to just sit down and read,
+and learn everything there is to know about the editor.
+I set up a basic configuration
+and got coding.
+.Bd -literal -offset indent
+export EXINIT='set ai ic sm sw=4 ts=4'
+.Ed
+.
+.Pp
+After I finished my
+.Xr pledge 2
+and
+.Xr unveil 2
+patch,
+I was so pleased with
+.Xr vi 1
+that I kept on using it
+yesterday and today
+for other work,
+and obviously to write this post.
+Despite the lack of editor amenities,
+its responsiveness and simplicity
+are enough to make using it
+.Em comfortable
+and perhaps
+.Em cosy .
+I'm not sure I'll ever use
+.Xr vi 1
+full-time,
+but for now I am much less likely
+to launch
+.Xr nvim 1 .
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/015-reusing-tags.7 b/www/text.causal.agency/015-reusing-tags.7
new file mode 100644
index 00000000..19546496
--- /dev/null
+++ b/www/text.causal.agency/015-reusing-tags.7
@@ -0,0 +1,155 @@
+.Dd January 17, 2021
+.Dt REUSING-TAGS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm reusing tags
+.Nd beyond ctags
+.
+.Sh DESCRIPTION
+I've tried to start writing this post a couple times now
+and I keep getting bogged down in explanations,
+so I'm just going to tell you
+about some cool things I did
+and hope they make sense.
+.
+.Pp
+When I wrote my first syntax highlighter,
+I decided that function definitions
+should have anchor links,
+because line number anchor links
+are entirely useless
+if you expect the file to change at all.
+Since the syntax highlighter
+was somewhat deliberately just a big pile of regex,
+I hacked in more regex to try
+to identify function and type definitions.
+It wasn't elegant and it didn't always work well.
+It did work though,
+and I found the links very useful.
+.
+.Pp
+Recently I was thinking about
+the lexer generator
+.Xr lex 1
+and decided to
+rewrite the syntax highlighter
+using it.
+Really syntax highlighting
+is no different than lexical analysis.
+I ran into a problem though,
+trying to preserve my anchor link function,
+because really that should involve
+some amount of parsing.
+Trying to port my regex hacks to
+.Xr lex 1
+made the lexers way more complicated
+and less reliable,
+so I gave up on it for a while.
+.
+.Pp
+And then,
+probably in the shower,
+I realized I was approaching it
+completely from the wrong direction.
+There's already a tool that does what I want,
+and I already use it:
+.Xr ctags 1 .
+All I need to do is use its output
+to insert anchor links
+into my syntax highlighter output.
+In an afternoon I wrote
+.Xr htagml 1 ,
+which loads tag definitions for its input file,
+then scans through the input for where they match.
+It can either HTML-escape
+the input as it goes,
+or use already formatted HTML
+being piped into it from a syntax highlighter.
+.
+.Pp
+The result is three simple tools
+working together to accomplish
+what a more complex tool
+couldn't reliably achieve.
+I'm very pleased with it,
+and I've updated my site and cgit
+to use the new
+.Xr lex 1 Ns -based
+highlighter,
+.Xr ctags 1
+and
+.Xr htagml 1 .
+I'm currently missing a lexer for
+.Xr sh 1 ,
+but I plan to write it eventually.
+I also want to write a tool
+to generate tags for
+.Xr make 1 ,
+.Xr mdoc 7
+and perhaps
+.Xr sh 1 .
+The cool thing about generating more kinds of tags
+is that they'll not only improve
+the HTML output,
+they'll also be usable in my editor.
+.
+.Pp
+Speaking of generating different kinds of tags,
+I also wrote some scripts not too long ago
+for reading IETF RFCs offline.
+The plain text files are available to
+.Xr rsync 1 ,
+but they're hard to navigate on their own.
+By scanning the files for headings
+and generating tags,
+it allows jumping to sections using
+.Ic :ta
+or
+.Ic ^]
+in
+.Xr vi 1 .
+For
+.Xr nvim 1
+I also added an
+.Ic :RFC
+command to open an RFC by number
+and set up
+.Ic ^]
+to work optimally for them.
+.
+.Pp
+I'm still using
+.Xr vi 1
+for most of my editing,
+by the way.
+And of course
+.Xr ctags 1
+was made to work with it!
+Simple old tools
+are really doing it for me lately.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://causal.agency/bin/htagml.html htagml
+.It
+.Lk https://causal.agency/bin/hilex.html hilex
+.It
+.Lk https://git.causal.agency/src/tree/doc/rfc rfctags
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Sh ADDENDUM
+.Xr catgirl 1 ,
+.Xr pounce 1 ,
+.Xr litterbox 1
+and
+.Xr scooper 1
+all have new releases,
+if you're using any of them.
+Also, this space is now
+available over gopher,
+if that's your sort of thing.
diff --git a/www/text.causal.agency/016-using-openbsd.7 b/www/text.causal.agency/016-using-openbsd.7
new file mode 100644
index 00000000..b843e3c3
--- /dev/null
+++ b/www/text.causal.agency/016-using-openbsd.7
@@ -0,0 +1,505 @@
+.Dd February 14, 2021
+.Dt USING-OPENBSD 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Using OpenBSD
+.Nd for real
+.
+.Sh DESCRIPTION
+Hello from
+.Ox !
+After wishing one too many times
+that I had a real BSD
+on a physical machine,
+I finally got around to
+just installing one on my
+mid-2014 MacBook Pro.
+I hadn't done it sooner
+because I didn't realize
+how easy it would be.
+It helped that I already had a
+.Dq Boot Camp
+partition with a disused Windows 8 install
+that I could replace.
+.
+.Pp
+I roughly followed an old jcs gist
+along with the
+.Ox
+Disk Setup guide.
+I'm once again happy
+that I bought a printer\(em
+they're very useful for instructions
+to install an operating system
+on your only usable computer.
+I set up encrypted softraid
+and the operating system
+installed smoothly.
+.
+.Pp
+Next I had to install rEFInd,
+since the default Mac boot manager
+is really not keen on booting much.
+Installing it requires using the
+macOS recovery partition these days.
+But there was a problem
+with my new boot menu:
+I was promised a picture of Puffy,
+and instead I just got some abstract coloured circles!
+Turns out a bunch of OS icons
+got removed from rEFInd at some point,
+and I had to rescue Puffy
+from the git history.
+.
+.Pp
+So I could happily boot
+.Ox
+by selecting Puffy,
+but I had no networking.
+I thought the wifi chip might be supported by
+.Xr bwfm 4 ,
+but I got unlucky and it's a BCM4360,
+which everything hates.
+Based on the jcs gist,
+I checked the list of hardware
+supported by the
+.Xr urtwn 4
+driver for a wifi dongle to order.
+Just having a clear list
+in the driver manual is wonderful.
+I went with the Edimax EW-7811Un v2,
+which I could get for around $20.
+It's nice and tiny,
+though it has a piercing blue LED
+(destroy all blue LEDs)
+which I had to cover with electrical tape.
+.
+.Pp
+I had to do one other thing
+before I could get it all working, though.
+When I had checked the
+.Xr urtwn 4
+hardware list,
+I had been looking at
+.Ox Ns -current ,
+but I had installed
+.Ox 6.8 ,
+and support for the v2 hardware
+I had bought was added after that release.
+So I downloaded a snapshot
+.Pa bsd.rd
+along with the
+.Xr urtwn 4
+firmware file
+to a USB drive
+and upgraded the system.
+.
+.Pp
+Connecting to wifi with
+.Xr ifconfig 8
+is a breeze, by the way,
+and then you just write the same thing to a
+.Xr hostname.if 5
+file to make it automatic.
+I wanted to use
+.Ox
+for exactly this reason:
+simple, consistent, cohesive, well-documented tools.
+.
+.Pp
+Finally, I got to configuring.
+The console is configured with
+.Xr wsconsctl 8 ,
+and similarly you can put the commands in
+.Xr wsconsctl.conf 5
+to have them run at boot.
+I added
+.Li display.brightness=50%
+to tone down the brightness,
+which is initially 100%,
+and
+.Li keyboard.backlight=0%
+to turn off those annoying lights.
+.Xr wsconsctl.conf 5
+is also where you can set
+trackpad settings if you're not using
+.Xr synaptics 4 .
+I ended up using:
+.Bd -literal -offset indent
+mouse1.tp.tapping=1
+mouse1.tp.scaling=0.2
+mouse1.reverse_scrolling=1
+.Ed
+.Pp
+This enables tapping with several fingers
+to simulate different mouse buttons,
+makes the cursor move at a reasonable speed
+and scrolling move in the right direction.
+I also set up my usual modified QWERTY layout.
+.
+.Pp
+For
+.Xr X 7
+I had enabled
+.Xr xenodm 1 ,
+which seems quite nice.
+It automatically prompts you to add your
+.Xr ssh 1
+keys to
+.Xr ssh-agent 1
+when you log in.
+One of the reasons I had not wanted
+to set up another graphical system
+is that I thought
+I would have to make too many choices,
+and that I would have to choose least bad options
+rather than actually good options,
+but
+.Ox
+already includes reasonable choices.
+I wanted to use
+.Xr cwm 1 ,
+so I started a basic
+.Pa .xsession
+file:
+.Bd -literal -offset indent
+\&. ~/.profile
+export LC_CTYPE=en_US.UTF-8
+xset r rate 175 m 5/4 0
+xmodmap ~/.config/X/modmap
+xrdb -load ~/.config/X/resources
+exec cwm -c ~/.config/cwm/cwmrc
+.Ed
+.
+.Pp
+The
+.Xr xset 1
+command sets keyboard repeat rate
+and mouse acceleration.
+I spent some time going through
+.Xr cwm 1 Ap s
+functions and writing up bindings
+that would get me something close enough
+to what I'm used to in macOS.
+Most importantly,
+putting everything on the 4 modifier (command key).
+.
+.Pp
+I also added key bindings on F1 and F2
+to adjust the brightness with
+.Xr xbacklight 1 ,
+and on F10, F11 and F12
+to adjust volume with
+.Xr sndioctl 1 .
+I'm not sure why the F keys
+just send regular F1, F2, etc.\&
+regardless of the Fn key.
+I don't use F keys for anything else though,
+so I'm not too concerned.
+Once again,
+.Xr sndioctl 1
+is such an easy straightforward tool:
+.Bd -literal -offset indent
+bind-key F10 "sndioctl output.mute=!"
+bind-key F11 "sndioctl output.level=-0.05"
+bind-key F12 "sndioctl output.level=+0.05"
+.Ed
+.
+.Pp
+For aesthetic configuration,
+I added a new output to my
+.Xr scheme 1
+colour scheme tool for
+.Xr X 7 Ns -style
+RGB and
+.Xr xterm 1
+resources.
+Normally I use the
+.Em Go Mono
+font,
+but since
+.Ox
+already includes
+.Em Luxi Mono ,
+which
+.Em Go Mono
+is based on,
+I used that.
+The most important configuration
+to make anything readable on a high-DPI display is:
+.Bd -literal -offset indent
+Xft.dpi: 144
+Xft.antialias: true
+Xft.hinting: false
+.Ed
+.
+.Pp
+I'm annoyed that I haven't found
+where these resources are actually documented.
+I would hope they'd be in
+.Xr Xft 3
+or something,
+but they're not.
+Anyway,
+turning off hinting
+seems absolutely necessary
+to prevent text from looking like garbage.
+.
+.Pp
+It seems that to get a reasonably sized cursor
+I need to install
+.Sy xcursor-dmz .
+I'd prefer if there wasn't this one
+extra package that I needed
+for a reasonable setup.
+Tangentially,
+I've never understood why
+the black versions of dmz cursors
+are called
+.Dq aa
+when it seems like that
+would stand for antialiasing
+or something.
+.Bd -literal -offset indent
+Xcursor.size: 64
+Xcursor.theme: dmz-aa
+.Ed
+.
+.Pp
+For a desktop background,
+I found a cute bitmap (little picture)
+of snowflakes already in the system
+and used colours from my usual scheme:
+.Bd -literal -offset indent
+xsetroot -bitmap /usr/X11R6/include/X11/bitmaps/xsnow \e
+	-bg rgb:14/13/0E -fg rgb:7A/49/55
+.Ed
+.
+.Pp
+Since I'd rather not install anything
+I don't have to,
+I went with the default
+.Xr xterm 1 .
+It seems more than adequate, honestly.
+I read through its RESOURCES
+section to configure it how I like.
+The important settings are
+.Sy XTerm*utf8
+and
+.Sy XTerm*metaSendsEscape .
+Since I'm used to copying and pasting on macOS,
+I added equivalent
+.Dq translations :
+.Bd -literal -offset indent
+XTerm*VT100*translations: #override \en\e
+	Super <Key>C: copy-selection(CLIPBOARD) \en\e
+	Super <Key>V: insert-selection(CLIPBOARD)
+.Ed
+.
+.Pp
+The next thing I needed
+was a clock and battery indicator.
+I actually had my battery die on me
+while I was doing all this,
+which reminded me.
+.Xr xclock 1
+would be perfect,
+but then I'd need something else
+for battery.
+There are a couple basic battery indicators
+for X in ports,
+but they're terribly ugly.
+I wanted something as simple as
+.Xr xclock 1 ,
+but that I could add some other text to.
+Then I realized I could just use
+.Xr xterm 1
+for that.
+To my
+.Pa xsession
+I added:
+.Bd -literal -offset indent
+xterm -name clock -geometry 14x1-0+0 -sl 0 -e clock &
+.Ed
+.Pp
+This places a little terminal
+in the top-right corner of the screen
+with no scrollback lines,
+running a script called
+.Pa clock .
+To have
+.Xr cwm 1
+treat it like a
+.Dq panel
+and show it on every desktop,
+I added this to my
+.Pa cwmrc :
+.Bd -literal -offset indent
+ignore clock
+autogroup 0 clock,XTerm
+.Ed
+.Pp
+The
+.Pa clock
+script simply uses
+.Xr date 1
+and
+.Xr apm 8
+to print the time and battery charge
+every minute:
+.Bd -literal -offset indent
+tput civis
+sleep=$(( 60 - $(date +'%S' | sed 's/^0//') ))
+while :; do
+	if [ $(apm -a) -eq 1 ]; then
+		printf '%3s%%' "$(apm -l)"
+	else
+		test $(apm -b) -eq 2 && tput setaf 1 bold
+		printf '%3.3sm' "$(apm -m)"
+		tput sgr0
+	fi
+	printf ' %s\r' "$(date +'%a %H:%M')"
+	sleep $sleep
+	sleep=60
+done
+.Ed
+.Pp
+The initial setting of
+.Va sleep
+is to align the updates
+with the minute ticking over.
+I made the battery output
+a bit fancier by showing
+percentage while charging,
+minutes left while discharging,
+and highlighting in red
+when the battery is
+.Dq critical .
+.
+.Pp
+Now is a good time to mention adding
+.Ql apmd_flags=-A
+to
+.Pa /etc/rc.conf.local
+to enable
+.Dq automatic performance adjustment ,
+or not running your battery flat
+as fast as possible mode.
+It seems like I can get up to 3 hours
+of battery life depending on the screen brightness,
+but this is quite an old battery by now.
+.
+.Pp
+The other thing I needed
+was something to tone down
+that awful, evil blue light from the screen.
+I asked around and someone told me about
+.Xr sct 1 ,
+originally written by tedu.
+The package also includes a little
+.Xr sctd 1
+script that you can add to your
+.Pa .xsession
+to have it automatically adjust
+the colour temperature throughout the day.
+My eyes are no longer being assaulted.
+.
+.Pp
+While I was doing all this,
+I of course needed to talk about it on IRC,
+and it was very nice to be able to
+install my own IRC client with:
+.Bd -literal -offset indent
+doas pkg_add catgirl
+.Ed
+.Pp
+I don't plan to do
+general Web Browsing on
+.Ox ,
+and there is definitely
+no good choice for browser,
+so I just installed
+.Xr imv 1 ,
+.Xr mpv 1 ,
+.Xr youtube-dl 1
+and
+.Xr w3m 1 .
+I wrote a script
+to open images by piping
+.Xr curl 1
+into
+.Xr imv 1 ,
+videos with
+.Xr mpv 1 ,
+and everything else with
+.Xr w3m 1
+in a new
+.Xr xterm 1 .
+Annoyingly,
+.Xr mpv 1
+seems incapable of exiting
+without segfaulting.
+That's quality.
+.
+.Pp
+One thing I am still missing
+is automatic brightness adjustment
+based on ambient light
+like macOS can do.
+I can read the sensor with
+.Xr sysctl 8
+.Cm hw.sensors.asmc0.illuminance0 ,
+which is measured in lux.
+I tried doing something with it in a script,
+but it seems tricky to map its value
+to brightness adjustments
+and to play nice with manual brightness changes,
+so I'll just keep doing it manually for now.
+.
+.Pp
+Update:
+prx sent mail to let me know about
+.Aq Lk https://github.com/jcs/xdimmer .
+I should've guessed jcs had written something.
+.
+.Pp
+And that's my current
+.Ox
+setup after a week of using it.
+I'm quite enjoying it,
+and still being pleasantly surprised
+by the quality-of-life from
+.Ox
+tools and documentation.
+For a small example,
+I can jump to sections
+or flag definitions in
+.Xr man 1
+using
+.Ic :t .
+Systems without basic usability like that
+should be ashamed.
+.
+.Pp
+I would post a screenshot,
+but this is
+.Li text.causal.agency
+;)
+.
+.Sh SEE ALSO
+.Lk https://gist.github.com/jcs/5573685
+.Pp
+My full configurations are in
+.Aq Lk https://git.causal.agency/src .
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Sh BUGS
+There's a red LED
+inside the headphone jack
+that is always on
+and I have no idea how to turn off.
+If anyone knows
+please send me an email.
diff --git a/www/text.causal.agency/017-unpasswords.7 b/www/text.causal.agency/017-unpasswords.7
new file mode 100644
index 00000000..f9643f2f
--- /dev/null
+++ b/www/text.causal.agency/017-unpasswords.7
@@ -0,0 +1,153 @@
+.Dd February 20, 2021
+.Dt UNPASSWORDS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Unpasswords
+.Nd password anti-management
+.
+.Sh DESCRIPTION
+Right away I want to say
+that I'm not trying to tell anyone
+how to manage their online authentication.
+This is just how I do it,
+and I haven't seen anyone else write about it.
+.
+.Pp
+I don't use a password manager.
+It's not a type of software
+I want to deal with.
+For the small handful of sites
+that I use regularly
+and that actually matter,
+I use strong passwords
+(stored in my noggin)
+and TOTP.
+For everything else,
+I simply do not know the password,
+and neither does any software.
+.
+.Pp
+I think I started doing this one time
+when I had legitimately forgotten
+the password to some old account.
+I clicked on
+.Dq forgot my password
+and opened the email,
+but I didn't want to
+come up with a new password
+I would just forget again.
+Instead I set a random one
+.Po
+I usually use
+.Ql openssl rand -base64 33
+for this
+.Pc
+and immediately used that to log in
+while it was still in my clipboard.
+Next time I wanted to log in,
+I could use
+.Dq forgot my password
+again.
+.
+.Pp
+Thinking about it,
+I realized that any web authentication
+with an email password reset flow
+is only ever as strong as
+the authentication for your email account.
+So what is the point of having
+all these passwords set on different sites?
+They all answer to your email account,
+and storing them in a password manager
+seems to add another potential point of failure.
+May as well have no other passwords at all,
+or as close as possible.
+.Po
+Shout out to sites like Liberapay
+and asciienema
+which let me not set a password at all.
+.Pc
+.
+.Pp
+So I started doing that for any site
+that I don't regularly log in to.
+Going through the password reset flow
+can be a bit slow,
+but it doesn't need to be done often.
+And I can do it from anywhere
+I have access to my email,
+which I feel is more easily reliable
+than syncing password management databases.
+It's quite stress-free.
+.
+.Pp
+After doing this manually for years,
+this week I finally got around to
+writing some automation for it.
+A while ago I had written
+.Xr imbox 1 ,
+a tool to directly export mail
+in mboxrd format from IMAP,
+along with
+.Xr git-fetch-email 1 ,
+a wrapper which offloads configuration to
+.Xr git-config 1 .
+It can match emails by
+Subject, From, To and Cc.
+This week I added a flag
+to use IMAP IDLE
+to wait for a matching message
+if there isn't one already,
+and a flag to move matching messages
+(for example to Trash)
+after exporting them.
+.
+.Pp
+With those two new flags,
+I started writing some shell scripts
+to automate the password reset flow
+using
+.Xr curl 1
+to submit forms and
+.Xr git-fetch-email 1
+with
+.Xr sed 1
+to pull the reset tokens
+from my inbox.
+At the end of the script,
+the random password it set
+is copied to the clipboard
+and the login page for the site is opened.
+So now logging in is as simple
+as running a command,
+waiting for the login page to open,
+and pasting.
+.
+.Pp
+The script isn't sophisticated,
+but I don't think it needs to be.
+I've written functions
+for a couple different sites already,
+and they all work in mostly the same way.
+Writing a new one is just a matter
+of identifying the form URLs and fields
+along with where the token is in the email.
+I'm not going to turn this automation
+into any kind of generally usable project,
+because I don't want to have to
+maintain functions for tonnes of different services.
+If you're interested in this idea,
+I encourage you to use my script as a template
+and implement the functions for services you use.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://git.causal.agency/imbox
+.It
+.Lk https://causal.agency/bin/sup.html
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/018-operating-systems.7 b/www/text.causal.agency/018-operating-systems.7
new file mode 100644
index 00000000..691102e2
--- /dev/null
+++ b/www/text.causal.agency/018-operating-systems.7
@@ -0,0 +1,86 @@
+.Dd February 22, 2021
+.Dt OPERATING-SYSTEMS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Operating systems
+.Nd criteria
+.
+.Sh DESCRIPTION
+Sometimes in conversation
+I use the term
+.Dq real operating system
+which people,
+perhaps rightfully,
+take as inflammatory.
+But I have actually thought about
+what I mean when I say
+.Dq real operating system
+and come up with
+this list of criteria.
+.
+.Pp
+An operating system should be...
+.Bl -bullet
+.It
+Consistent and cohesive:
+all parts of the system should have similar
+usage, configuration, documentation and so on.
+Parts of the system should naturally work together,
+because they were designed to do so.
+.
+.It
+Documented:
+the system should include its own documentation.
+A user should not have to
+search some external wiki
+to learn how the system works.
+It should be obvious
+where to find documentation
+on a particular topic.
+.
+.It
+Programmable:
+the system should provide
+a way to program the computer.
+A computer which cannot be programmed
+is not a computer at all.
+Usually this takes the form
+of a C compiler
+and the tools that go with it.
+In earlier times,
+it might have been
+a BASIC interpreter.
+.
+.It
+Examinable and modifiable:
+the full source tree
+for the system should be included,
+or easily obtainable
+through official means.
+A user should have no trouble
+finding the corresponding source
+for a part of the system.
+Together with the previous point,
+the source tree should be
+compiled by the included toolchain,
+allowing local modification.
+.El
+.
+.Pp
+Some things that may be parts
+of real operating systems,
+but are not themselves operating systems:
+a kernel,
+a package manager,
+a collection of packages.
+.
+.Pp
+I will leave it as an
+.Dq exercise for the reader
+to guess which operating systems
+meet these criteria
+and which don't.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/019-mailing-list.7 b/www/text.causal.agency/019-mailing-list.7
new file mode 100644
index 00000000..b3490a94
--- /dev/null
+++ b/www/text.causal.agency/019-mailing-list.7
@@ -0,0 +1,286 @@
+.Dd March  4, 2021
+.Dt MAILING-LIST 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Mailing List
+.Nd a small-scale approach
+.
+.Sh DESCRIPTION
+When I initially published
+some software I expected
+other people to use,
+I just asked that patches
+be mailed directly to me,
+but I figured that
+if more people were interested,
+it would be better
+to have a mailing list.
+Unfortunately
+email software,
+mailing list options in particular,
+are quite daunting.
+I wanted a light-weight option
+that would require me to host
+as little software as possible.
+.
+.Pp
+My regular email is hosted by Fastmail,
+and I poked around its settings
+to see what I could do.
+It turns out Fastmail lets you
+configure address aliases to
+.Dq also send to all contacts in
+a contacts group.
+That's a mailing list!
+I created a group called
+.Dq List
+and an alias called
+.Mt list@causal.agency
+configured to deliver to that group.
+So it's really just an alias
+for my regular address
+that happens to also
+deliver to another group of people.
+.
+.Pp
+It's easier to just configure
+and manage one mailing list,
+so what I do is ask patches and feedback
+to be sent to
+.Mt list+catgirl@causal.agency ,
+for example.
+Fastmail treats any
+.Ar +suffix
+the same as the base address,
+but the full address can be used
+by subscribers to filter mail by topic
+if they wish.
+.
+.Pp
+To subscribe someone to the list,
+I add their contact to the group.
+For a long time I was planning
+to write some software
+to manage these subscriptions.
+It should be possible
+to process subscription requests from IMAP
+and manipulate the contact group with CardDAV.
+When I went to start implementing this,
+however,
+I found CardDAV (and WebDAV in general)
+completely inscrutable.
+It's the kind of protocol
+that is split across like 20
+different RFCs
+and you can't understand anything
+by just reading
+the one you actually care about.
+So I've given up on that
+and will keep manually subscribing people
+on request.
+.
+.Pp
+The only thing missing, then,
+is a way for people to read
+mail sent to the list
+while they aren't subscribed.
+All the existing
+mailing list archive software
+I know of
+expects to have the mail locally,
+but I'd rather keep all my mail in IMAP.
+First,
+in order to make sure
+I keep a complete archive
+of the mailing list in IMAP,
+I added a small amount
+of Sieve code
+to my Fastmail filters configuration:
+.Bd -literal -offset indent
+if address :matches ["To", "Cc"] "list*@causal.agency" {
+	fileinto :copy :flags "\e\eSeen" "INBOX.List";
+}
+.Ed
+.
+.Pp
+Sieve is a small standard language
+specifically for filtering mail.
+This bit of code matches
+anything sent to the list
+and adds a copy of it
+(the original is going into my inbox)
+to the
+.Dq List
+folder
+and marks the copy as read.
+.
+.Pp
+With a pristine IMAP mailbox
+to export from,
+I wrote a new archive generator.
+It's called
+.Xr bubger 1
+kirg (have it in a way).
+My goal was to render directly from IMAP
+and produce only static files as output,
+making it not only easy to serve,
+but also to run in one place
+and copy the files elsewhere.
+That's important to me
+because it has access to my email,
+so I'd rather run it
+on my local network and
+.Xr rsync 1
+its output into The Cloud.
+The static files are in
+HTML, Atom and mboxrd formats.
+.
+.Pp
+The architecture of
+.Xr bubger 1
+is that for each piece of mail,
+identified by its UID in the mailbox,
+HTML and Atom fragments
+are exported along with the mboxrd.
+Those fragments are then stitched together
+using the IMAP SORT and THREAD extensions
+to make full pages and feeds
+for each thread.
+The fragments act as a cache
+for subsequent runs.
+.
+.Pp
+I admit I did some
+pretty questionable things
+to achieve this.
+Namely,
+I wrote a small string templating engine in C.
+I use it to produce the HTML
+and XML for Atom,
+as well as to generate URLs
+and paths.
+I'm really happy with how it works, actually.
+This is also where
+I really started using
+one of my favourite C hacks:
+.Bd -literal -offset indent
+#define Q(...) #__VA_ARGS__
+.Ed
+.
+.Pp
+I quote all my HTML/XML templates
+with this and it's lovely.
+.
+.Pp
+I've been working on
+.Xr bubger 1
+on and off for almost a year now,
+and it's been interesting.
+I learned a lot about how email
+works from having to deal with
+all the ways a message can be.
+Thankfully a lot of that dealing
+is done by the IMAP server.
+.
+.Pp
+As for running it,
+I initially just ran it with
+.Xr cron 8 ,
+and that's still a good way to go.
+To hook it up to
+.Xr rsync 1 ,
+pipe it like so:
+.Bd -literal -offset indent
+bubger -C list [...] | rsync -a --files-from=- list remote:list
+.Ed
+.
+.Pp
+Later,
+I got a little annoyed
+with having to wait
+for the next run
+if I wanted to link
+to some mail I just received.
+I added an option
+to use IMAP IDLE
+to wait for new mail continuously
+and I started running it
+under my process supervisor,
+.Xr catsitd 8 .
+.
+.Pp
+The setup is a little more complex
+to feed the list of updated files to
+.Xr rsync 1 .
+I added the
+.Xr catsit-watch 1
+utility to run a command
+when a file changes,
+and in my
+.Xr catsit.conf 5
+I have the following:
+.Bd -literal -offset indent
+bubger	~/.local/libexec/bubger
+rsync	catsit-watch -i -f ~/list/UIDNEXT ~/.local/libexec/rsync
+.Ed
+.
+.Pp
+The
+.Pa ~/.local/libexec/bubger
+script runs
+.Xr bubger 1 ,
+writing the list of updated paths to
+.Pa ~/list/FILES :
+.Bd -literal -offset indent
+exec bubger -i -C ~/list [...] >~/list/FILES
+.Ed
+.
+.Pp
+And the
+.Pa ~/.local/libexec/rsync
+script gets run each time a
+.Xr bubger 1
+update completes
+.Po
+.Pa UIDNEXT
+is always the last file written
+.Pc
+and copies the listed files
+to the remote host:
+.Bd -literal -offset indent
+exec rsync -a --files-from=$HOME/list/FILES ~/list remote:list
+.Ed
+.
+.Pp
+I haven't tagged any
+.Xr bubger 1
+releases yet
+because it hasn't gotten
+a huge amount of testing,
+and I'm not sure anyone but me
+would even want to use it.
+But I'm happy
+with how it's working right now,
+so I might tag 1.0 soon
+just for fun.
+.
+.Sh SEE ALSO
+.Bl -item -compact
+.It
+.Lk https://causal.agency/list/
+.It
+.Lk https://git.causal.agency/bubger/about
+.It
+.Lk https://git.causal.agency/catsit/about
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Sh BUGS
+Almost every time
+I try to type
+.Dq mailing list
+I instead type
+.Dq mailist list .
diff --git a/www/text.causal.agency/020-c-style.7 b/www/text.causal.agency/020-c-style.7
new file mode 100644
index 00000000..9816dbc3
--- /dev/null
+++ b/www/text.causal.agency/020-c-style.7
@@ -0,0 +1,172 @@
+.Dd March 16, 2021
+.Dt C-STYLE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm C Style
+.Nd a rough description
+.
+.Sh DESCRIPTION
+This is a rough description
+of the style in which I write C,
+since it's uncommon
+but some people seem to like it.
+I don't have any hard rules,
+it just needs to look right.
+.
+.Ss Superficialities
+I use tabs
+and they're set to 4 characters wide
+in my editor.
+I keep my lines shorter than 80 columns,
+which I enforce by
+not resizing my terminal's width.
+I use block indentation only,
+meaning I write long function calls
+like this:
+.Bd -literal -offset indent
+fprintf(
+    imap.w, "%s UID THREAD %s UTF-8 %s\er\en",
+    Atoms[thread], algo, search
+);
+.Ed
+.Pp
+Anything that can be sorted
+should be sorted,
+with trailing commas where possible.
+This and block indentation
+make for simpler diffs.
+.Pp
+I either write single-line ifs
+or always use braces.
+I put parentheses
+around ternary expressions.
+I use camelCase
+for functions and variables,
+and PascalCase for types and constants.
+When an acronym appears
+in an identifier,
+it's in either all lower case
+or all upper case.
+The despicable SCREAMING_SNAKE_CASE
+is reserved for macros.
+I don't set globals or statics to zero
+since that is already the default.
+I don't compare against zero or NULL
+unnecessarily.
+.
+.Ss \&No side-effects in control flow
+I never write a function call
+with side-effects
+inside the condition of an if statement.
+I find this makes following the
+.Dq happy path
+through functions
+much easier.
+I write things like this:
+.Bd -literal -offset indent
+pidFile = open(pidPath, O_WRONLY | O_CREAT | O_CLOEXEC, 0600);
+if (pidFile < 0) err(EX_CANTCREAT, "%s", pidPath);
+
+error = flock(pidFile, LOCK_EX | LOCK_NB);
+if (error && errno != EWOULDBLOCK) err(EX_IOERR, "%s", pidPath);
+if (error) errx(EX_CANTCREAT, "%s: file is locked", pidPath);
+.Ed
+.Pp
+I do write side-effects
+inside for and while statement heads,
+since that's generally expected.
+For some reason
+I like to write the constant first
+if I'm comparing the result of an assignment
+with a side-effect.
+.Bd -literal -offset indent
+for (ssize_t len; 0 <= (len = getline(&buf, &cap, file)); ++line)
+.Ed
+.
+.Ss Paragraphs
+I leave blank lines
+between logical chunks of
+.Dq things happening .
+This is usually between side-effects
+with their related error handling,
+or between groups of closely related side-effects.
+I try to keep variable declarations
+glued to the top of the bit of code
+they're used in.
+.
+.Ss Leading break
+I've mentioned this previously.
+I write my switch statement breaks
+before each case label.
+Doing this aligns nicely,
+and being in the habit
+means I always avoid
+accidental fallthrough.
+.Bd -literal -offset indent
+switch (opt) {
+    break; case 'a': append = 1;
+    break; case 'd': delay = strtol(optarg, NULL, 10);
+    break; case 'f': watch(kq, optarg);
+    break; case 'i': init = 1;
+    break; default: return EX_USAGE;
+}
+.Ed
+.
+.Ss Function type definitions
+Function types are always typedef'd,
+and it's the function type itself
+that is defined,
+not a function pointer type!
+I put the typedef above any functions
+that are supposed to be of that type
+so it's clear what the pattern is.
+.Bd -literal -offset indent
+typedef void Action(struct Service *service);
+Action *fn = NULL;
+.Ed
+.
+.Ss Constants
+I prefer enums over #defines
+for integer constants,
+and static const strings over #defines
+unless I want to do concatenation.
+.Bd -literal -offset indent
+enum { Cap = 1024 };
+.Ed
+.Pp
+I avoid the preprocessor
+wherever possible,
+with the notable exception of X macros,
+which I've talked about previously.
+Doing things in the actual language
+makes for easier debugging.
+.
+.Ss Organization
+I usually use only one header file
+in each project.
+The dependency is easy to declare
+and the complete rebuild
+when the header changes
+isn't a problem for small projects.
+Unless it's a single-file program,
+I name the file which contains main
+something generic,
+since the name of the project
+isn't relevant to its function.
+I name functions like
+.Ar nounVerb ,
+and all the functions for a 
+.Ar noun
+are defined in
+.Pa noun.c .
+Not really to do with C,
+but I always put a FILES section
+in my README pages
+to briefly describe
+the layout of the code
+for anyone looking to
+read or make changes to it.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/021-time-machine.7 b/www/text.causal.agency/021-time-machine.7
new file mode 100644
index 00000000..93d35c1e
--- /dev/null
+++ b/www/text.causal.agency/021-time-machine.7
@@ -0,0 +1,144 @@
+.Dd April 25, 2021
+.Dt TIME-MACHINE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Time Machine
+.Nd an awful one
+.
+.Sh DESCRIPTION
+If, like me,
+you have a Raspberry Pi 3 at home
+that you've just upgraded to
+.Fx 13.0
+which has a hard drive
+from an old laptop
+attached to it by USB adapter
+with ZFS on it
+and you want to
+use that as a Time Machine
+backup destination
+over SMB using
+.Xr samba 8 ,
+despite
+.Xr samba 8
+being awful software
+and using ZFS on a system
+with only 1 GB of RAM
+being a terrible idea,
+this is how to do it.
+.
+.Pp
+In
+.Pa /usr/local/etc/smb4.conf :
+.Bd -literal -offset indent
+[global]
+vfs objects = zfsacl catia fruit streams_xattr
+fruit:metadata = stream
+fruit:model = Macmini
+
+[TimeMachine]
+read only = no
+path = /media/zhdd/backup/TimeMachine
+fruit:time machine = yes
+fruit:time machine max size = 250G
+.Ed
+.
+.Pp
+The important thing here is
+.Sy zfsacl
+in the vfs objects list.
+Most pages will tell you about the others,
+but without
+.Sy zfsacl
+Time Machine will just fail to
+create the backup
+and not provide any useful error.
+I'm not actually sure if the
+.Sy fruit:metadata
+setting is required,
+but a bunch of pages recommend it.
+The
+.Sy fruit:model
+just makes it look nice in Finder.
+The rest creates an SMB share called
+.Dq TimeMachine
+that macOS will be willing to use.
+You can limit the size of the share that
+.Xr samba 8
+reports so that Time Machine
+doesn't fill up the whole drive.
+.
+.Pp
+The other important thing to do
+is to create some swap space.
+When I first tried backing up
+to this share,
+it stopped after a while
+because
+.Xr smbd 8
+got killed
+when there was nowhere to swap pages to.
+A wiki page told me to
+create swap on ZFS like this:
+.Bd -literal -offset indent
+zfs create -V 2G \e
+	-o org.freebsd:swap=on \e
+	-o checksum=off \e
+	-o compression=off \e
+	-o dedup=off \e
+	-o sync=disabled \e
+	-o primarycache=none \e
+	zhdd/swap
+swapon /dev/zvol/zhdd/swap
+.Ed
+.
+.Pp
+To be fair to
+.Xr samba 8 ,
+most of the memory
+is being used by the ZFS ARC
+.Po
+which you can see in
+.Xr top 1
+.Pc ,
+but
+.Xr smbd 8
+still seems to be using
+far more memory than is reasonable.
+It's interesting seeing processes
+with 0 RES in
+.Xr htop 1
+because they're all being swapped out
+while the ARC takes half the available RAM.
+And having to wait for my shell
+to be paged back in when I quit
+.Xr htop 1 .
+.
+.Pp
+Anyway,
+as expected this whole thing
+is terribly slow.
+On my initial backup,
+I'm currently at 26.49 GB
+of 104.22 GB
+with an estimate of 8 hours remaining.
+Normally transfer time estimates
+are wildly inaccurate,
+but I think in this case it's right.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Sh BUGS
+.Fx
+doesn't seem to want to mount
+the ZFS volumes on the hard-drive-over-USB
+automatically at boot.
+I have to
+.Xr zpool-import 8
+the drive manually each time.
+I don't know if there's a workaround for this,
+but I don't have anything essential
+to the system on the drive,
+and it doesn't need to reboot often.
diff --git a/www/text.causal.agency/022-swans-are-dead.7 b/www/text.causal.agency/022-swans-are-dead.7
new file mode 100644
index 00000000..8664e886
--- /dev/null
+++ b/www/text.causal.agency/022-swans-are-dead.7
@@ -0,0 +1,164 @@
+.Dd May  5, 2021
+.Dt SWANS-ARE-DEAD 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Swans Are Dead
+.Nd album by Swans
+.
+.Sh DESCRIPTION
+Swans Are Dead
+is the best Swans album.
+Among my favourites are
+Soundtracks for the Blind,
+To Be Kind
+and Love of Life,
+but Swans Are Dead
+is the one I come back to
+most consistently.
+I'm always in the mood
+to listen to these tunes.
+.
+.Pp
+It's interesting to me
+that I enjoy it so much,
+I think because I had the expectation
+that live albums
+are not of the same quality
+as studio albums,
+but that's just completely untrue
+in the case of Swans.
+The performances are excellent
+and the recording is
+for the most part perfect.
+The album feels live,
+without any distracting deficiencies
+of live recording
+that would take you out
+of just enjoying the music.
+.
+.Bl -ohang
+.It Dq Feel Happiness
+This track feels kind of special
+since it's the only song on the album
+that was never released
+as part of another project.
+I absolutely love this format of song.
+It's like 10 minutes of build
+before any lyrics happen,
+which you only get after
+the wave of the first part
+of the song collapses.
+It bookends the first disc nicely with
+.Dq Blood Promise,
+I think,
+which is sort of the reverse.
+.
+.It Dq Blood On Yr Hands
+This is such a great start
+to the Jarboe-focused
+section of the black disc.
+A cappella apart from the hum
+of the equipment on stage,
+I love this vocal performance.
+I sing this song,
+terribly,
+in the shower.
+The lack of instrumental
+seems to make it stick in my mind even more.
+.
+.It Dq I Crawled
+This is another great vocal performance
+by Jarboe.
+It's so much more dynamic and intense
+than the version of this song
+released much earlier on Young God
+with Gira's vocals.
+I remember seeing a bad comment
+somewhere online
+from someone who couldn't stand
+any Swans song Jarboe sang on.
+They must have never heard
+this version of
+.Dq I Crawled.
+Incredible.
+.
+.It Dq Blood Promise
+My favourite track on
+Swans Are Dead,
+by far.
+I had actually never heard of
+.Dq The Whiffenpoof Song
+until I looked up
+the recording they use
+to introduce this song
+and indicate it's the last of the show.
+Anyway,
+this track highlights
+what makes Swans live albums
+so interesting.
+This performance of the song
+has evolved so much
+from the studio recording on
+The Great Annihilator.
+They share the same lyrics,
+but the earlier version is only 4:15,
+to the live version's fifteen and a half minutes!
+And it sucks me in the whole time.
+As the song winds down
+you can hear an audience member yell,
+.Dq Don't stop!
+and I agree.
+.
+.It Dq The Sound
+One of my all-time favourite songs.
+It's the one that got me to listen to
+Soundtracks for the Blind,
+and might've gotten me into Swans altogether.
+I don't quite remember
+what order I started listening to things in.
+This version of it is great.
+I don't think I could choose
+between this and the studio recording.
+There are just
+two ways to enjoy it.
+I love how frantic the guitars get
+at the height of this track.
+.
+.It Dq I See Them All Lined Up
+This version of the song
+is way more harsh
+than the version on Soundtracks.
+It loses some contrast
+between the verses
+and the explosions of sound
+punctuating them,
+it just hits hard
+the whole time.
+.
+.It Dq Yum Yab
+An absolute banger.
+The drums sound so good on this
+and they really get me moving.
+The whole thing is delightfully nasty and fun.
+.El
+.
+.Pp
+Everything else on the album
+is good too,
+of course,
+I just don't have as much to say.
+There's almost two and a half hours of music
+on this thing!
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+I want to try writing
+about different kinds of things here,
+and this is my first attempt
+at doing so.
+There's more music
+I want to write about,
+and maybe some other
+non-computer topics.
diff --git a/www/text.causal.agency/023-sparse-checkout.7 b/www/text.causal.agency/023-sparse-checkout.7
new file mode 100644
index 00000000..925bc043
--- /dev/null
+++ b/www/text.causal.agency/023-sparse-checkout.7
@@ -0,0 +1,144 @@
+.Dd June  9, 2021
+.Dt SPARSE-CHECKOUT 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Sparse Checkout
+.Nd a cool git feature
+.
+.Sh DESCRIPTION
+I was going to write a post about
+.Xr git-subtree 1
+(and I still plan to!)
+but while talking about it
+with a friend
+I came across another command:
+.Xr git-sparse-checkout 1 .
+I got pretty excited because
+I already had a use case for it.
+.
+.Pp
+.Xr git-sparse-checkout 1
+does pretty much what it sounds like.
+It lets you only have
+a subset of files in the repository actually
+.Dq checked out .
+This is really useful
+for huge respositories
+where you are only interested in
+some part of it.
+Any operation touching the working tree
+is much faster because
+it can skip all the files you don't care about.
+.
+.Pp
+My use case is with the
+.Fx
+.Xr ports 7
+tree,
+which recently moved to git
+and contains almost 14 thousand files.
+Working with the whole repository
+was super painful.
+.Xr git-status 1 ,
+which I run as a habit
+when my shell is idle,
+would take dozens of seconds
+to check the whole working tree
+and report back.
+(I didn't get any real time measurements
+before enabling
+.Xr git-sparse-checkout 1 ,
+and I'm not about to disable it now,
+since it'd have to check out
+all those files again.)
+I'm only actually working on
+a small handful of ports,
+so all that work is wasted.
+Time to turn on sparse checkout:
+.Bd -literal -offset indent
+git sparse-checkout init --cone
+.Ed
+.
+.Pp
+The
+.Fl \-cone
+option here
+(which I keep reading as
+.Dq clone
+because it's git)
+restricts the kinds of patterns
+you can use to select files to check out,
+but makes the calculation more efficient.
+Basically it means you can only select
+paths along with everything below them,
+which I think is pretty much
+always what you want anyway.
+Enabling sparse checkout
+can take quite a while
+because it has to do a lot of un-checking-out.
+I should mention
+that you can pass
+.Fl \-sparse
+to
+.Xr git-clone 1
+to avoid ever checking out
+the whole tree.
+.
+.Pp
+The default selection when you run
+.Cm init
+is to check out all the files
+at the root of the repository,
+but none of the subdirectories.
+For
+.Xr ports 7 ,
+I also want to check out
+the shared scripts and Makefiles:
+.Bd -literal -offset indent
+git sparse-checkout add Keywords Mk Templates Tools
+.Ed
+.
+.Pp
+And then I can selectively check out
+just the ports I'm working on:
+.Bd -literal -offset indent
+git sparse-checkout add irc/catgirl irc/pounce
+.Ed
+.
+.Pp
+After enabling sparse checkout,
+.Xr git-status 1
+takes what I'd call
+a normal amount of time.
+I also did this on
+a couple-weeks-out-of-date copy of the
+.Xr ports 7
+tree,
+and when I ran
+.Xr git-pull 1
+it was also really quick,
+because it didn't have to bother
+updating all those files
+I'm not interested in.
+It still downloads all the git objects,
+of course,
+and you can just add any new paths you need
+to the sparse checkout list.
+My disk usage also went down
+by about a gigabyte.
+.
+.Pp
+I'm super pleased to discover this part of git,
+because it makes working with huge
+and/or monorepo-style repositories
+so much more feasible!
+You can see how I came across it,
+since
+.Xr git-subtree 1
+is also a useful tool for monorepos.
+Stay tuned for that post,
+I guess :)
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/024-seprintf.7 b/www/text.causal.agency/024-seprintf.7
new file mode 100644
index 00000000..d1af2e1a
--- /dev/null
+++ b/www/text.causal.agency/024-seprintf.7
@@ -0,0 +1,137 @@
+.Dd June 12, 2021
+.Dt SEPRINTF 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm seprintf
+.Nd an snprintf alternative
+.
+.Sh SYNOPSIS
+.Ft "char *"
+.Fn seprintf "char *ptr" "char *end" "const char *fmt" "..."
+.
+.Sh DESCRIPTION
+While discussing string building in C recently,
+mcf pointed out
+.Xr seprint 2
+from Plan 9,
+and it kind of blew my mind.
+I had implemented my own function in
+.Xr catgirl 1
+for building up strings using
+.Xr snprintf 3
+and a struct containing
+pointer, length and capacity,
+but it felt out of place.
+.Fn seprintf
+(I add the
+.Dq f ,
+Plan 9 doesn't)
+is a much simpler
+and more
+.Dq C-like
+interface with really nice usage patterns.
+.
+.Pp
+The obvious difference from
+.Xr snprintf 3
+is that
+.Fn seprintf
+takes an
+.Fa end
+pointer
+rather than a size.
+This means you need only calculate it
+once for each buffer,
+rather than subtracting
+the running length from the buffer size.
+.Fn seprintf Ap s
+return value is a pointer
+to the terminating null
+of the string it wrote,
+so you can pass that back in
+to continue appending
+to the same buffer.
+.
+.Pp
+I'm not sure of the exact behaviour on Plan 9,
+but my implementation indicates truncation occurred
+by returning the
+.Fa end
+pointer.
+That makes it both easy to check,
+and perfectly fine to keep calling
+.Fn seprintf
+anyway.
+It just won't write anything if
+.Fa ptr
+==
+.Fa end .
+.
+.Pp
+In the case of formatting failure
+(which should be prevented by
+compile-time format string checking,
+but should still be considered),
+.Fn seprintf
+returns
+.Dv NULL .
+I'm again not sure if this matches Plan 9.
+I like this a lot better than
+.Xr snprintf 3
+returning -1,
+because an unchecked
+.Dv NULL
+is likely to quickly cause a crash,
+while blindly adding
+.Xr snprintf 3 Ap s
+return value
+to your running length
+is a non-obvious logic error.
+.
+.Sh EXAMPLES
+Here's an example of what some code using
+.Fn seprintf
+might look like:
+.Bd -literal -offset indent
+char buf[4096];
+char *ptr = buf, *end = &buf[sizeof(buf)];
+ptr = seprintf(ptr, end, "argv: ");
+for (int i = 1; i < argc; ++i) {
+	ptr = seprintf(
+		ptr, end, "%s%s",
+		(i > 1 ? ", " : ""), argv[i]
+	);
+}
+if (ptr == end) errx(1, "truncation occurred :(");
+.Ed
+.
+.Pp
+And here is the very short implementation of it against
+.Xr vsnprintf 3
+which I copy into my project header files:
+.Bd -literal -offset indent
+#include <stdarg.h>
+#include <stdio.h>
+static inline char *
+seprintf(char *ptr, char *end, const char *fmt, ...)
+	__attribute__((format(printf, 3, 4)));
+static inline char *
+seprintf(char *ptr, char *end, const char *fmt, ...) {
+	va_list ap;
+	va_start(ap, fmt);
+	int n = vsnprintf(ptr, end - ptr, fmt, ap);
+	va_end(ap);
+	if (n < 0) return NULL;
+	if (n > end - ptr) return end;
+	return ptr + n;
+}
+.Ed
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Another short one before
+.Xr git-subtree 1 .
+I just think this function
+is really neat.
diff --git a/www/text.causal.agency/025-v6-pwd.7 b/www/text.causal.agency/025-v6-pwd.7
new file mode 100644
index 00000000..90bfd6ac
--- /dev/null
+++ b/www/text.causal.agency/025-v6-pwd.7
@@ -0,0 +1,330 @@
+.Dd September  1, 2021
+.Dt V6-PWD 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm V6 pwd
+.Nd deciphering old code
+.
+.Sh DESCRIPTION
+We were talking about
+.Xr wall 1
+on IRC
+and how long it had been annoying users.
+My manual page says
+.Xr wall 1
+appeared in
+.At v6 ,
+which means that
+.Xr wall 1
+has been annoying users for 46 years!
+.
+.Pp
+The Wikipedia page links to the source for
+.At v6 ,
+so I was curious to see how the very first
+.Xr wall 1
+was implemented.
+It's not that surprising,
+except that it is hardcoded
+to handle only 50 logins,
+and it forks to write to each tty,
+waiting one second between each.
+I think the forking must be to avoid
+any of the terminals being opened
+from becoming the controlling terminal
+of the original
+.Xr wall 1
+process.
+.
+.Pp
+Then I started looking
+at some of the other source files
+and found the implementation of
+.Xr pwd 1 ,
+which was surprising.
+There's no
+.Xr getcwd 3
+function
+(the earlier form of which,
+.Xr getwd 3 ,
+appeared in
+.Bx 4.0 ) ,
+so
+.Xr pwd 1
+has to figure out
+the path to the working directory itself.
+It took me a while to figure out how it works.
+.
+.Pp
+To make it easy to talk about,
+I'm just going to include the whole thing here:
+.Bd -literal
+char dot[] ".";
+char dotdot[] "..";
+char root[] "/";
+char name[512];
+int file, off -1;
+struct statb {int devn, inum, i[18];}x;
+struct entry { int jnum; char name[16];}y;
+
+main() {
+	int n;
+
+loop0:
+	stat(dot, &x);
+	if((file = open(dotdot,0)) < 0) prname();
+loop1:
+	if((n = read(file,&y,16)) < 16) prname();
+	if(y.jnum != x.inum)goto loop1;
+	close(file);
+	if(y.jnum == 1) ckroot();
+	cat();
+	chdir(dotdot);
+	goto loop0;
+}
+ckroot() {
+	int i, n;
+
+	if((n = stat(y.name,&x)) < 0) prname();
+	i = x.devn;
+	if((n = chdir(root)) < 0) prname();
+	if((file = open(root,0)) < 0) prname();
+loop:
+	if((n = read(file,&y,16)) < 16) prname();
+	if(y.jnum == 0) goto loop;
+	if((n = stat(y.name,&x)) < 0) prname();
+	if(x.devn != i) goto loop;
+	x.i[0] =& 060000;
+	if(x.i[0] != 040000) goto loop;
+	if(y.name[0]=='.')if(((y.name[1]=='.') && (y.name[2]==0)) ||
+				(y.name[1] == 0)) goto pr;
+	cat();
+pr:
+	write(1,root,1);
+	prname();
+}
+prname() {
+	if(off<0)off=0;
+	name[off] = '\en';
+	write(1,name,off+1);
+	exit();
+}
+cat() {
+	int i, j;
+
+	i = -1;
+	while(y.name[++i] != 0);
+	if((off+i+2) > 511) prname();
+	for(j=off+1; j>=0; --j) name[j+i+1] = name[j];
+	off=i+off+1;
+	name[i] = root[0];
+	for(--i; i>=0; --i) name[i] = y.name[i];
+}
+.Ed
+.
+.Pp
+First, some syntax trivia:
+it seems you don't need
+.Sy =
+to give globals values.
+I guess that makes sense.
+I also noticed that
+it avoids giving
+.Va inum
+and
+.Va jnum
+the same name.
+I think that's because in old C,
+struct field names all shared the same namespace.
+The last difference I noticed
+is the operator
+.Sy =&
+rather than
+.Sy &= .
+Honestly I think the former makes more sense,
+but I can see that the one we have now
+is less ambiguous.
+.
+.Pp
+To get
+.Fn prname
+and
+.Fn cat
+out of the way,
+it's building up a path from the bottom.
+At first I thought it must be
+starting at the end of its buffer
+and moving back as it adds components,
+but no,
+it moves the entire path-so-far over
+every time it adds a new component
+onto the front.
+.Fn cat
+is just a bunch of manual string copying.
+It also gives up
+if the new component
+would make the path longer than 511 characters.
+Fair enough.
+.
+.Pp
+So how does it build up the path?
+The loop in
+.Fn main
+first calls
+.Xr stat 2
+on the current directory
+.Pa \&.
+in order to get its inode number.
+I love that
+.Vt struct statb
+is just declared at the top of this file.
+Clearly this code predates the C preprocessor.
+.
+.Pp
+It then opens the parent directory
+.Pa ..
+and reads directory entries from it.
+The inner loop is looking for
+a directory entry with the same inode number
+as the current directory,
+to figure out what the current directory is called.
+Curiously,
+it reads 16-byte directory entries,
+despite declaring a larger struct.
+The preprocessor can't be invented soon enough.
+.
+.Pp
+Once it finds the matching directory entry,
+it adds the name of the entry
+onto the front of the path,
+changes directory to
+.Pa ..
+and starts over.
+It stops when the current directory
+has an inode number of 1,
+which must be the root of a file system,
+but then it does something else.
+It took me a while to decipher what
+.Fn ckroot
+is doing.
+.
+.Pp
+The loop in
+.Fn main
+stops when it gets to the root
+of a file system,
+but that's not necessarily
+.Pa / .
+I think what
+.Fn ckroot
+is doing is trying to figure out
+where that file system is mounted.
+It starts by checking the device number
+that the current directory is on.
+Or really it calls
+.Xr stat 2
+on the name of the directory entry that
+.Fn main
+just found,
+which I think must be
+.Pa \&.
+at this point anyway since it's at a root.
+.
+.Pp
+Anyway,
+it then changes directory to and opens
+.Pa /
+and starts reading directory entries from that,
+calling
+.Xr stat 2
+on each of them
+and checking for a matching device number.
+I think this implies that file systems
+can only be mounted in
+.Pa /
+and not at any lower level,
+at least not if you want
+.Xr pwd 1
+to understand it.
+I'm not sure what the check for
+an inode number of 0 is skipping over
+in this loop.
+Some kind of special entry in
+.Pa /
+perhaps.
+.
+.Pp
+Once it finds an entry
+with a matching device number,
+it checks the flags
+to make sure the entry is a directory.
+It does so with hardcoded constants,
+but it seems they haven't changed
+in all these years.
+According to
+.Xr stat 2 ,
+040000 is
+.Dv S_IFDIR .
+The number of file types
+clearly has grown since then though,
+since
+.Dv S_IFMT
+is now 0170000 rather than 060000.
+.
+.Pp
+I think the reason it checks
+that the entry is a directory
+is because if it actually is
+on the root file system already,
+then any regular file
+would have a matching device number.
+If the entry is indeed a directory,
+it then checks if the entry is
+.Pa \&.
+or
+.Pa \&.. ,
+which indicates that it really is already at
+.Pa / .
+If it's not,
+it adds the mount point that it found
+to the front of the path.
+.
+.Pp
+Finally,
+it prints
+.Pa /
+followed by the path it built up.
+If it failed at any point before that,
+it would print the path it had built so far
+with no leading
+.Pa / .
+Better than nothing!
+.
+.Pp
+So that's how I think
+.Xr pwd 1
+works in
+.At v6 .
+It was a fun puzzle to work through,
+and it was interesting to see
+the assumptions it makes.
+How simple things were back then...
+Actually I find it really cool
+that code from 1975
+can still be read and understood
+using knowledge of modern C and UNIX-likes.
+.
+.Sh SEE ALSO
+.Lk https://minnie.tuhs.org/cgi-bin/utree.pl?file=V6
+.Pp
+.Pa pwd.c
+appears in
+.Pa V6/usr/source/s2 .
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+I regret saying in two previous posts
+what I planned to write next,
+because this is still not that.
diff --git a/www/text.causal.agency/026-git-comment.7 b/www/text.causal.agency/026-git-comment.7
new file mode 100644
index 00000000..fefb497e
--- /dev/null
+++ b/www/text.causal.agency/026-git-comment.7
@@ -0,0 +1,190 @@
+.Dd September 10, 2021
+.Dt GIT-COMMENT 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm git-comment
+.Nd add comments from commit messages
+.
+.Sh SYNOPSIS
+.Nm git comment
+.Op Fl \-all
+.Op Fl \-comment-start Ar string
+.Op Fl \-comment-lead Ar string
+.Op Fl \-comment-end Ar string
+.Op Fl \-min-group Ar lines
+.Op Fl \-min-repeat Ar lines
+.Op Fl \-no-repeat
+.Op Fl \-pretty Ar format
+.Op Ar options ...
+.Op Fl \-
+.Ar file
+.
+.Sh DESCRIPTION
+The
+.Nm
+command
+adds comments to a file
+showing the commit messages
+which last modified
+each group of lines.
+By default only commit messages with bodies
+and which modified groups of at least 2 lines
+are added.
+Each comment contains
+the abbreviated commit hash
+and the commit summary,
+followed by the commit body.
+.
+.Pp
+.Nm
+accepts all the options of
+.Xr git-blame 1
+in addition to the following:
+.Bl -tag -width Ds
+.It Fl \-all
+Include all commit messages.
+The default is to include
+only commit messages with bodies
+(lines after the summary).
+.
+.It Fl \-comment-start Ar string
+Start comments with
+.Ar string .
+The default is the value of
+.Cm comment.start
+or
+.Ql /* .
+.
+.It Fl \-comment-lead Ar string
+Continue comments with the leading
+.Ar string .
+The default is the value of
+.Cm comment.lead
+or
+.Ql " *" .
+.
+.It Fl \-comment-end Ar string
+End comments with
+.Ar string .
+The default is the value of
+.Cm comment.end
+or
+.Ql " */" .
+.
+.It Fl \-min-group Ar lines
+Add comments only for groups of at least
+.Ar lines .
+The default is 2 lines.
+.
+.It Fl \-min-repeat Ar lines
+Avoid repeating a comment
+if it occurred in the last
+.Ar lines .
+The default is 30 lines.
+.
+.It Fl \-no-repeat
+Avoid repeating comments entirely.
+.
+.It Fl \-pretty Ar format
+Set the pretty-print format
+to use for commit messages.
+The default is the value of
+.Cm comment.pretty
+or
+.Ql format:%h\ %s%n%n%-b .
+See
+.Xr git-show 1 .
+.El
+.
+.Sh EXAMPLES
+For files with
+.Ql #
+comments:
+.Bd -literal -offset indent
+git config comment.start '#'
+git config comment.lead '#'
+git config comment.end ''
+.Ed
+.
+.Pp
+Add as many comments as possible:
+.Bd -literal -offset indent
+git comment --all --min-group 1 --min-repeat 1
+.Ed
+.
+.Pp
+Some examples of output from
+.Xr catgirl 1 :
+.Bd -literal
+/* 347e2b4 Don't apply uiThreshold to Network and Debug
+ *
+ * Messages don't really need to be hidden from <network> and I think
+ * it could be confusing. Debug messages are all Cold so everything
+ * would be hidden, and I want to keep them that way so that <debug>
+ * doesn't clutter the status line needlessly.
+ */
+if (id == Network || id == Debug) {
+	window->thresh = Cold;
+} else {
+	window->thresh = uiThreshold;
+}
+
+/* b4c26a2 Measure timestamp width using ncurses
+ *
+ * This allows for non-ASCII characters in timestamps, and simplifies
+ * things by including the trailing space in the width.
+ */
+int y;
+char buf[TimeCap];
+struct tm *time = localtime(&(time_t) { -22100400 });
+size_t len = strftime(buf, sizeof(buf), uiTime.format, time);
+if (!len) errx(EX_CONFIG, "invalid timestamp format: %s", uiTime.format);
+waddstr(main, buf);
+waddch(main, ' ');
+getyx(main, y, uiTime.width);
+(void)y;
+
+/* 43b1dba Restore toggling ignore with M--
+ *
+ * So that pressing M-- repeatedly maintains the previous behavior.
+ */
+if (n < 0 && window->thresh == Ice) {
+	window->thresh = Cold;
+} else {
+	window->thresh += n;
+}
+
+/* 1891c77 Preserve colon from previous tab-complete
+ *
+ * This fixes the case when pinging multiple nicks and one of them needs to
+ * be cycled through.
+ */
+bool colon = (tab.len >= 2 && buf[tab.pos + tab.len - 2] == L':');
+.Ed
+.
+.Sh SEE ALSO
+.Lk https://git.causal.agency/src/tree/bin/git-comment.pl
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Pp
+In case it's unclear,
+this is a
+.Xr git 1
+subcommand I wrote.
+Did you know you can add new
+.Xr git 1
+subcommands just by
+adding executables named
+.Pa git-*
+to somewhere in
+.Ev PATH ?
+.
+.Pp
+This is also,
+I think,
+my third Perl script ever.
+It's an interestingly shaped language.
+Quite neat.
diff --git a/www/text.causal.agency/027-openbsd-linode.7 b/www/text.causal.agency/027-openbsd-linode.7
new file mode 100644
index 00000000..9f40de42
--- /dev/null
+++ b/www/text.causal.agency/027-openbsd-linode.7
@@ -0,0 +1,202 @@
+.Dd September 26, 2021
+.Dt OPENBSD-LINODE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Installing OpenBSD on Linode
+.Nd a guide
+.
+.Sh DESCRIPTION
+I've been thinking for a while
+about moving my servers to Linode,
+and also about moving them to
+.Ox .
+I actually originally got into
+.Fx
+(and from there,
+.Ox )
+only because DigitalOcean
+started offering it as a
+.Dq droplet
+image.
+I've been running those servers fine for years,
+but now I prefer to run
+.Ox ,
+and some recent DigitalOcean outages
+had me thinking about it more,
+so I'm giving it a shot.
+.
+.Pp
+As an aside,
+running
+.Ox
+on DigitalOcean
+is not really a good option.
+It seems more awkward to install your own OS there,
+and if you do,
+I've heard that IPv6 won't work
+because they don't know how to run SLAAC.
+Also,
+now that I've used
+the Linode control panel and LISH a bit,
+DigitalOcean kind of feels like a toy
+in comparison.
+.
+.Pp
+Here's what I did to install
+.Ox
+on Linode:
+.Bl -enum
+.It
+Create a Linode with the
+.Dq Choose a Distribution
+box blank.
+.
+.It
+Under the Storage tab,
+create a disk called
+.Dq miniroot
+of type raw
+with size 8 MB.
+This will hold the install image.
+.
+.It
+Create another disk called
+.Dq root
+of type raw
+using the remaining available storage.
+.
+.It
+Boot the Linode in rescue mode
+from the option in the three-dots menu
+next to
+.Dq Power On .
+Attach
+.Dq miniroot
+to
+.Pa /dev/sda .
+.
+.It
+Log into the LISH console
+and obtain the install image:
+.Bd -literal
+curl -O https://cdn.openbsd.org/pub/OpenBSD/6.9/amd64/miniroot69.img
+dd if=miniroot69.img of=/dev/sda
+.Ed
+.Pp
+Power off the Linode.
+.
+.It
+Under the Configurations tab,
+create a configuration called
+.Dq install
+in full virtualization mode.
+Paravirtualization works fine once installed,
+but for some reason the installer
+can't see the root disk
+without full virtualization.
+Under boot settings,
+select direct disk.
+Attach
+.Dq root
+to
+.Pa /dev/sda ,
+.Dq miniroot
+to
+.Pa /dev/sdb
+and set the root device to
+.Pa /dev/sdb .
+.
+.It
+Create a similar configuration called
+.Dq boot
+but using paravirtualiztion
+and without
+.Dq miniroot
+attached.
+Set the root device to
+.Pa /dev/sda .
+.
+.It
+Boot the
+.Dq install
+configuration,
+launch the LISH console
+and switch to Glish.
+It's possible
+to have the installer use serial console,
+but it requires entering commands
+at the boot prompt
+before the timeout,
+and I never managed it.
+If you do manage it,
+run:
+.Bd -literal
+stty com0 9600
+set tty com0
+boot
+.Ed
+.
+.It
+Proceed through the
+.Ox
+installer.
+When asked to
+change the default console to com0,
+answer yes
+so that regular LISH will work.
+Power off the Linode.
+.
+.It
+Boot the
+.Dq boot
+configuration
+and log in to LISH.
+Since the installer configured networking
+in full virtualization,
+rename the file to the paravirtualized interface:
+.Bd -literal
+mv /etc/hostname.em0 /etc/hostname.vio0
+.Ed
+.Pp
+In order to get the right public IPv6 address,
+disable privacy extensions
+by changing the inet6 line of
+.Pa hostname.vio0
+to:
+.Bd -literal
+inet6 autoconf -temporary -soii
+.Ed
+.
+.It
+Bring networking up
+and run
+.Xr syspatch 8
+since
+.Pa rc.firsttime
+couldn't do it:
+.Bd -literal
+sh /etc/netstart
+syspatch
+.Ed
+.
+.It
+Reboot.
+.El
+.
+.Pp
+I guess I'll be slowly moving things over
+to the new servers
+for the next little while.
+With any luck the next post here
+will not say
+.Fx
+in its header!
+.
+.Sh SEE ALSO
+I learned the basic idea
+of how to do this from
+.Lk https://www.subgeniuskitty.com/notes/openbsd_on_linode .
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/028-names.7 b/www/text.causal.agency/028-names.7
new file mode 100644
index 00000000..de47c074
--- /dev/null
+++ b/www/text.causal.agency/028-names.7
@@ -0,0 +1,81 @@
+.Dd October 30, 2021
+.Dt NAMES 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Names
+.Nd three types
+.
+.Sh DESCRIPTION
+There are (at least) three
+different types of names
+a person has.
+.
+.Pp
+First, there are normie names.
+These are names usually made up
+of several words
+each of which is capitalized.
+Most people have one of these,
+but it's possible to have more.
+They're names that might appear on
+various types of Documents.
+A
+.Dq legal name
+(dubious)
+is a normie name,
+but normie names need not be
+.Dq legal
+(dubious).
+I list this category first
+not because it's more important,
+but because it is by far the most boring.
+.
+.Pp
+Next, there are Real Names.
+Most people have at least a few
+and will probably go through
+different ones over time.
+Your Real Names are anything people
+use to refer to you.
+On the internet these are often not capitalized.
+Sometimes that is the only distinction
+between a Real Name
+and a normie name.
+.
+.Pp
+There was a period of time
+when I was playing a lot of TF2
+and not really leaving my apartment.
+I had set my steam name to
+.Dq gluten product
+(yeah, from that dril tweet)
+and I talked in the game's voice chat
+quite a bit.
+Naturally other Gamers in voice chat
+called me
+.Dq gluten
+and at some point I realized
+that over the span of months
+I had been refered to as
+.Dq gluten
+more often than any other name.
+So that was a Real Name of mine.
+People used it and I responded to it.
+.
+.Pp
+Last, there are the True Names.
+The kind of name that knowledge of
+gives one power over a person.
+I don't think any humans
+know their own True Names,
+but I do believe they exist.
+It's possible that other animals
+know theirs.
+It's probably best not to know though, right?
+I think if I knew mine
+I would always worry
+about accidentally revealing it.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/029-topics.7 b/www/text.causal.agency/029-topics.7
new file mode 100644
index 00000000..d071eb67
--- /dev/null
+++ b/www/text.causal.agency/029-topics.7
@@ -0,0 +1,116 @@
+.Dd January  8, 2022
+.Dt TOPICS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Topics
+.Nd a bit of a mess
+.
+.Sh DESCRIPTION
+Shortly after my last post
+I started writing another one
+but I never finished it.
+I don't think I had enough to say,
+or if I did it meant going into
+a whole extra thing.
+I also had a list in my mind
+of other things to write about,
+but inspiration hasn't really struck
+for any of them.
+I'm currently in the mood
+to write something anyway,
+so I'm just going to write a bit
+about the topics I have in mind.
+I may or may not ever
+write more about any of them.
+.
+.Pp
+The post I had started writing
+(twice, actually)
+was about voices.
+I like them a lot
+and I'm fascinated by them.
+The problem is
+I don't actually have much
+to say about it
+without getting into Gender,
+which as I say is a whole extra thing,
+and not something I've written about here before.
+.
+.Pp
+When I started writing here,
+I didn't want to blog about
+personal topics or LGBTQ stuff.
+But more recently
+I want to move away from
+only writing about computers.
+Or maybe away from
+writing about computers entirely.
+There are more interesting things,
+but I don't have experience
+writing about them.
+Yet,
+I should say.
+.
+.Pp
+I'm honestly still not sure
+if writing about gender here
+is at all a good idea.
+But it turns out to feel like
+a bit of a prerequisite
+for other things.
+I find gender perception
+in particular
+to be fascinating.
+It's interesting.
+It's neat.
+And I don't know if I can
+write anything coherent about it.
+.
+.Pp
+Related to that,
+I've been thinking of writing
+about how the pandemic
+has had a strangely positive effect
+on my life.
+Or at least,
+I've made a lot of positive changes
+during it.
+I'm in a better place emotionally now
+than ever before,
+and that obviously runs counter
+to most people's experiences.
+Additionally with that positive outlook
+I want to write about the meaning
+of my domain name.
+I'm proud of it.
+.
+.Pp
+This week the topic of fetish
+has been on my mind.
+That actually feels
+a bit less risky
+to write about than gender.
+And it may honestly be more interesting.
+I don't know.
+There's not enough sex
+on computer blogs,
+or whatever this is.
+Although my main ideas
+are not about sex at all.
+.
+.Pp
+Just this turned out to be
+harder to write than I thought it would be.
+I think I want to populate this space
+with more short posts like the previous one.
+I wrote that while very sleepy
+after 3 AM though,
+and I don't exactly
+want to repeat that regularly.
+We'll see.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Listening to Kate Bush \(em Hounds of Love.
diff --git a/www/text.causal.agency/030-discs.7 b/www/text.causal.agency/030-discs.7
new file mode 100644
index 00000000..df73a750
--- /dev/null
+++ b/www/text.causal.agency/030-discs.7
@@ -0,0 +1,114 @@
+.Dd January  8, 2022
+.Dt DISCS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Desert Island Discs
+.Nd we're doing three in this one
+.
+.Sh DESCRIPTION
+In typical fashion
+I'm going to write about something
+completely different instead.
+Something short and simple.
+I got thinking about this
+after reading a little interview thing
+this week.
+The question is
+which three albums would you want to have
+if you were stranded on a desert island.
+What could you listen to
+for the rest of time?
+It's surprisingly easy
+to take this question very seriously.
+.
+.Pp
+My immediate thought was
+.Em Music for 18 Musicians.
+I've literally said this about it
+in conversation before.
+That's an album
+I'd want to have on a desert island.
+I find it incredibly soothing,
+almost hypnotic.
+I really do feel like
+I could listen to it forever.
+And then maybe I could finally determine
+which of its eleven sections
+is the best.
+.
+.Pp
+My next thought was
+.Em Soundtracks for the Blind .
+We already know I'm a huge SWANS fan.
+Despite what I've written about
+.Em Swans Are Dead ,
+I instead jumped to SFTB.
+I still think that
+.Em Dead
+has better tunes,
+but
+.Em Soundtracks
+is definitely the better cohesive album.
+It has such atmosphere and mood on it.
+Like
+.Em 18 ,
+it's an album that sucks me in.
+Also,
+either SWANS album
+is an economical choice
+in this hypothetical
+since they're each 2 hours and 20 minutes long.
+.
+.Pp
+Choosing a third album is a lot harder.
+There's so much other music I like
+and only one slot left.
+There's no other single album
+that stands out above the rest
+like the previous two,
+for me.
+.Em Wildlife ,
+maybe?
+Or 
+.Em Jane Doe ?
+Perhaps a classic like
+.Em Aeroplane ,
+or a boomer classic like
+.Em The Wall .
+But would I really want to
+listen to any of those
+to the exclusion of everything else?
+They're too mood-dependent.
+.
+.Pp
+Then I realized the perfect choice
+for third album.
+.Em Mouth Moods .
+A mashup album is the perfect wildcard,
+and
+.Em Moods
+is just fun as hell to listen to.
+I get songs from it stuck in my head
+instead of the originals.
+The final track,
+.Em Shit ,
+always gets me moving.
+It's a masterpiece.
+.
+.Bl -enum
+.It
+Steve Reich Ensemble \(em
+.Em Music for 18 Musicians
+.It
+SWANS \(em
+.Em Soundtracks for the Blind
+.It
+Neil Cicierega \(em
+.Em Mouth Moods
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Listening to Steve Reich Ensemble \(em Music for 18 Musicians.
diff --git a/www/text.causal.agency/031-books-2021.7 b/www/text.causal.agency/031-books-2021.7
new file mode 100644
index 00000000..d7b46f17
--- /dev/null
+++ b/www/text.causal.agency/031-books-2021.7
@@ -0,0 +1,127 @@
+.Dd January 12, 2022
+.Dt BOOKS-2021 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Books 2021
+.Nd a review, I guess
+.
+.Sh DESCRIPTION
+In 2021 I read 26 books.
+Finished the 26th right on December 31st.
+It's not a lot but it's more than last year.
+Here are the ones I loved
+(in the order I read them).
+I will avoid spoilers,
+of course.
+.
+.Ss Network Effect by Martha Wells
+I've been reading the
+.Em Murderbot Diaries
+series for a while.
+They're fun stories.
+I liked this full-length novel entry a lot.
+I guess it felt like it had more room
+for the characters to develop.
+This is probably when I started
+asking my friends if they'd read it
+because I wanted to talk about
+Murderbot gender vibes.
+.Pp
+You may like if: you're trans.
+.
+.Ss The Once and Future Witches by Alix E. Harrow
+Um,
+it's about witches!
+One of them has the same name as me.
+Kind of has some similar vibes to
+.%T The Future of Another Timeline ,
+which was my favourite book I read in 2020.
+.Pp
+You may like if: you like women.
+.
+.Ss A Desolation Called Peace by Arkady Martine
+I was so excited for this sequel to
+.%T A Memory Called Empire ,
+another previous favourite
+and something I've been wanting more of.
+I kinda wish there was more fucking in it though honestly.
+.Pp
+You may like if: you like women.
+.
+.Ss Piranesi by Susanna Clarke
+Really something different.
+It turned out to be a different story
+than I expected
+from reading the first few pages.
+.Pp
+You may like if: you like statues, I guess?
+.
+.Ss A Psalm for the Wild-Built by Becky Chambers
+Ok yes I do give 3/3 stars
+to every Becky Chambers book.
+They're so fucking good.
+I'm looking forward to
+more entries in this novella series.
+(Also I'm currently reading
+the fourth
+.Em Wayfarers
+book
+and loving it too!)
+.Pp
+You may like if: your pronouns are they/them <3
+.
+.Sh HONOURABLE MENTIONS
+.Ss Her Body and Other Parties by Carmen Maria Machado
+I really enjoyed the short story
+.Dq Especially Heinous: 272 Views of Law & Order SVU
+in this collection.
+It goes on a bit too long
+but the format is unique.
+You can read that one online,
+actually.
+.
+.Ss The Hobbit by J. R. R. Tolkien
+Yeah I hadn't read this until last year.
+I borrowed it after marathoning
+the extended editions of the
+.%T Lord of the Rings
+trilogy during a heat wave.
+As I said at the time,
+pretty good for something
+written by a man
+like a hundred years ago.
+Kind of hilarious that women
+just don't exist
+in the world of
+.%T The Hobbit .
+.
+.Ss Earthlings by Sayaka Murata
+Pretty fucking wild.
+I'd recommend it,
+but I have to say it
+.Em extremely
+needs a child sexual abuse content warning on it.
+.
+.Ss Six Months, Three Days, Five Others by Charlie Jane Anders
+A surprising number of these short stories
+are actual stories!
+They have beginnings,
+middles
+and ends!
+.
+.Ss The City in the Middle of the Night by Charlie Jane Anders
+It's got some
+.Em Xenogenesis
+series vibes.
+Sophie is a goddamn lesbian idiot though
+and she never even realizes it.
+.
+.Sh SEE ALSO
+.Lk https://git.causal.agency/src/tree/txt/books.txt
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Listening to
+.Em Ear Massage with Latex Gloves 100% Sensitivity 40 minute (No Talking) .
diff --git a/www/text.causal.agency/032-albums-2021.7 b/www/text.causal.agency/032-albums-2021.7
new file mode 100644
index 00000000..72c1d0d2
--- /dev/null
+++ b/www/text.causal.agency/032-albums-2021.7
@@ -0,0 +1,173 @@
+.Dd January 13, 2022
+.Dt ALBUMS-2021 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Albums 2021
+.Nd a review
+.
+.Sh DESCRIPTION
+Every year I create a new playlist
+in iTunes
+(Music dot app, whatever)
+for the albums I listen to that year.
+It's usually embarrassingly short.
+I don't listen to new music
+as much as I'd like,
+and usually only one or two
+are actually from the current year.
+Not that the playlist
+is limited to new (to me) music.
+If I get really into an album
+I've heard before,
+more than before,
+I also add it to the list.
+Anyway,
+this is a review
+of my 2021 albums playlist.
+.
+.Ss Black Country, New Road \(em For the First Time (2021)
+I first heard the single
+.Em Sunglasses
+from someone sharing it on IRC,
+and I loved it,
+so I was looking forward to this album.
+What a let down though.
+The version of
+.Em Sunglasses
+on the album is just plain worse
+than the single version.
+I still got some decent listening
+out of the album,
+but that just sours it for me.
+.Pp
+Favourite track:
+.Em Track X .
+.
+.Ss Black Dresses \(em Forever \&In Your Heart (2021)
+I fucking love Black Dresses.
+.Em Peaceful as Hell
+is one of my all-time favourite albums.
+I'm glad they put out another one
+after it looked like they wouldn't.
+The sounds are just so good.
+Exactly what my ears crave.
+The texture of it
+tickles my brain clit.
+.Pp
+Favourite tracks:
+.Em Waiting42moro ,
+.Em Mistake .
+.
+.Ss Low \(em Drums and Guns (2007)
+I've long loved the song
+.Em Breaker
+and its music video,
+but I only listened to the album
+it's on last year.
+Something I didn't realize,
+I guess because I usually pulled up
+the music video
+without headphones on,
+is how aggressively this album
+uses stereo panning.
+Vocals are generally
+panned hard right throughout,
+with much of the instrumentation
+panned centre or hard left.
+It's bold
+and it really works for me.
+I especially love the vocal harmony on
+.Em Breaker
+all the way on the opposite channel.
+Bring back stereo separation!
+.Pp
+Favourite tracks:
+.Em Breaker ,
+.Em Murderer ,
+.Em Violent Past .
+.
+.Ss The Armed \(em Ultrapop (2021)
+I have to admit
+I didn't actually listen to this one much.
+I listened to the previous album,
+.Em Only Love ,
+a lot in 2020.
+I think this album is good,
+but I'll probably only really get into it
+in some future year.
+.
+.Ss Lingua Ignota \(em Caligula (2019)
+Dear lord,
+why did I wait so long
+to listen to this one.
+I had heard
+.Em "Do You Doubt Me Traitor"
+back when it came out,
+but somehow I didn't realize
+just how much this album
+would be my shit.
+Fucking incredible vocals.
+Lovely sometimes minimal,
+sometimes extreme
+instrumentals
+and exquisite percussion.
+The sound of,
+I believe,
+a lightbulb rolling around on the floor on
+.Em Fragrant
+is such an interesting addition.
+.Pp
+Favourite tracks:
+.Em "Do You Doubt Me Traitor" ,
+.Em "Fragrant Is My Many Flower'd Crown" ,
+.Em "If the Poison Won't Take You My Dogs Will" .
+.
+.Ss Black Dresses \(em LOVE AND AFFECTION FOR STUPID LITTLE BITCHES (2019)
+I wanted even more Black Dresses
+and fortunately there was still more
+I hadn't yet listened to!
+I've already gushed about Black Dresses
+so I'll spare you.
+They're so good though.
+.Pp
+Favourite tracks:
+.Em STATIC ,
+.Em HERTZ ,
+.Em MY HEART BEATS OUT OF TIME .
+.
+.Ss Barenaked Ladies \(em All Their Greatest Hits: Disc One 1991-2001
+What?
+Yeah,
+late last year I decided to revisit BNL.
+My parents listened to them a lot
+when I was growing up,
+and I liked them too.
+The first show I ever went to was the
+.Dq Barenaked for the Holidays
+tour with my parents.
+It turns out
+I still think their '90s stuff
+is pretty darn good!
+Steven Page is really a great singer.
+This is also the first time
+I'm listening to these tunes
+with fancy headphones
+and it sounds great.
+Honestly
+.Em The Old Apartment
+can totally compete
+with the favourites
+I've accumulated more recently.
+\&'90s alt rock was good actually?
+.Pp
+Favourite tracks:
+.Em The Old Apartment ,
+.Em Brian Wilson ,
+.Em What a Good Boy ,
+.Em Too Little Too Late .
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Listening to all my favourite tracks :)
diff --git a/www/text.causal.agency/033-jorts.7 b/www/text.causal.agency/033-jorts.7
new file mode 100644
index 00000000..001f877c
--- /dev/null
+++ b/www/text.causal.agency/033-jorts.7
@@ -0,0 +1,485 @@
+.Dd February  2, 2022
+.Dt JORTS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Introducing Jorts
+.Nd june's ports
+.
+.Sh DESCRIPTION
+Alright so I've gone off the deep end,
+maybe.
+After continual frustration with MacPorts
+culminating in not being able to install
+.Xr nvi 1
+on my work MacBook,
+I have just gone ahead
+and started my own personal ports tree
+for macOS.
+After a couple of weeks,
+I have 32 ports in my tree
+and only two remaining requested ports
+installed from MacPorts.
+.
+.Pp
+I set out with a couple ideas in mind:
+.Bl -bullet
+.It
+This will be my own personal ports tree.
+It only has to work for me.
+Since I'm using it on both
+my personal Intel MacBook Pro
+still running Catalina
+and my work M1 MacBook Pro
+running Monterey,
+it is at least that portable.
+.
+.It
+It's ok to rely on
+system libraries and tools
+provided by macOS.
+I'm not creating a distro,
+so it doesn't need to be totally isolated.
+This lets me skip really annoying things
+like compiler toolchains.
+.
+.It
+Sources get vendored,
+either from release tarballs
+or with
+.Xr git-subtree 1 .
+This allows totally pain-free
+local patching,
+and boy has this paid off.
+I can just do what I need to do
+to get the thing to build how I want
+and commit it in git like anything else.
+.Pp
+It also means that the tree itself
+is entirely self-contained
+and doesn't rely on any external sources
+or network access.
+Honestly with some old and obscure software
+it feels like upstream could disappear at any moment,
+so this gives me peace of mind too.
+.Pp
+Another advantage of vendoring upstream sources
+is that all of the code installed on my system
+(in
+.Pa /usr/local
+anyway)
+is easily inspected,
+much like
+.Pa /usr/src
+on a BSD.
+This can be super useful for debugging
+or just for reference.
+.
+.It
+Produce simple package tarballs.
+They're just the contents of
+.Ev DESTDIR
+after a staged install.
+They get installed for real
+by untarring them in
+.Pa / .
+They can then be uninstalled
+(or upgraded)
+by removing the paths contained
+in the tarball from the system.
+.
+.It
+Track installed packages with symbolic links
+to specific package tarballs.
+Keep old tarballs around for rollbacks.
+This means I can see what's installed
+with plain old
+.Xr ls 1 !
+.Bd -literal
+$ ls */Installed
+\&...
+libretls/Installed          toilet/Installed
+mandoc/Installed            tree/Installed
+
+$ ls -l toilet/Installed
+lrwxr-xr-x  1 root  staff  19 17 Jan 21:45 toilet/Installed -> toilet-0.3~1.tar.gz
+.Ed
+.
+.It
+Use
+.Xr bmake 1 .
+It's scrutable.
+It also knows how to bootstrap itself
+pretty well.
+Since
+.Xr bmake 1
+is itself a port in my tree
+that would require
+.Xr bmake 1
+to build and install,
+I wrote a small
+.Pa Bootstrap
+shell script
+to install
+.Xr bmake 1
+.Dq manually
+then use that
+.Xr bmake 1
+to build and install its own port.
+It also requires a bit of care
+when upgrading the
+.Xr bmake 1
+port since macOS
+rather doesn't like a binary
+deleting itself while it's running.
+.
+.It
+No GNU software.
+I simply refuse to do it.
+To that end,
+prefer configuring/building with
+.Xr cmake 1
+where at all possible.
+I fell into this early on
+since I originally just wanted to install
+.Xr nvi 1
+and
+.Sy lichray/nvi2
+is a better upstream source these days
+that uses
+.Xr cmake 1 .
+.Pp
+With a port and support for
+.Xr cmake 1
+in
+.Pa Port.mk ,
+I can make changes to
+.Pa CMakeLists.txt
+files without issue.
+I can also vendor upstreams
+directly from git
+rather than having to find
+release tarballs with generated
+.Pa configure
+scripts and so on.
+When I need to make changes
+to the build systems of projects using autotools,
+I either have to have autotools installed
+(from outside my tree)
+or painstakingly reflect my edits by hand
+in the generated files,
+both of which suck hard.
+.El
+.
+.Pp
+Ok so that's actually quite a number of ideas.
+But they have come together
+into something surprisingly usable
+surprisingly quickly!
+Like I said,
+this is only intended to be
+my own personal ports tree,
+but I hope that some of these ideas
+are interesting
+and maybe inspire others
+to explore similar approaches.
+.
+.Pp
+But wait,
+I'm not done yet!
+There are some other interesting things
+that I came up with along the way,
+and also some complaints
+about some upstreams,
+but I'll try to keep those to a minimum.
+.
+.Pp
+So it turns out that dependencies are hard.
+Who knew?
+It's easy enough to enforce
+direct dependencies
+at build time
+by just checking for the required
+.Pa Installed
+symlinks.
+It's less straightforward
+to do this recursively,
+which you need if
+you want to be able to say,
+.Do
+Install
+.Xr nvi
+for me!
+.Dc
+and get
+.Xr ncurses 3 ,
+.Xr cmake 1
+and
+.Xr pkgconf 1
+installed first
+if they aren't already.
+.
+.Pp
+Rather than trying to do all that in
+.Xr bmake 1 ,
+I wrote a shell script called
+.Pa Plan ,
+which itself produces a shell script.
+Given a list of ports
+to install or upgrade,
+it recursively gathers their dependencies
+and feeds them to
+.Xr tsort 1 ,
+which is a neat utility
+which topologically sorts a graph.
+In other words,
+it determines the order
+in which the graph of dependencies
+should be installed.
+The
+.Pa Plan
+script produces a list of
+.Xr bmake 1
+commands to make that happen
+on standard output,
+which can be piped to
+.Xr sh 1 .
+So,
+the way to say the above is:
+.Bd -literal -offset ident
+$ ./Plan -j4 nvi | sh -e
+.Ed
+.
+.Pp
+Now,
+what's missing from this approach
+is the ability to automatically
+uninstall no-longer-needed dependencies.
+It's something I've criticized Homebrew for lacking
+and one of the reasons I started using MacPorts,
+so it's somewhat ironic that
+my own system lacks it as well.
+However,
+I don't think it's much of a problem,
+since I'm only packaging
+what I actually want installed
+in the first place.
+On my personal computer,
+I have all 32 of my ports installed,
+and I expect that to continue.
+I can always keep using MacPorts
+to install things I only intend
+to use temporarily.
+.
+.Pp
+Another thing I was slightly concerned about
+from the beginning was disk usage.
+I think the benefits of vendoring sources
+far outweigh the cost in storage,
+but it would be nice to at least minimize that cost.
+Previously,
+I wrote about
+.Xr git-sparse-checkout 1 ,
+which allows you to only have certain paths
+checked out in your git working tree.
+Since port sources aren't always interesting
+and only
+.Em required
+while actually building the port,
+it makes sense to not have them always checked out.
+.
+.Pp
+Rather than manipulate
+.Xr git-sparse-checkout 1
+myself,
+I added support for it
+directly into
+.Pa Port.mk .
+If sparse checkout is enabled,
+building a port will automatically
+add its source tree to the checkout list,
+and cleaning that port will
+remove it from the list.
+At rest,
+only the port system itself
+and the package tarballs
+need to be present on the file system.
+.
+.Pp
+It turns out that upstream
+build system behaviour
+is super inconsistent,
+even among projects using
+the same tools.
+I started collecting a list of checks
+to perform on the output of my port builds
+to make sure they didn't do anything weird.
+They live in
+.Pa Check.sh ,
+which gets run
+when a package tarball is created.
+The current list of checks is:
+.Bl -bullet
+.It
+Check for directories not included by
+.Ev PACKAGE_DIRS .
+In other words,
+make sure the port isn't
+trying to install anything
+outside of
+.Pa /usr/local .
+Sometimes this makes sense,
+though,
+which is what
+.Ev PACKAGE_DIRS
+is for.
+.It
+Check for references to PWD,
+i.e. the build directory.
+This can mean the build
+didn't understand
+.Ev PREFIX
+and
+.Ev DESTDIR
+correctly,
+or that it built with debug info.
+.It
+Check for binaries without manuals.
+If your software installs an executable in
+.Pa bin
+but not a manual page,
+your software is incomplete!
+Sometimes this just means
+I missed an extra documentation install target.
+.It
+Check for dynamic linking to outside objects.
+In other words,
+if something ended up linking to
+a library installed by MacPorts
+rather than the one from
+.Nm jorts
+or macOS.
+.It
+Check for dynamic linking
+to system libraries
+.Nm jorts
+provides instead.
+Similar to the last one,
+if both macOS and
+.Nm jorts
+provide a library,
+check that ports link with the latter.
+.It
+Check for scripts with outside interpreters.
+This is analogous to the linking checks
+but for scripts,
+checking that their shebang lines
+refer to interpreters installed
+by macOS or
+.Nm jorts .
+.El
+.
+.Pp
+A number of my ports
+still fail some of these checks,
+but I have fixed a lot of problems
+the script called out.
+.
+.Pp
+Speaking of problem ports...
+git's build system is truly awful.
+I'm sorry,
+it's just really disappointing.
+On the upside though,
+I did manage to patch it
+to use
+.Xr asciidoctor 1
+directly to generate manual pages
+from asciidoc source,
+rather than generating docbook or whatever
+then converting that.
+One less build dependency!
+I also fixed up curl's
+.Pa CMakeLists.txt
+(which I guess are normally only used on Windows)
+to build and install documentation properly.
+And I got libcaca's Cocoa driver working again!
+Very important to be able to run
+.Xr cacafire 1
+in a Cocoa window.
+.
+.Pp
+Shout out to SDL2,
+which didn't require any patching
+or extra options beyond
+.Ev USE_CMAKE=yes .
+Model upstream.
+.
+.Pp
+Some other odds and ends:
+I like being able to name ports how I want
+(for example,
+.Sy ag )
+and use my own port version convention,
+using
+.Ql +
+to append VCS revisions
+and
+.Ql ~
+to append port revisions.
+I don't think those are likely
+to ever clash with upstream versioning schemes.
+Not that I even need to follow upstream versioning.
+There is no reason the version number of
+.Xr dash 1
+should start with a zero.
+.
+.Pp
+Speaking of versions,
+a big downside of maintaining your own ports tree
+is that you actually need to update it.
+Thankfully,
+once I packaged
+.Xr curl 1
+and
+.Xr jq 1
+(which needs a new release dammit,
+it's been 4 years and the build is broken
+on macOS),
+I could use the Repology API
+to check if I'm behind everyone else.
+Far more reliable than
+trying to automate checking upstreams
+for new versions.
+That lives in the
+.Pa Outdated
+shell script.
+.
+.Pp
+Phew!
+I wrote a lot about this.
+It feels a little self-indulgent,
+but I've had fun working on this
+and want to share.
+If anyone else tries anything similar,
+or is weird enough to give
+.Nm jorts
+a try themselves,
+I'd love to hear about it!
+.
+.Sh SEE ALSO
+.Lk https://git.causal.agency/jorts/
+.Pp
+.Lk https://youtu.be/Sx3ORAO1Y6s
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+Listening to
+.Em Arcade Fire \(em Arcade Fire (EP) ,
+.Em Arcade Fire \(em The Suburbs .
+.Pp
+Typed on a brand new
+Leopold FC660M
+with Cherry MX Red switches.
+Lovely keyboard.
diff --git a/www/text.causal.agency/034-voices.7 b/www/text.causal.agency/034-voices.7
new file mode 100644
index 00000000..4990295d
--- /dev/null
+++ b/www/text.causal.agency/034-voices.7
@@ -0,0 +1,56 @@
+.Dd March  5, 2022
+.Dt VOICES 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Voices
+.Nd more kinds of them
+.
+.Sh DESCRIPTION
+Welcome to the third time
+I started writing this post!
+I think the first time
+was after watching a jan Misali video
+that had clips of audio interviews in it.
+It got me thinking about
+how interesting it was
+to hear someone's voice
+without knowing anything else about them.
+.
+.Pp
+That's pretty much all I managed to write
+the first two times I started this.
+If I get past this next sentence,
+then I can probably finish the post.
+What stopped me was that
+all my thoughts and feelings about voices
+are influenced by being trans
+(and being a fan of other trans people),
+and I thought,
+.Dq I don't write about that here,
+but why don't I?
+I don't have to come out to my blog.
+.
+.Pp
+So really what I have been wanting to say is this:
+every trans woman's voice that I have heard
+has sounded genuinely wonderful to me.
+Especially if you're reading this
+and we've been on a voice call before.
+I know,
+voices are the object of so much self-consciousness,
+but I really wish they didn't have to be.
+Most of us do not sound like cis women
+and to me that is fine.
+Good, actually.
+Trans women sound like trans women.
+As a voice appreciator,
+I am so happy to hear more kinds.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+I've been watching some Vektroid streams lately,
+and I love her voice.
+It was another thing
+reminding me to write this.
diff --git a/www/text.causal.agency/035-addendum-2021.7 b/www/text.causal.agency/035-addendum-2021.7
new file mode 100644
index 00000000..262f2178
--- /dev/null
+++ b/www/text.causal.agency/035-addendum-2021.7
@@ -0,0 +1,111 @@
+.Dd March 18, 2022
+.Dt ADDENDUM-2021 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Addendum 2021
+.Nd missed music
+.
+.Sh DESCRIPTION
+I just realized that I totally forgot
+some important music for me from last year
+in my Albums 2021 post,
+because it wasn't in my playlist.
+Last year I watched
+the Berserk anime from 1997,
+and its soundtrack is incredible.
+.
+.Pp
+Actually the only reason
+I started watching it at all
+was because of the music.
+I was watching the wayneradiotv stream,
+.Do
+Mon repas durant un temps de tristesse;
+un pizza je n'oublierai jamais
+.Dc
+and I was mesmerized by the Guts theme.
+I had to find out what it was from.
+This was also around the time
+that Kentaro Miura died
+so people were really talking about it.
+Anyway just hearing
+that part of the soundtrack
+got me to start watching the anime,
+since you can find it all on youtube.
+.
+.Pp
+The anime in general did not disappoint.
+Actually it's really fucking good,
+and so is the rest of the soundtrack.
+The title sequence and credits tracks
+are so good that I let them play
+every episode even though
+I watched the series over only like 2 days.
+.
+.Pp
+I absolutely love whatever genre this stuff is.
+Is '90s anime intros its own genre?
+Something about combining
+acoustic and electric guitars,
+maybe.
+I'm also fond of
+the poorly written english lyrics.
+They're poetic in a distinctive way.
+I feel the same about
+that Shinsei Kamattechan
+song that was used for the credits of
+Attack on Titan season 2.
+Honestly awesome to write lyrics
+in a second language you haven't mastered.
+.
+.Pp
+So,
+the intro track,
+.Em Tell Me Why .
+First off,
+that sword sound effect
+near the beginning rules.
+Put that in more songs.
+What I really can't get enough of
+on this track are
+the quiet shouty vocals
+a bit off to the left
+during the chorus.
+It's such a cool idea
+to have clean lead vocals
+and shouting in the background.
+.
+.Pp
+And the credits track,
+.Em "Waiting So Long" .
+That first low note is so good.
+This is really a perfect credits song
+for the atmosphere of the show.
+It's creeping.
+The dual vocals
+the whole way through
+are such an interesting texture.
+Both of these tracks
+have really cool vocal sounds.
+And that dirty final guitar chord
+is a great sound to end on.
+.
+.Sh SEE ALSO
+These aren't great quality uploads
+but this stuff is sadly hard to find.
+.Bl -tag -width Ds
+.It "Guts"
+.Lk https://youtu.be/vZa0Yh6e7dw
+.It "Earth"
+.Lk https://youtu.be/5iAViNf9Z4Y
+.It "Penpals \(em Tell Me Why"
+.Lk https://youtu.be/I2rV8oKWSdM
+.It "Silver Fins \(em Waiting So Long"
+.Lk https://youtu.be/70GD2SBCq64
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+.Dq I like swinging my sword in battle.
diff --git a/www/text.causal.agency/036-compassion.7 b/www/text.causal.agency/036-compassion.7
new file mode 100644
index 00000000..9d0d887d
--- /dev/null
+++ b/www/text.causal.agency/036-compassion.7
@@ -0,0 +1,105 @@
+.Dd March 31, 2022
+.Dt COMPASSION 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Compassion
+.Nd better world fiction
+.
+.Sh DESCRIPTION
+Recently I watched the film
+.Em Margarita With a Straw .
+I'm not sure how to feel
+about some aspects of it,
+but it tries to do a lot,
+and I was still thinking about it
+a couple days later.
+.
+.Pp
+What really sticks out about it,
+to me,
+is that it is
+better world fiction,
+for lack of a better term.
+It's a film about two characters
+with disabilities,
+but it doesn't play into tropes.
+There's no big dramatic scene
+where a character gets treated unfairly.
+It doesn't really happen.
+In the world of the movie,
+most people are accepting,
+patient
+and compassionate.
+That's not to say
+there is no conflict.
+The film is just telling a different story.
+.
+.Pp
+The story takes place
+in a better world.
+Or maybe it takes place
+in a world that exists
+within our own,
+hidden between the worse parts.
+It's wonderfully subversive.
+Because I went into the film
+expecting at least one deeply upsetting
+scene of discrimination.
+What else would you expect
+of a story like this one,
+right?
+But instead of being upset,
+I was warmed.
+It was so nice to see
+the characters work through
+their own problems
+surrounded by simple kindness.
+And when it was over,
+I was left wanting
+to move our world
+closer to that one.
+.
+.Pp
+That's what I love about this kind of fiction.
+It's why I love the books of Becky Chambers so much.
+They give me hope,
+and guidance.
+I count the
+.Em Murderbot Diaries
+series in this as well,
+which shows a sort of bad world,
+and an alternative.
+I think it's so important
+to see the good that exists
+and the good that could exist.
+Rather than something to fight against,
+these stories show something to fight for.
+A more compassionate world.
+.
+.Pp
+I know,
+one person can't change the world.
+But they can change their own world,
+and the worlds of those around them.
+And slowly,
+good things can spread.
+I'll strive to be
+more patient,
+more understanding,
+more compassionate,
+and I hope you will too.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Pp
+I can't help but worry,
+when I write something like this,
+that someone I know will read it
+and think that I'm lying
+because I've hurt them.
+If that's the case,
+I am sorry,
+and I promise
+I am trying to do better.
diff --git a/www/text.causal.agency/037-care.7 b/www/text.causal.agency/037-care.7
new file mode 100644
index 00000000..052a4727
--- /dev/null
+++ b/www/text.causal.agency/037-care.7
@@ -0,0 +1,167 @@
+.Dd April  3, 2022
+.Dt CARE 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Care
+.Nd trans stuff in Montreal
+.
+.Sh DESCRIPTION
+This kind of info
+is frustratingly hard to find
+even from support orgs
+and the like.
+I think it's unlikely
+that anyone in my blog's audience
+is also someone who needs this info,
+and my blog isn't easy to find either,
+but I want to at least
+make it available somewhere.
+Really this is just like
+the posts where I figure out
+how to do something with a computer
+then I write it down.
+.
+.Pp
+Prices obviously change,
+by which I mean they inevitably go up,
+but I'm gonna give the amounts I paid
+in 2021\(en2022.
+Also if you want more details
+about any of this
+please email me.
+I will be happy to tell you all about it.
+.
+.Ss Medication
+I get HRT through
+Dr. Gabrielle Landry
+at La Clinique A,
+which is a private clinic.
+I've done everything over the phone.
+After the first consultation,
+I signed an informed consent form
+and had a prescription the next day,
+which I could start
+after I got an initial blood test.
+The information I found
+said to contact a specific person
+at the clinic with a direct phone number,
+which is what I did.
+Email me if you want that number.
+.
+.Pp
+I paid $300 for the first consult,
+$195 for the first followup,
+and $75 for further followups.
+I think annual appointments
+are more expensive
+than the followups.
+I've been getting blood tests done at a CLSC,
+which is free.
+On the public drug insurance plan,
+I paid $30-$35
+for my prescriptions
+as my dosage increased.
+I have private insurance now
+that entirely covers prescriptions,
+so I'm not sure what I'd be paying
+for my current prescription
+on the public plan.
+.
+.Ss Hair removal
+I tried laser hair removal,
+for longer than I should have.
+It was a waste of time and money.
+Do not believe any arguments about
+its convenience over electrolysis.
+.
+.Pp
+I've started getting electrolysis done
+with Dimi.
+Again,
+feel free to email me for contact info.
+He is very good and can do long sessions.
+I really don't find it very painful,
+which I think is partly my own pain tolerance
+and partly good equipment and skill.
+I've also found that taking acetaminophen beforehand
+and dressing warmly to keep my body relaxed help.
+I've paid $85 for hour-long sessions
+and $160 for two-hour sessions.
+I'm still early in treatment,
+but I'm really happy with the results so far!
+.
+.Ss Sex & name change
+The form for this is
+.Do
+Application to Change the Sex Designation
+of a Person 18 Years of Age and Over
+.Dc
+from the
+.Em Directeur de l'\('etat civil .
+It's self-ID,
+but you have to get it signed by
+someone you know
+and a commissioner for oaths.
+Julien at P10 is qualified for that
+and was super nice.
+We did it over Zoom.
+It's a free service,
+so I made a donation to P10.
+.
+.Pp
+I paid $144 to file mine
+but it's now FREE
+the first time you do it.
+Also $17 to mail it.
+Surprisingly,
+I got an acknowledgment letter
+.Po
+just saying they got it
+and would start looking at it
+.Dq shortly
+.Pc
+like a week and a half
+after I mailed the application.
+My cheque was cashed
+39 days after the date
+on the acknowledgment,
+and I got a
+.Dq favourable decision
+a week later.
+It takes another 30 days
+to get the certificate of change,
+after which you can
+order a new birth certificate
+and RAMQ will (slowly) send you a form
+to get a new card.
+In all it took about 4 months
+from when I mailed the application
+to having ID with my name on it.
+.
+.Ss Therapy
+I'm not seeking therapy
+for gender specifically,
+but I would like to find a good therapist
+that's aware of it.
+I'll update this
+if I find one.
+.
+.Ss Piercings
+Ok I know this isn't trans-specific
+but at least for me getting piercings
+was gender-affirming.
+Cuz I got nipple piercings lol.
+Anyway,
+I went to Mauve.
+They're super nice,
+really know what they're doing,
+and their website has lots of info.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Pp
+If somehow you did find this useful,
+I'd love for you to email me
+and let me know how things went for you.
diff --git a/www/text.causal.agency/038-agency.7 b/www/text.causal.agency/038-agency.7
new file mode 100644
index 00000000..f99a070b
--- /dev/null
+++ b/www/text.causal.agency/038-agency.7
@@ -0,0 +1,85 @@
+.Dd April 14, 2022
+.Dt AGENCY 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Agency
+.Nd origin of a name
+.
+.Sh DESCRIPTION
+When I registered this domain name,
+it was aspirational.
+Intentionally so.
+I wanted a new domain
+for a new identity,
+and I was thinking about personhood.
+That's what causal agency means.
+.
+.Pp
+It really was aspirational
+for me at the time.
+I spent a lot of time
+wishing I could be a person,
+because I didn't feel like one.
+I didn't feel real,
+like everyone else was.
+I didn't have any power
+over my own life.
+Things just happened to me,
+and I watched.
+There wasn't really a
+.Dq me
+there.
+The world was something that happened
+but that I couldn't interact with.
+I felt like that
+for most of my life.
+.
+.Pp
+But at some point
+I decided that,
+even if I wasn't now,
+one day I hoped to be an actual real life person.
+Like most programmers
+I am dreadful at naming things,
+so I didn't come up
+with this clever domain name
+myself.
+I typed
+.Dq person
+into some thesaurus,
+and it gave back
+.Dq causal agent ,
+and I realized
+agency is a TLD now.
+.
+.Pp
+Maybe it's a little dramatic
+to label myself with the thing
+I didn't think I had.
+But who knows,
+maybe it helped.
+Because it took a few years,
+but I did become a person.
+I feel real now.
+I can change my own life
+and the world around me.
+I have causal agency.
+.
+.Pp
+I am really proud of this domain name.
+I'm proud to put it on everything I make.
+Every instance of it
+is a reminder
+that I did what I set out to do,
+and that I'm still doing it.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.
+.Pp
+If anything in this post resonates with you,
+I want you to know that,
+whatever you think you can't do,
+it is possible,
+and you'll get there one day.
diff --git a/www/text.causal.agency/039-apologies.7 b/www/text.causal.agency/039-apologies.7
new file mode 100644
index 00000000..1b15076a
--- /dev/null
+++ b/www/text.causal.agency/039-apologies.7
@@ -0,0 +1,81 @@
+.Dd September 19, 2022
+.Dt APOLOGIES 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Apologies
+.Nd making them
+.
+.Sh DESCRIPTION
+Apologies are very important to me.
+Unfortunately
+I've only recently realized
+how valuable they are.
+I've tried to think about
+what makes a good apology,
+since it's not something
+I was ever taught.
+This is the advice
+I came up with for myself,
+on how to apologize.
+.
+.Bl -enum
+.It
+Make the apology.
+This is the most important part.
+If you feel guilty
+for something you've done,
+or think you might have hurt someone,
+apologize.
+Even if they don't need an apology,
+saying sorry won't hurt.
+And start with that.
+Literally say
+.Dq I'm sorry .
+Sometimes people forget that.
+.Pp
+On the other side,
+if you've been hurt by someone,
+and you trust them,
+let them know.
+Give them a chance to apologize.
+People don't always realize
+they've made a mistake.
+.
+.It
+Explain what you did wrong.
+I think it's important
+for the other person
+to know you understand
+how you've messed up.
+Really think about this!
+It's what will help you learn.
+If you know you've hurt someone
+but you're not sure why,
+you can try asking them.
+Take their answer seriously.
+.
+.It
+Don't make excuses.
+Do not talk about yourself.
+Don't even mention
+how you were feeling stressed that day,
+or whatever.
+It's not relevant.
+We all make mistakes,
+we all have bad days.
+.
+.It
+Commit to doing better.
+Try to learn from your mistakes.
+Say it won't happen again.
+Literally say
+.Dq I won't do that again .
+And then try your hardest to make that true.
+An apology is a commitment,
+not something you're done with
+once you've said it.
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/040-sound-memory.7 b/www/text.causal.agency/040-sound-memory.7
new file mode 100644
index 00000000..c995de08
--- /dev/null
+++ b/www/text.causal.agency/040-sound-memory.7
@@ -0,0 +1,165 @@
+.Dd November 14, 2022
+.Dt SOUND-MEMORY 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm Sound Memory
+.Nd associations
+.
+.Sh DESCRIPTION
+.Ss Talking Heads \(em "Remain In Light"
+The first time I gave this album a serious listen
+was when I was going for several-hour walks
+at 4 in the morning in,
+I think,
+fall 2020.
+I would stay up all night,
+go out walking at 4am
+for a couple hours,
+come home,
+eat
+.Dq breakfast
+and go to sleep.
+I listened to this album
+walking on completely empty
+big city streets
+in the dark.
+.
+.Ss Buffy Sainte-Marie \(em Up Where We Belong
+I started listening to this album
+after hearing it many mornings
+walking into the cafe on my block
+back in 2019.
+I could tell Vincent was working
+if I heard this when I opened the door.
+.
+.Ss Molasses \(em Trilogie: Toil & Peaceful Life
+I listened to this when I had 8am classes
+in CEGEP.
+In particular my first semester philosophy course,
+which was in the forum.
+I usually got there even earlier
+because of how the bus schedules worked out.
+There was another girl in my class,
+who I always sat next to,
+who also got there early,
+but we never spoke outside of class.
+.
+.Ss Arcade Fire \(em Funeral
+This album just feels like walking outside
+in fresh snow in early winter,
+you know?
+.
+.Ss Molasses \(em Trouble at Jinx Hotel
+I listened to this when I was looking for an apartment.
+I specifically remember listening to it
+walking down Clark toward my new place
+to pick up my keys.
+.
+.Ss Arcade Fire \(em Neon Bible
+The song
+.Dq "No Cars Go"
+is strongly associated for me
+with my earliest gender feelings.
+It's how I date when I first
+started to feel like something was wrong.
+The Suburbs was released in 2010,
+so I was probably listening to Neon Bible
+in 2011.
+Ten years between that
+and coming out.
+.
+.Ss "Do Make Say Think" \(em "You You're a History In Rust"
+I remember hearing
+.Dq "A Tender History In Rust"
+for the first time
+at the office of my first job.
+Me and my coworkers stayed late,
+probably on a Friday night,
+drinking free tech startup booze.
+.
+.Ss mewithoutYou \(em It's All Crazy! It's All False! It's All a Dream! It's Alright
+I exclusively listened to this album
+on a high school trip to Europe.
+Every morning when we got on the bus,
+I heard
+.Dq Every Thought a Thought of You
+and every night before bed
+I listened to
+.Dq The King Beetle on a Coconut Estate .
+.
+.Ss Arcade Fire \(em The Suburbs
+I listened to this album a tonne
+when I was playing
+Minecraft and Urban Terror
+with my online friends
+while I was in high school.
+In particular I remember
+a backyard shed World of Padman map
+and the apartments Minecraft world.
+.
+.Ss Arcade Fire \(em Reflektor
+I associate
+.Dq Afterlife
+with the walk between Laurier metro
+and my first job,
+in the winter.
+Must've just been how the timing worked out
+with my commute at the time.
+.
+.Ss Swans \(em To Be Kind
+I listened to this on one of my playthroughs
+of Half-Life 2.
+In particular I associate
+.Dq Bring the Sun / Toussaint L'Ouverture
+with the Water Hazard chapter.
+.
+.Ss Wrekmeister Harmonies \(em Light Falls
+For a while I put this on whenever I
+left my apartment to go somewhere
+and it was already dark,
+so probably winter.
+.
+.Ss St. Vincent \(em MASSEDUCTION
+This,
+along with the next one,
+I think were all I listened to
+on a family vacation
+to Quebec City and New Brunswick
+some years ago.
+.
+.Ss SOPHIE \(em Oil of Every Pearl's Un-Insides
+Many hours on the road
+on that family vacation.
+Two albums on repeat.
+.
+.Ss Julia Holter \(em Aviary
+This is another album
+I listened to when I was taking
+walks at 4am.
+I wasn't in a good place.
+Yet.
+.
+.Ss Beep Test \(em Laugh Track
+A tape from the first act
+at one of my favourite shows
+I've ever been to,
+at La Sotterenea
+in Suoni 2019.
+I wish I had been out already.
+.
+.Ss The Armed \(em Only Love
+The third of the albums I listened to
+on those dark walks.
+I listened to it loud,
+this album's mixing needs it.
+.
+.Ss Eliza Kavtion \(em The Rez That Summer
+A favourite local artist.
+I remember vividly the first time
+I heard her play,
+opening for Wrekmeister Harmonies
+at La Vitrola in 2018.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/041-albums-2022.7 b/www/text.causal.agency/041-albums-2022.7
new file mode 100644
index 00000000..48bd3c3d
--- /dev/null
+++ b/www/text.causal.agency/041-albums-2022.7
@@ -0,0 +1,185 @@
+.Dd December 21, 2022
+.Dt ALBUMS-2022 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm albums 2022
+.Nd review
+.
+.Sh DESCRIPTION
+it's the year-end review
+of albums I listened to.
+same as last year,
+I added any albums I got into
+this year to a playlist.
+I've actually done that
+every year since 2018.
+maybe I'll review
+those old playlists some time.
+.
+.Ss ZHAOZE \(em SUMMER INSECTS TALK ABOUT ICE (2021)
+it's a five-and-a-half-minute album!
+you can loop it however long you want.
+it's really lovely.
+.Pp
+favourite track:
+ON HORSEBACK, TO FARAWAY
+.
+.Ss KATE BUSH \(em HOUNDS OF LOVE (1985)
+first of all I do not watch that one show.
+I've known that track for a while actually.
+I mean I probably first heard the CHROMATICS cover.
+but anyway,
+I think someone mentioned this album
+on IRC at just the right time
+and I put it on.
+the second half really shines tbh.
+love a concept album.
+.Pp
+favourite tracks:
+RUNNING UP THAT HILL,
+HOUNDS OF LOVE,
+AND DREAM OF SHEEP,
+WATCHING YOU WITHOUT ME,
+THE MORNING FOG.
+.
+.Ss GODSPEED YOU! BLACK EMPEROR \(em ALL LIGHTS FUCKED ON THE HAIRY AMP DROOLING (1994)
+didn't expect to hear this probably ever.
+still wild that it finally got uploaded.
+and to be honest I'm a little mad
+that it's actually good.
+like yeah it's not a godspeed album
+but it holds up as a tape on its own.
+it's the kind of shit I listen to.
+also can't believe some people
+still thought it was fake.
+like have you not heard
+any other efrim menuck projects?
+.Pp
+favourite tracks:
+$13.13,
+DIMINISHING SHINE,
+DADMOMDADDY,
+333 FRAMES PER SECOND,
+ALL ANGELS GONE.
+.
+.Ss BLACK DRESSES \(em FORGET YOUR OWN FACE (2022)
+woops I think I only listened to this like twice.
+will need to revisit it later for sure.
+I'll like it.
+.
+.Ss BACKXWASH \(em I LIE HERE BURIED WITH MY RINGS AND MY DRESSES (2021)
+only got into this album
+after hearing it live this summer.
+was the first show I went to in years
+and it was really fucking good.
+gotta listen to this shit loud.
+sampling godspeed for a beat fucks.
+honestly back to back bangers.
+.Pp
+favourite tracks:
+I LIE HERE BURIED WITH MY RINGS AND MY DRESSES,
+TERROR PACKETS,
+SONG OF SINNERS,
+BURN TO ASHES.
+.
+.Ss PHILIP GLASS ENSEMBLE \(em EINSTEIN ON THE BEACH (1979)
+actually just the knee plays
+because I can't be bothered
+listening to all of it.
+and I'm embarrassed by how much
+I enjoy this avant-garde bullshit.
+like ok just sing repeating numbers at me
+and my brain is happy.
+what is this?
+my kink?
+anyway I also have kind of an obsession
+with the
+.Dq story of love
+in knee 5.
+I fucking hate it.
+but it's delivered so well.
+and that violin though!
+.Pp
+favourite tracks:
+KNEE 1,
+KNEE 5.
+.
+.Ss KANYE WEST \(em YEEZUS (2013)
+ok look I listened to this
+before recent events.
+what the fuck.
+it's a really good album though?
+pretty sure I listened to it
+because bound 2 kept getting in my head,
+because of that minecraft parody parody
+wayne did ages ago.
+.Pp
+favourite tracks:
+BLACK SKINHEAD,
+HOLD MY LIQUOR,
+BLOOD ON THE LEAVES,
+BOUND 2.
+.
+.Ss FLYING RACCOON SUIT \(em AFTERGLOW (2021)
+I've listened to the whole album
+a few times
+but I'm mostly just here
+for the title track.
+this also happened to be
+dropped in IRC at just the right time.
+good ska-punk-type shit.
+and I like lisps ok.
+.Pp
+favourite track:
+AFTERGLOW.
+.
+.Ss RAMSHACKLE GLORY \(em LIVE THE DREAM (2011)
+one of those albums
+I don't know why I took so long
+to get to.
+I've been listening to johnny hobo
+since I was like in high school.
+ramshackle is a little more hopeful
+and I love that.
+your heart is a muscle the size of your fist.
+keep on loving.
+keep on fighting.
+.Pp
+favourite tracks:
+WE ARE ALL COMPOST IN TRAINING,
+NEVER COMING HOME,
+YOUR HEART IS A MUSCLE THE SIZE OF YOUR FIST.
+.
+.Ss LES RALLIZES D\('ENUD\('ES \(em THE OZ TAPES (2022)
+a pleasant surprise in someone's playlist.
+lately I've been listening to this
+in the metro to or from electrolysis.
+it's good listening for that.
+bold to have two versions
+of the same 24-minute song
+on the same release.
+.Pp
+favourite tracks:
+A SHADOW ON OUR JOY,
+THE LAST ONE_1970 (ver.2).
+.
+.Ss LINGUA IGNOTA \(em SINNER GET READY (2021)
+another I'm only getting into
+after hearing it live.
+just last sunday actually.
+was a good show.
+people will go wild
+to hear a cover live for real.
+.Pp
+favourite tracks:
+I WHO BEND THE TALL GRASSES,
+PENNSYLVANIA FURNACE,
+PERPETUAL FLAME OF CENTRALIA.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+I started writing this
+before I saw LINGUA IGNOTA.
+good thing I waited.
diff --git a/www/text.causal.agency/042-comfort-music.7 b/www/text.causal.agency/042-comfort-music.7
new file mode 100644
index 00000000..445e04c3
--- /dev/null
+++ b/www/text.causal.agency/042-comfort-music.7
@@ -0,0 +1,62 @@
+.Dd February 23, 2024
+.Dt COMFORT-MUSIC 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm comfort music
+.Nd feel better
+.
+.Sh DESCRIPTION
+it's been a while.
+and I'm on almost no sleep
+and haven't eaten a real meal
+since noon.
+which is a state I've written
+at least a couple posts in before,
+so what better time
+to return to what has apparently
+become this blog's format:
+lists of some music I like.
+.
+.Pp
+this is a list of music that comforts me.
+.
+.Bl -bullet
+.It
+knee play 5, from einstein on the beach.
+I like the organ and the counting and the cadence of the story.
+.It
+low \(em words.
+and I'm tired.
+.It
+godspeed you! black emperor \(em storm.
+this is like my original comfort music.
+been listening to it since I was teenage.
+the grooves are worn deep in my mind.
+.It
+set fire to flames \(em love song for 15 ontario (w/ singing police car).
+I like how it ends.
+.It
+va, from the beginner's guide.
+I think that's the whole point.
+though maybe it's too sad
+to be truly comforting.
+.It
+undertale, from undertale.
+what can I say?
+.It
+wrekmeister harmonies \(em covered in blood from invisible wounds.
+I find quite a bit of the album comforting really.
+I'm picking this one because I like the cadence
+of the lyrics.
+.It
+lingua ignota \(em pennsylvania furnace and perpetual flame of centralia.
+these are really my go to in recent times.
+I like waiting for the next line.
+.El
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
+.Pp
+I don't think I've said anything
+very interesting here.
diff --git a/www/text.causal.agency/043-little-blessings.7 b/www/text.causal.agency/043-little-blessings.7
new file mode 100644
index 00000000..957c6289
--- /dev/null
+++ b/www/text.causal.agency/043-little-blessings.7
@@ -0,0 +1,78 @@
+.Dd March 24, 2024
+.Dt LITTLE-BLESSINGS 7
+.Os "Causal Agency"
+.
+.Sh NAME
+.Nm little blessings
+.Nd life's
+.
+.Sh DESCRIPTION
+today I went out to go around.
+run some errands and do some shopping.
+along the way I was given
+several of life's little blessings.
+.
+.Pp
+while walking on ste-cath
+between berri and complexe desjardins,
+there was a somewhat disheveled man
+walking in the same direction and singing.
+he had a beautiful voice.
+he was singing a sad song in french,
+and he sung it well and enunciated every word.
+.
+.Pp
+in the mcdonald's at complexe desjardins,
+while waiting for my order,
+there were what appeared to be
+a teenager and her younger brother,
+who must have been
+looking at the display of
+current happy meal toys.
+the teenager was playing smash or pass,
+to the amusement of the younger one.
+they got ice cream
+and ate it across the room from me downstairs.
+.
+.Pp
+later,
+taking the 24 home from atwater
+carrying my new vacuum cleaner,
+the bus got lost.
+I think the driver missed the stop
+and tried to compensate
+by turning north onto peel
+and stopping there.
+but then he had to keep going up peel.
+he turned right onto docteur-penfield,
+which just brings you further up the mountain.
+when it met des pins,
+he turned left and pulled over,
+asking for guidance over the radio.
+we got moving again,
+back towards peel.
+that's how I ended up
+on a 24
+.Dq sherbrooke
+east,
+facing west on des pins.
+it was actually quite scenic.
+and amusing.
+I was in no rush.
+.
+.Pp
+after getting back onto sherbrooke,
+the bus had to take another detour,
+this one planned.
+so my ride on the 24,
+which normally only drives on sherbrooke,
+ended up going on peel,
+docteur-penfield,
+des pins,
+de bleury,
+ren\('e-l\('evesque
+and saint-laurent.
+it was a very exciting bus trip.
+.
+.Sh AUTHORS
+.An june Aq Mt june@causal.agency
diff --git a/www/text.causal.agency/Makefile b/www/text.causal.agency/Makefile
new file mode 100644
index 00000000..a8683a20
--- /dev/null
+++ b/www/text.causal.agency/Makefile
@@ -0,0 +1,66 @@
+WEBROOT = /var/www/text.causal.agency
+
+TXTS += 001-make.txt
+TXTS += 002-writing-mdoc.txt
+TXTS += 003-pleasant-c.txt
+TXTS += 004-uloc.txt
+TXTS += 005-testing-c.txt
+TXTS += 006-some-libs.txt
+TXTS += 007-cgit-setup.txt
+TXTS += 008-how-irc.txt
+TXTS += 009-casual-update.txt
+TXTS += 010-irc-suite.txt
+TXTS += 011-libretls.txt
+TXTS += 012-inability.txt
+TXTS += 013-hot-tips.txt
+TXTS += 014-using-vi.txt
+TXTS += 015-reusing-tags.txt
+TXTS += 016-using-openbsd.txt
+TXTS += 017-unpasswords.txt
+TXTS += 018-operating-systems.txt
+TXTS += 019-mailing-list.txt
+TXTS += 020-c-style.txt
+TXTS += 021-time-machine.txt
+TXTS += 022-swans-are-dead.txt
+TXTS += 023-sparse-checkout.txt
+TXTS += 024-seprintf.txt
+TXTS += 025-v6-pwd.txt
+TXTS += 026-git-comment.txt
+TXTS += 027-openbsd-linode.txt
+TXTS += 028-names.txt
+TXTS += 029-topics.txt
+TXTS += 030-discs.txt
+TXTS += 031-books-2021.txt
+TXTS += 032-albums-2021.txt
+TXTS += 033-jorts.txt
+TXTS += 034-voices.txt
+TXTS += 035-addendum-2021.txt
+TXTS += 036-compassion.txt
+TXTS += 037-care.txt
+TXTS += 038-agency.txt
+TXTS += 039-apologies.txt
+TXTS += 040-sound-memory.txt
+TXTS += 041-albums-2022.txt
+TXTS += 042-comfort-music.txt
+TXTS += 043-little-blessings.txt
+
+all: colb ${TXTS}
+
+.SUFFIXES: .7 .fmt .txt
+
+.7.txt:
+	mandoc -T utf8 $< | ./colb > $@
+	touch -m -r $< $@
+
+.fmt.txt:
+	fmt $< | sed '1,/^$$/d' > $@
+	touch -m -r $< $@
+
+feed.atom: feed.sh colb ${TXTS}
+	sh feed.sh > feed.atom
+
+clean:
+	rm -f colb ${TXTS} feed.atom
+
+install: colb ${TXTS} feed.atom
+	install -p -m 644 ${TXTS} feed.atom ${WEBROOT}
diff --git a/www/text.causal.agency/colb.c b/www/text.causal.agency/colb.c
new file mode 100644
index 00000000..5faabc3a
--- /dev/null
+++ b/www/text.causal.agency/colb.c
@@ -0,0 +1,16 @@
+#include <locale.h>
+#include <stdio.h>
+#include <wchar.h>
+int main(void) {
+	setlocale(LC_CTYPE, "en_US.UTF-8");
+	wint_t next, prev = WEOF;
+	while (WEOF != (next = getwchar())) {
+		if (next == L'\b') {
+			prev = WEOF;
+		} else {
+			if (prev != WEOF) putwchar(prev);
+			prev = next;
+		}
+	}
+	if (prev != WEOF) putwchar(prev);
+}
diff --git a/www/text.causal.agency/feed.sh b/www/text.causal.agency/feed.sh
new file mode 100644
index 00000000..71bbf662
--- /dev/null
+++ b/www/text.causal.agency/feed.sh
@@ -0,0 +1,58 @@
+#!/bin/sh
+set -eu
+
+readonly Root='https://text.causal.agency'
+
+updated=$(date -u '+%FT%TZ')
+cat <<-EOF
+	<?xml version="1.0" encoding="utf-8"?>
+	<feed xmlns="http://www.w3.org/2005/Atom">
+	<title>Causal Agency</title>
+	<author><name>June</name><email>june@causal.agency</email></author>
+	<link href="${Root}"/>
+	<link rel="self" href="${Root}/feed.atom"/>
+	<id>${Root}/</id>
+	<updated>${updated}</updated>
+EOF
+
+encode() {
+	sed '
+		s/&/\&amp;/g
+		s/</\&lt;/g
+		s/"/\&quot;/g
+	' "$@"
+}
+
+set -- *.txt
+shift $(( $# - 20 ))
+for txt; do
+	entry="${txt%.txt}.7"
+	test -f "$entry" || entry="${txt%.txt}.fmt"
+	date=$(grep '^[.]Dd' "$entry" | cut -c 5-)
+	title=$(grep -m 1 '^[.]Nm' "$entry" | cut -c 5- | encode)
+	summary=$(grep '^[.]Nd' "$entry" | cut -c 5- | encode)
+	published=$(date -ju -f '%B %d, %Y %T' "${date} 00:00:00" '+%FT%TZ')
+	mtime=$(stat -f '%m' "$entry")
+	updated=$(date -ju -f '%s' "$mtime" '+%FT%TZ')
+	cat <<-EOF
+		<entry>
+		<title>${title}</title>
+		<summary>${summary}</summary>
+		<link href="${Root}/${txt}"/>
+		<id>${Root}/${txt}</id>
+		<published>${published}</published>
+		<updated>${updated}</updated>
+		<content type="xhtml">
+		<div xmlns="http://www.w3.org/1999/xhtml">
+	EOF
+	printf '<pre>'
+	encode "$txt"
+	cat <<-EOF
+		</pre>
+		</div>
+		</content>
+		</entry>
+	EOF
+done
+
+echo '</feed>'