I think someone asked. But is it possible to classically convert a PDF into JSON? And the answer is yes, yes you can. The feats of programming know no end (it’s out of my explanatory scope, sorry).

qpdf --json-output my.pdf my.json

I’m finally realizing that in the stack link of this post, there’s actually an entry about “grepping” the device. Everything’s a file or something. It feels comical.

A quick wrapper script for normalized git commit dates. Commit dates and timezones are normalized or truncated to the day. This is useful for avoiding rebasing and/or not having to think about system or git date/time settings when on another device or in another timezone. If precise timestamps matter, then it is easily disabled.

#!/bin/sh -eu

GIT=$(which git --all | grep --invert-match "local/bin" | head --lines 1)
GIT_COMMITTER_DATE="$(date --utc --date '0' '+%a %b %d %H:%M:%S %Y %z')"

export GIT_COMMITTER_DATE

[ "${1:-}" = "commit" ] && export DATE=1 && $GIT "$@" --date="$GIT_COMMITTER_DATE"

[ "${DATE:-}" = 1 ] && exit

$GIT "$@"

There’s truly nothing more liberating than deleting all of your work by mistake (or having it deleted). That’s why it never hurts to have multiple distribution (and backup) strategies. A newer way is invented every day:

gcc -o delete-this-file.c

Why dabble with Firefox? Everything is customizable. How long will this be so? Who knows. E.g, I have custom stylesheets that hide comments everywhere (Firefox/Chrome) among other things. “Experts” only?

about:debugging#/runtime/this-firefox

chrome://browser/content/browser.xhtml

toolkit.legacyUserProfileCustomizations.stylesheets true