I don't know what your exact criteria for "sportswatch" is, but I've been very happy with my Bangle.js 2: https://banglejs.com/

Here you go: https://adnauseam.io/

These days I think OMEMO is a better choice than OTR, if your client supports it.

Zulip is pretty great.

This is not true. In fact, Guix doesn't use systemd at all. When managing an operating system installation Guix uses its own init system called the Shepherd.

Elogind is a project extracted from systemd originally for use in Guix systems.

Another application of Parse, don't validate. The vulnerability isn't really about the single-line regex, it's that the validation doesn't match the use. If the regex extracted (read: parsed) the valid bit with a group, then passed that through, it would be fine.

I think IETF BCP 47 is the way to go for this (and is rougly a superset of what you're saying). They also define a "more/less specific" selationship between tags so you can ask for "en-CA" and have the system decide "en" is still appropriate.

I think your summary is less useful than reading the whole article. Teaching well involves more than just stating a refined idea.

I also think the ideas in the article are worth the 25 minutes.

If you wanted to do it in one query I think you could do something like

UPDATE local_user AS u
  SET u.totp_2fa_url = null,
      u.totp_2fa_secret = null
 FROM person AS p
WHERE p.id = u.person_id
  AND p.local
  AND p.name = 'guineapig';

I assume the p.local is optional, too, because the id match against the local_user table will presumably limit it to only local users. 🤷

What makes you think that Christians, as a whole, deny science?

Okay. I don't really want to keep arguing about this. Your negativity seems entirely unwarranted, so I'm not really sure what you're trying to achieve. Given the only thing I wanted to say is "this article is worth reading", I don't think there's a productive line for this conversation to go down.