The video I saw had a disclaimer at the end saying that no device has been built yet.

Reminds me of this one, that made rounds some years ago and turned out to be fake as well.

It’s a mix of --codec copy to not reencode the streams, --map 0:1 to select the audio stream of the input file and for merging back you need multiple -i input files and then --map 0:0 to map the first input file’s video stream and --map 1 to map the second input file (English audio stream). Mappings may vary - use ffprobe to check your files.

You might want to look into Tdarr. Configure it once and let your old computer transcode away automatically.

For audio stream replacement - if you get the exact same video (same intro length, same scene lengths, same total length, etc.), you can easily replace the audio streams and it should work out. ffmpeg is your friend to do it lossless.

database schema improvements, which is a particularly tricky thing to work on

Not, if you know what you’re doing.

I dont like AI, I wish I could go back to the world before it, but I cant.

You do know that you can use LLMs to get inspiration and suggestions but then still implement those things by hand? Which is also a great way to make sure it doesn’t change something else in the background. Also, if you use Claude like you’ve used StackOverflow and Google before, you don’t even have to mention it as a Co-Author.

Pro-tip: Block the user “claude” on GitHub and it’ll show you a warning for every repository where they forgot to remove the Co-Author tags:

v3 Migration Guide

[!NOTE] Upgrading to Paperless-ngx v3 can only be performed from version 2.20.15. If you are running an older version, please upgrade to v2.20.15 before proceeding with the v3 upgrade.

dovecot/dovecot and solr should do the trick. At least they do for me.

EDIT: Two advantages of this setup for me: I can easily drag&drop emails from my other email accounts to this “archive” account in Thunderbird. And once I want to “seal” an archive folder, I can gzip the underlying mbox file and make it read-only (dovecot needs to be configured to store mails in mbox files).

Eh, there’s not really a better/worse between Gitea and ForgeJo. Gitea is targeting business customers, ForgeJo is targeting the open source community.

Yes, I’m using that after a short foray into Pixelfed. And together with @mczachurski@mastodon.social 's Impressia iOS app it’s a really good experience.

You can tune SearXNG results in a similar way. There’s just no easy UI for it.

And yes, it’s often slower than commercially available search engines - but it’s completely free, without trackers, and it’s combining results from various different engines. I can live with that.

Residential IP (runs in a Docker container on a Raspberry Pi 5 here), pretty much vanilla (apart from Marginalia).

Kagi is still $10+tax a month. That’s $120 a year (plus tax).

SearXNG is $0 and a few minutes of my time whenever I tweak the configuration.

No problems with my SearXNG here at all (just have to look into Qwant):

Do you know about the MAILTO= lines in crontabs? That’s exactly what they’re there for. (And your script needs to output error messages on failure, of course.) You’ll need a local mail forwarder like ssmtp or exa, though.

That’s what I settled for as well. Keeping my *.kdbx file in iCloud, doing nightly backups to my NAS, and using Strongbox as a client (there’s also KeePassium with similar features) - which beautifully integrates into Apple’s AutoFill API, so it feels native - i.e. as if you’re using Apple’s Passwords app. And I can access all passwords from Windows using KeePassXC as well.

If you host it on a VPS and stop paying the invoices, it’ll go down anyway. However, AFAIK, Bitwarden client apps cache all passwords - so your family would still be able to access them, but there would be no sync back, of course.

Have you considered using something like a KeePass database on a shared drive? Most modern client apps can sync changes seamlessly and there are browser plugins for all the major browsers.

For the curious people like me: AirTrail

I can get almost two Punkt MP02 for that.

From the linked article:

Two things that aren’t changing: every Strava athlete can still access and download their data for free, at any time – and wearable and device integrations are not affected.

Manually backdating is one thing, but importing them from Instagram with the original timestamp is a different. That’s why #2871 is still open and #4558 was closed as “not planned”, I guess.