bjoern_tantau

cross-posted from: https://discuss.online/post/17161733

http://www.smbc-comics.com/comic/cosmology-3

Alt text

I think we could increase the number of stem phds just by changing the title from Doctor to Ultimate Grandmaster.

Bonus panel

Ich hoffe dieser offene Brief kann etwas erreichen. ME ist die häufigste Ausprägung von Long Covid. War aber schon vorher Jahrzehnte lang bekannt. Trotzdem wissen noch nicht einmal die meisten Ärzte, dass es existiert oder tun es als Depression oder andere psychische Erkrankung ab.

Einfach nur mehr Bekanntheit würde mir und anderen Betroffenen viel bringen.

In my quest to tame my pict-rs instance I've disabled thumbnail proxying, deleted all thumbnails and images not uploaded by me and still the files folder is several GB big. And growing.

I am the only user of my instance. But iotop shows constant disc activity for pict-rs. It's writing stuff at about 1 MB/s. The webserver logs only show my own activity and occasionally someone requesting an image. But not nearly enough to explain this constant activity.

When I'm looking at the actual files at random it seems to be mostly memes and stuff, but not what I uploaded myself.

Does anyone have an idea how I can find out what is causing all this activity?

Edit: I've opened an issue at https://git.asonix.dog/asonix/pict-rs/issues/79

I'm currently migrating pictrs' database from sled to PostgreSQL and even during that it seems to be constantly writing data to sled. Doesn't feel right to me.

it's night.

Sorry, could not resist.

She bought a Steam Deck to play PC games because she joined BAFTA.

Ein wenig Hoffnung in diesen schwarzen Tagen.

By default Lemmy seems to be mirroring every link's preview image. Even when that link was originally posted on an external Lemmy instance. That can put quite a strain on small servers. Is there a way to disable this? Ideally I would have pict-rs just be used to host the images my users upload and nothing else.

Solved:

Add

  pictrs: {
    image_mode: "None"
  }

to lemmy.hjson.

Does anyone know what's up with that? Couldn't find anything via Google. Seems really fishy to me.

Edit: Got the official explanation from the dev on Reddit:

AutoClick Feature of JDownloader works as following. Open the browser and wait few seconds, then take screenshot and search for the Recaptcha click area and auto click on it. Screenshot is needed so JDownloader knows where to click. You can disable this feature, see https://support.jdownloader.org/de/knowledgebase/article/jd-opens-my-browser-to-display-captchas

On Linux, JDownloader creates screenshot to find out the color of tray area so it can try to find it's tray icon and calculate the correct background color for transparency. You see the JDownloader icon having white background. You can disable this via Settings->Advanced Settings->Tray.gnometrayicontransparentenabled

cross-posted from: https://midwest.social/post/21295973

Fame (2025-01-03)

http://smbc-comics.com/comic/fame

Alt text

Really every theorem should either describe what it does or have ten thousand names, chained by hyphens, going back to an early hominid named Garg.

Bonus panel

Tja

Hätt ja keiner Ahnen können.

Dr Becky, PBS Space Time and Angela Collier all release on the same days. Giving me one big day filled with quality astro physics. But leaving all other days devoid of good content. It would be nicer if they staggered a little.

Posting here for a lack of !verymildlyinconvenient.