Yup, crazy. I record all my coding screencasts at 15 fps and it looks fine while the video file is tiny.

For anyone else looking for a reason to stay away from plebbit- uh, look at their X account https://x.com/getplebbit
It’s 4chan crypto hashwash

Could it be something to do with the order of nouns and verbs in the sentence? English is quite different from the order of others.

Ok this is fixed now, for future posts from Mbin. Thanks for the mention.

@yoasif@fedia.io

I see the issue. Just a small fix, really. Will release 1.3.8 shortly.

@wjs018@piefed.social is this that thing mbin did with the images and links?

Nice one 👍

You might want to make the container name a command line parameter or env var, incase that changes in future.

Llama 4 Maverick did not blackmail

I don’t feel like they explored this well enough, probably because it did not fit the narrative they’re trying to build.

I am suspicious of Anthropic’s motives in publishing this. Maybe in a couple of weeks/months they’ll “solve” this problem they created, making themselves looking smart and relevant.

Or maybe Llama 4 is too stupid to be evil or useful. I don’t know enough about the models to judge.

I have just added code which makes the instance vote weight visible in the UI - https://codeberg.org/rimu/pyfedi/commit/04b54e25f44f54420c247152323c90c4a3360a02

You can go to https://piefed.social/instances and check the vote weight of any instance. It’ll be very boring tho because they’re all set to “Votes are weighted at the normal level.” Yes, even lemmy.ml.

I can see the background tasks framework getting a lot of use!

We’re all tired of you doing it. Please stop.

I’ve never used Sync but the screenshots on Google Play remind me of Blorp.

https://github.com/Blorp-Labs/blorp

https://blorpblorp.xyz/

No it is not ok, if that is the justification the therapist is giving to not to provide you with the medicine you need.

Same thing is happening in Iraq.

After all the fighting, suffering and genocide in the region, it’ll all be for nothing once climate change makes it uninhabitable.

I appreciate you showing us this interesting tool once or twice but it’s getting a bit overly self-promotional at this point.

Yes, it’s tricky. I don’t remember which instance it was but it was one of the smaller lemmy instances with roughly the same MAU as Piefed.social, tho. Not any of the top 6 Lemmy instances.

It could have been an outlier. Looks like lemmy.mods4ever.com is much more reasonable although that’s a single-user instance so it’s going to be the absolute minimum.

I can’t find/remember the post but recently an admin shared some resource usage graphs. The thing that jumped out at me was their Lemmy instance was using 32 GB of ram while their PieFed instance was using 2 GB of ram. Maybe a little more, but it was low single digits.

Maybe their piefed graph did not include postgresql, which should be 2 GB just by itself… That would bring the total up to around 4 or 5 GB. piefed.social uses between 4 and 5 GB, too.

CPU usage was quite similar between the two.

I really noticed that problem on Spotify. Lots of 2 minute tracks/songs that are just one musical idea with no development or iteration of it and really limited recommendations.

Tidal was much better.

You won’t be the only one, surely interviews are going to have to change once no one can do white boarding anymore!

A recent JetBrains survey I saw found that 85% of devs are using AI in some capacity.

I guess they get the corrections and adjustments done by telling the LLM what corrections and adjustments to make?