live coding interviews are just humiliation rituals

@benis_redux that can help formulate an opinion I'll give you that, I still think there's more than just numbers

@nuko just formulate your own opinion

@pernia you wish you knew where I lived

@mischievoustomato absolutely not never no

I love people that say "X place sucks" when they've never been there. If you've been there and say "X place sucks" then at least you were able to independently formulate an opinion which is infinitely more respectable than mindlessly regurgitating someone else's opinion

@splitshockvirus they're missing the blockchain

@meso @mischievoustomato as someone who has writers block all the time it might be ok for outlines but the only stuff I would directly copy and paste from the slop trough is something I would not care about whatsoever

@mischievoustomato @meso they're using it for its current main intention, cheating on assignments

@meso @mischievoustomato the good thing at least is most people don't pay for AI garbage they usually use it for free

@mischievoustomato also Intel is kind of a complete joke in the consumer x86 space and has been since basically Zen 3 came out in 2019, their mobile processors are distinctly "not shit" though

@mischievoustomato Apple cheats because their unified memory architecture allows macOS to cram as much memory into the GPU as humanly possible

@meso it is a grift, looting from the gullible (VC and investor types looking for the next BIG THING (tm)) and making garbage and not actually improving products whatsoever

@meso and a money pit but now I can yell at my GPU and make it do funny text dances with ollama so I guess it was worth a few billion dollars

@benis_redux that was four months ago I noticed it now with the new Framework laptops

AMD labelling their new mobile CPUs as "Ryzen AI" when they have a little NPU or something is very funny and definite proof that we are not in a bubble whatsoever

@EllisDee @mischievoustomato ubuntu doesn't work sometimes either, 25.04 broke wireguard for me until I moved the wireguard config to where Canonical wanted it to be

@phnt @hj I have no goddamn idea because the error is kind of vague and downgrading to 3.12.5 fixes the issue but when running 3.13 I get this error when anyone tries to block or mute:

1) test blocking the conversation with the blocked user is not marked as unread on a reply (Pleroma.Conversation.ParticipationTest)
     test/pleroma/conversation/participation_test.exs:315
     ** (ArgumentError) empty list of fields to update, use the `:replace` option instead
     code: {:ok, _user_relationship} = User.block(blocker, blocked)
     stacktrace:
       (ecto 3.13.2) lib/ecto/repo/schema.ex:927: Ecto.Repo.Schema.on_conflict/6
       (ecto 3.13.2) lib/ecto/repo/schema.ex:479: anonymous fn/15 in Ecto.Repo.Schema.do_insert/4
       (pleroma 3.15.2-1016-g31bf1ec2-froth-zone+test) lib/pleroma/user.ex:2664: Pleroma.User.add_to_block/2
       test/pleroma/conversation/participation_test.exs:326: (test)

I don’t know ecto or elixir enough to figure out the fix but I think it has something to do with https://github.com/elixir-ecto/ecto/issues/4633 because that is the same exact error

@phnt @hj neither could I on the akko side my issue was ecto being weird which I fixed by making it no longer weird it had nothing to do with the frontend

@phnt @hj Good news, I fixed the block/mute issue so that just works on pleroma-fe without issues.

Notifications don't though without disabling status notifications, documented at https://git.pleroma.social/pleroma/pleroma-fe/-/issues/1388