Claude 4.0 review so far: Its definitely the best coding model, really solid stuff there. It doesn't necessarily generate better code on a function by function level, but it stays on track for longer tasks much better, and its a lot better at spotting issues.
The β€œlets tell anyone we dont like to kill themselves” crowd are not welcome here sorry
LLMs are enough of an improvement to my workflows that i find people who call them useless really funny. Yes they make mistakes, but so do i, and the loop to fixing those mistakes is shorter with their help.
So, thing I learned today that Claude is really good at. Log Parsing. Apparently you can just shove a ton of haproxy logs into it and ask "whats going on??" and it does a great job of diagnosing issues
If you want to give feedback on a feature first, inform yourself on how it works, dont make dumb assumptions second, dont just spam shit on a pull request that the dev working on the project has to deal with
[deer.social]( ) looks pretty cool, appears to be a fork of our open source app with more robust controls around labelers, neat!
Behold, my soon-to-be full bluesky index https://morel.us-east.host.bsky.network/xrpc/com.atproto.sync.getBlob?did=did:plc:vpkhqolt662uhesyj6nxm7ys&cid=bafkreiff2dgi7pm2ro4endvsyohc2snadrevvjebiwl6brghnihuipe6gi
Backfill is probably going to take a while, but its running! RE: https://morel.us-east.host.bsky.network/xrpc/com.atproto.sync.getBlob?did=did:plc:vpkhqolt662uhesyj6nxm7ys&cid=bafkreieznd7zl2g4s5peatt5cog5yyolzlgg23k2rm35yonrgxns47qvmm View quoted note β†’
ordered an rpi5 with an nvme drive, annoyed by people saying "its impossible to run an appview unless you have a billion dollars so just stay on mastodon where we can whine about things together"
My unified theory of political discourse centers around the fact that the quote post was invented in 2015