blurbleblurble

Did you consider incorporating ast-grep or gritql?

Congratulations, great work.

The author posted their own confession right here: https://pbs.twimg.com/profile_banners/591273520/1719711719/1...

"The author's confession is above..."

It's a rap on claude

Opus 4.6 was incredible but Opus 4.7 is genuinely frustrating to me so far. It's really sharp but can be so lazy. It's constantly telling me that we should save this for tomorrow, that it's time for bed (in the middle of the day), and very often quite sloppy and bold in its action. These adjustments are getting old. The next crop of open models seems ready to practically replace the big ones as sharp orchestrator agents.

I didn't say anything about illegality, I'm speaking strictly in terms of the ethics of please stop shitting in the pool shame on you.

Don't poop in the public swimming pool and don't feed perverse war games that result in children and their parents being maimed and killed.

what a waste of money!

just kidding this is amazing

"Information" is not neutrally reflective of some objective, deterministic reality. Processes involving information are slippery and reflexive, and that's the danger people are all but shouting for you to see. The feedback loops have extreme potential. The danger isn't in "exposing information" it's in people coercing reality to fit their bet. Peoples' lives are on the line.

The wild thing is that prediction market dabbler rationalists who think they're betting on the outcome of a pre-written fiction story are actually just upping the reward for someone to come along and bomb people to death. Your participation is not neutral. You are culpable in the outcome of the situation you're "gambling" on which has not been determined yet and you are literally throwing fuel on the fire. No matter which way you bet it's fuel.

I'm sorry, I don't believe you personally deserve it but for the betterment of the everyone we have no choice but to shame people like you (prediction market rationalizers) via downvotes and hot replies until you can see through the limits of your overly simple model of reality.

I don't think you're a bad or stupid person, but when it comes to your views on prediction markets, they are very bad and very stupid. Thankfully you are not your beliefs and you are free to change them at any time. I beg you to reconsider this one.

4.7 has been incredibly frustrating vs 4.6. Not sure what's going on but it keeps dropping stuff and getting stuck in weird side quests. Hope it gets fixed cause 4.6 was awesome.

I was thinking the exact same thing just now as I load up qwen3.6 into hermes agent and all while fantasizing that it will replace opus 4.7. It might not actually but seems like we're on the verge of that.

Lately I've been wondering too just how large these proprietary "ultra powerful frontier models" really are. It wouldn't shock me if the default models aren't actually just some kind of crazy MoE thing with only a very small number of active params but a huge pool of experts to draw from for world knowledge.

It only has 17b active params, it's a mixture of experts model. So probably a lot more people than you realize!

Elon Musk

Your gaslighting doesn't work on me

https://www.bbc.com/news/articles/c5yx8knpr5no

Buying a newspaper has always been a political act

DOGE posts had everything to do with technology and silicon valley

You're aging well

More should follow them. That website is a complete cesspool at this point and if you're not noticing it I worry about how it's gonna effect your psychological wellbeing later in life. The internet is bad enough as it is, but that site is at another degree of awful.

"Israel should extend its border with Lebanon up to the Litani River deep inside the country's south, Israel's finance minister said on Monday as Israeli troops bombed bridges and destroyed homes in an escalating military assault. The comments by Finance Minister Bezalel Smotrich were the most explicit yet by a senior Israeli official on seizing Lebanese territory in a fight Israel says targets Iran-backed Hezbollah militants."

Am I missing something?

hate? who's full of hate?

Please share more

They're utterly embarrassed it's just that they've been persistently encouraged via their amygdalas to project their own shame and insecurities onto others, as well as to swallow insane rationales as to why even though these people are evil it's a necessary bitter pill for the worldly government to swallow in order to bring in the eternal kingdom

I'd also be depressed if I were in your shoes account created 8 minutes ago

https://www.reuters.com/world/middle-east/israeli-minister-c...

https://www.timesofisrael.com/nearly-20-nations-slam-israels...

https://www.timesofisrael.com/expansion-of-west-bank-rule-ap...

> fact I can't run Opus locally

Yet

idk seems accurate from where I'm sitting

I think some kind of graph-capable model directly on the AST or a lower level IR would be the way to go, with bidirectionality so that changes propagate back up to the syntax without squandering LLM resources.

Any reason not to use it with the agent client protocol instead?

Zed is incredible, such a relief

It'd be hilarious if it was a fiction, some bitter comedic cautionary satire

Personally I don't think the ELIZA effect is the interesting part of this. For me it's how the incentives set this dynamic up right from the start, and how quickly they've been taken to the extreme.