w0m

> If that ain't a moat, then what exactly is a moat?

And Anthropic has still generally been eating their lunch at the top end.

i have active disagreements with teammates on the value of compaction/months-long sessions.

Same teamates also post 'Sol deleted my git repo!' or 'Sorry, ignore those 300 PR comments i was just looking!' ~once a month.

I take that more as the user is mixing business with pleasure. When i joined a company using `gcp` heavily, I didn't attach my personal gmail to it - I created a 'business' account and used that. Slightly inconvenient - agree, but it's on the individual to draw the distinctions.

I'd agree if it wasn't so much faster than ~all the other models. Maybe they nerfed reasoning to get that speed.

if the answer to 'the model is bad at X' is "you're over-reliant on it" - then yes, the model is bad at X in comparison to alternatives.

I'd feel guilty, but you're likely correct.

This surprises me a bit. I've found much better token usage with a proper/efficient MCP as even with a deep /skill defining usage, parsing MCP results is generally just better/more efficient than parsing CLI results. I say this having written a CLI tool explicitly for harness usage, and leveraging MCPs for the same.

I'm sure there are bad MCPs and great CLI tools that parse poorly/well via harness, but I'd be curious on an better research study.

someone needs to do this for the arm Surface Pro 11.

> Are you sure they even will want it back?

75 y/o MIL is still fighting my wife to take her middle/highschool 'stuff' back to this day. She won't throw it out though.

> He's looking at how it used to be done

I mean, you're probably thinking 30 years ago; not any time recent. Indy devs are still a thing though.

HEDDphone D1 is my current happy place without going into the thousands. Compare it to an iBud on a spotify lossless track and it's kind of shocking.

'ai slop' is hard for me to complain about when it's better than 95% of code prior. The issue is more people trusting it too much to go too far, and you end up with lots of extraneous bullcrap you don't need. Doing too much and too big of a change is still a problem in the agentic world.

the ~only reason I don't have a macbook in active rotation currently is feeling burnt by the terminal experience on my last go-round. I admit to not having recent experience (~5-7 years) - maybe it's gotten better?

> His sudden flip to being anti-Mac after years of being pro-Mac follows the pattern

I mean, I used Mac on ~every machine i touched for about a decade, got annoyed and haven't touched one in ~5 years now. It's not unheard of.

the tag line for the distro on it's home page.

> Beautiful, Fun & Agentic Linux

'agentic' linux

This is a solid summary of Apple to me. Generally great hardware (outside of when they fuck up the keyboard or for ~5 years or something), and you have to piece general usability in from a myriad a third party apps while hoping Apple doesn't generically decide to lock out whatever API they're relying on unceremoniously.

I ran my ~2011 i7 MBA for 12 years as primary laptop before the battery finally killed it (had to run linux on it due to apple killing support the last few years).

Great hardware. I don't miss OSX though.

If a coworker sends me a copy/paste from a chatbot - I can't bring myself to read it. I know it probably* has something useful; but it's just a visceral reaction.

> Why do you feel that creating the LLM from scratch is sufficient to trust it

If you're doing the training yourself, you at least have a verifiable supply chain and an audit trail. Today, we have no idea if a black-box model handed to us is coded to recognize specific domains or patterns and back door an application in a sneakily targeted way.

Black box is a black box. Many open-weight models clearly haven't been trained or created in the manner their creators claim---which raises the obvious question: if they lied about the recipe, what else did they lie about? Putting those models in a production capacity scares the living crap out of me.

That said, building from scratch isn't about achieving mathematical perfection or manually auditing 15 trillion tokens---that's impossible. It's about eliminating third-party supply chain risk and having actual governance over the pipeline.

Of course, that doesn't mean we can magically guarantee gradient descent won't produce weird emergent behaviors, or that we can blindly trust OpenAI not to backdoor things. But at least with the latter, you're making a calculated operational decision rather than blindly trusting an opaque black box of entirely unknown provenance.

> I don’t trust hosted LLMs for anything that needs to be private

I'd update this to

'I don’t LLMs for anything that needs to be private'

What's to prevent the LLM from sliding a heavily obfuscated binary blob into the application that does nefarious things? If you aren't creating the LLM itself from scratch, I don't feel it can be trusted.

To be fair - I'd much rather the name be overwrought but clear (once you know or have the decryption key to translate) than to have to look up an even uglier chart that i have ~no chance of memorizing to go 'ok, the Tiberian has X, Y, Z and Q and the Cyrian has Q, Y, but Z is nerfed'

i mean; that sounds like a win for GH right?

> but everyone sees the writing on the wall

i'm... not sure? This assumes ~stagnation in task-possibility. We've had ~exponential progress for like 3+ years now; I'd have never dreamed the tooling I hammer daily would exist in my lifetime just.. 3? years ago. And it's improving daily.

Maybe Open will win, maybe Closed will keep pushing the envelope. The world here is raw enough i don't think anyone can make any significant claim other than 'holy shit this is useful and moving Fast'.

and honestly that's my entire point. There is no Good Guy here.

> and were not called the attackers.

This is the mental mental leaps I'm struggling with here. Did you not live through that era where they were explicitly and repeatedly called out as 'attacks'? They were generally tolerated/hardenee around as they provided value-in-discoverability.

i mean, i got 5 replies in a minute of asking, and none deny it's an "attack", they simply say "good". HN should be better discourse.

> to give me a cheaper service that fits my use case?

Because they aren't giving you a cheaper service that fits your use case.

Best Case scenario, it's a trillion-dollar behemoth stealing from a billion-dollar behemoth so they can add their own explicit restrictions/weights on top to influence the masses.

There is no 'robin hood' here, any perceived value you get is clearly and explicitly tainted. "I don't care if it doesn't show me non-party-line results - It makes me a cheap UI !". Ethics/morals be damned.

>We are just very pleased with it regardless

You don't trust the multi-billion dollar behemoth, but you trust the militarized multi-trillion dollar behemoth to play 'robin hood'?

i can't get my brain around the mental loops here.

> Distillation is NOT an attack.

From the article -

> 28.8 million exchanges with Claude through almost 25,000 fraudulent accounts

wouldn't that be considered an attack? Not sure what I'm missing here.

> are you seeing large token usage by GLM 5.2

the statement isn't "GLM 5.2 has large token usage", it's "GLM 5.2 has large token usage vs modern Opus".

I haven't used it, but this wouldn't surprise me. I see ~30% lower token usage for better results with Opus 4.8 vs 4.6 (and i had great results with 4.6)