Hacker Newsnew | past | comments | ask | show | jobs | submit | jonfw's commentslogin

One token of GPT seems to go farther than an opus token in my experience, those books are a little cooked

If you're interested in keyboard navigation of websites, consider a browser or extension with link hinting support! It worked really well in my experience a few years ago, although I've since became much more of a mouse guy and stopped using it.

Qutebrowser was my favorite browser for keyboard navigation but firefox, chrome, etc. have extensions for this as well.


Vimium extension does that. Works well too. Works on Chrome and Firefox.

try out surfingkeys if vimmium isnt ur cup of tea

I am currently trying something called ShortCat, this is not just for the browser but works in other Mac applications too!

Look Ma, No mouse !


Link hinting - love it

We are discussing a gas tax, and there is a strong correlation between gas consumption and weight, which implies more taxes for trucks and SUVs


That's true, but gas consumption by weight is more of a linear function, while road wear follows the fourth power law by axle weight.

See: https://en.wikipedia.org/wiki/Fourth_power_law

As an example:

A 2026 Honda Accord LX has a combined gas mileage around 32 mpg and a curb weight of 3,239 lbs.

See: https://automobiles.honda.com/accord-sedan/specs-features-tr...

A 2025 Ford F-150 XLT has a combined gas mileage around 20 mpg and a curb weight of 4,941 lbs.

See: https://www.edmunds.com/ford/f-150/2025/features-specs/

Keeping things simple and calculating the axle weight to the fourth powers of both vehicles, the F-150 causes 5.4x the road wear of the Honda Accord while using only 1.6x the gas.

The reason this doesn't matter so much, though, is that the types of trucks used for shipping goods, when loaded, cause on the order of 10^4 the road wear, dwarfing any differences between standard commuter vehicles, which is why commercial trucks have to stop at weigh stations.


The big trucks also have a lot more tires / tire surface area, to mitigate that. IIUC, the weigh stations are to ensure they aren’t overloading the truck, so that road wear is comparable instead of being that vastly greater


I think that this has the benefit of both raising a non trivial amount of money as well as reducing housing demand from the wealthy in a supply constrained market


I used to work with a guy who would always say "if you're looking for trouble, you are going to find it"

When I hear that "we found X bugs using some new tool", where the standard for bugs is low and doesn't neccessarily require user impact in realistic scenarios, I think to myself- duh! You went looking for bugs, of course you found them.

For a sufficiently complicated product, in my experience, you don't have to look far.


Sure, but the bugs were found in an automated process. They just let an LLM scan. That's very impressive finding 100s of needed code changes. And it's even better if those needed code changes are bugs / vulnerabilities. The part no one is talking about comes from the bill. I'm sure Anthropic let Mythos analyze possibly for US$10,000s in tokens. A similar phenomenon happened back when an LLM scored well on some math olympiad competition. Yeah, it got all the answers right, but it was a frontier model running for 8 hours straight. That'll hurt the budget quite a bit. We're likely not at a stage where big corporate systems can just throw Mythos at it willy nilly for a complete analysis unless they have a ton of money.


Well it helps if 'looking for bugs' doesn't cost $300 per hour per set of eyes.


how much does it cost? my understanding of Mythos is that it runs a lot to find issues


The things I’ve read from various open source orgs with access to it is that Anthropic is giving them unmetered access for now as part of Glasswing. I’d bet that the corporate partners have to pay though.


> if you're looking for trouble, you are going to find it

That's the "'No Way to Prevent This,' Says Only Nation Where This Regularly Happens" of unsafe languages.

There are huge swathes of problems we know how to categorically prevent, but some people won't do it because they're more comfortable believing it was never preventable than accepting any culpability for not preventing it previously.


Humans can certainly be self improving, both on an individual basis and in aggregate.

In humans, it seems that improvement in a new domain seems to follow a logarithmic scale.

Why wouldn’t this be the same for an AI?


> Could be. But 99% of the repos are static garbage with no PR nor actions.

But the 1% of repos that do have PRs and actions are likely going to be seeing enormous increases in volumes

I have been a part of two very large companies with self hosted gits and I've seen enough to be confident that this is an incredibly hard thing to manage


Ya but they are owned by freaking microsoft and have billions of dollars and employees to throw at the problem. The outage problems shouldn't be happening period.


Easy to say that! Some problems are legitimately hard to solve though. Github is likely seeing usage patterns that have never been seen before and I bet some of these failure modes are novel

If you are at the limits of your architecture you may need to re-write things, and if you are rewriting things you can not arbitrarily speed that up by throwing dollars at it.


That's entirely a predicament of Microsoft's own making, though. Don't forget that they're the ones who launched "AI" programming into the hype cycle to begin with. So it's entirely reasonable to hold them as a company responsible for the resulting outages, which indeed shouldn't be happening. Dogs, fleas, and so on.


It is not like MS is involved with AI and say they can make anything in minutes with AI too


> If I can use an open source highly effective LLM locally, and have it do all of the things ChatGPT can do

And if my grandmother had wheels, she'd be a bicycle!

We are miles off from open models having parity with what the AI labs are putting out.

And that's putting aside the compute requirements to run the top open models and the poor economics of running these models locally.


Not all prompts require the same compute, and Gemma-4B runs on our phones with parity output for ordinary 1-5 sentence queries. The common use case of Google-style queries is already solved locally, saying we're miles off is ridiculous.


It's not "miles off", the gaps are narrowing instead of widening and they are already more than enough for the majority of use cases.

Getting people to install a different chatgpt app on their phones is a lot easier than getting them to change their search engine.


With a good CLI, an agent may be able to do something outside of the scope of it's skill fairly easily, by running help commands or similar. With even a well written API it is not as easy.

I suppose that curl + API docs could replace a CLI but that's really token inefficient


I dislike that the conversation about wealth is increasingly tied to comparisons with the wealth of others.

Just because I am not likely to join the class of billionaires, does not mean I can not become wealthy! I prefer to measure wealth in terms of quality of life rather than by comparison to others.

I do not have to join the capital class to produce a great life for my family and live in stability and happiness


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: