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

Which is quite the contrast to Mercedes new axial flux electric motor, which goes all in on rare earths- the design relies on the highest end high-grade permanent magnets.

Still, presumably Mercedes ambitions are for few motors than BMW or Renault.


Vastly different target market and/or features there. Mercedes are chasing maximum power density, minimum weight for high performance deployments, with seemingly little concern for cost or supply chain.

Renault is going after the consumer market with these motors, where minimising cost and maximising availability is more important than pushing past 95% efficiency or cramming a 700kW power output in a motor that is small and light enough to fit inside of a wheel hub.


I'm a Terence Tao fan, but yes, OpenAI should at the very least just be telling him to go crazy with the latest models on our dime.

I liked this article because it helped me understand the javacript engine flow better than I had.

The rust flow is so much more natural to me.


I moved to the UK in 2016.

The public sector, simple, no frills, accessible, no flashy graphics, websites were a massive eye-opener.

They just worked. They had a job. They did it. I wasn't going to buy more from them because of it, and they didn't care. It was great.

I've heard that recently they've dismantled the centralised team that wrote all the rules, enforced it, and started moving to decentralised hosting, but so far the whole still seems to hold to together really well. I think, I hope, they have embedded the expectation that the local council, the tax office, your visa status, etc, should just be utilitarian in nature, and work for everyone.

I worry how long it will last...


Damnit. No WFH option.

Unless Stonehenge is your home

“Work From Henge”

> 3. AI Companies will be profitable

but many of the current crop will never return money to investors.

I largely agree with you, but the huge investments currently being made will be very hard to get a return on. Token costs will come down, performance will go up, and you want to be in the business of selling the picks & shovels, not doing the mining.

Which is of course why nvidia, google & TSMC are in pretty good positions, but even their valuations have some bubble in them.


Respectfully, do you want a bet that AI companies like OpenAI and Anthropic can't become profitable?

I mean this is a sort of conspiracy theory and I genuinely don't know why people think AI is particularly hard to get money back from?

> I largely agree with you, but the huge investments currently being made will be very hard to get a return on.

Why do you find it huge? Anthropic went from $1B to $44B revenue in a few months and this is unprecedented.

1. The margins on inference are huge

2. There is genuine moat because AI models have personalities strengths and weaknesses that's so they are definitely not fungible

I think a lot of handwaving goes on but it comes in the form of some latent concern that AI might just be profitable. But the reality is that it will be.

None of the "selling picks and shovels" analogies will stick.


wtf is this comment section?

The author of these commits were tridge & claude.

What does tridge have to do to convince the open source community that he might be a legit programmer & have a clue?

Samba? Whats that? Rsync? Never heard of it. Tivo? No clue (maybe more Australian context here than others, but still).

Even the comments on the github issue, are totally devoid of the context that this is a very senior open source contributer who has maintained this project since he came up with the diff algorithm during his Phd, started the project and now chooses to acknowledge that he's using claude.

Is there any evidence that the bug rate on rsync is any worse than it used to be? or just a screenshot from mastadon?

It is just so bizarre to me.


I'm not sure about tridge personally, but I've regularly seen real competent engineers introduce obvious hallucinations when using coding agents. Review fatigue is real, and you just cannot own the code you didn't write to the same degree as the one you wrote

My consistent observation is that seniors who do only or mostly code reviews for several months end up making worst and worst code review. They nitpick thinks that dont matter more and more ... and miss big architectural issues, maintennability issues and bugs more amd more.

There is no reason to think reviewing AI code more then writing own wont have the same effect.


> this is a very senior open source contributer who has maintained this project since he came up with the diff algorithm during his Phd

People change. You can be Linus Torvalds for all I care, if one day you wake up and start pushing 9000 line commits created by LLM and with regressions, you're not that person anymore.


Also, a real example is Steve Yegge.

I still have no idea what on earth why or is Gas Town. Also, he facilitated a crypto rug pull around it because he claimed the money to help support Gas Town was too good to pass up. People have lost their minds with this.

This is just a tool and it’s making people have brain damage. I don’t want this reality. It’s too stupid.


But DID anything change?

Of course I know that some people can just becoming psychotic out of nowhere. But why would I assume it?


According to the thread rsync broke for incremental backups and increases the cpu load heavily. The whole thread only started because people noticed regressions and were wondering what happened.

Since I quite a few users are using distros that won't update for a while it gets even better: this trend may continue and as soon as the update actually happens we'll be so far down the road that it will be too late to take a step back and reconsider due to the delayed feedback. This is pretty much about the few people _already_ having issues with it.

That being said, if the creator wants to use AI to work on the project they are free to do so. I just hope nothing of value is lost because of it.

P.S.: If you stop writing by hand and start delegating - to AI or other people - something has changed. There shouldn't be any discussion about it. Delegation is different than writing it yourself.


Yes, according to the git diff and the comment here https://github.com/RsyncProject/rsync/issues/929#issuecommen....

A change in the sys calls that are used. That's pretty sensitive in general I think; I can see if it were introduced by an LLM why people would be upset if they experienced data loss from it.


Do you think LLM use is evidence of psychosis? I think it's a widespread problem but probably not psychosis.

At least once a week we have a post on HN about CEOs having AI paychosis. Maybe there is something to it.

I like the term paychosis

> Is there any evidence that the bug rate on rsync is any worse than it used to be? or just a screenshot from mastadon?

There's plenty of evidence that rsync 3.4.3 has broken a bunch of features like incremental copies, yes.

Which is why your post is a great proof of how AI derangement can make previously great engineers output broken dangerous slop.


Has rsync never broken features before, without AI?

Have features been broken in patch versions?

I don't remember of any examples, no. I'm sure you can use your clanker to answer that though and also plot out the number of occurences.

[flagged]


Those posts may not have been visible to everyone. The posts you're referencing are hidden for me behind a link "33 Remaining Items (load more)". Without the update, I didn't know to go look for them.

And honestly I noped out of scanning the entire comment thread by about #5 or #6... I could tell there was nothing productive in the remainder of the comments.


Actually it is, someone compiled allll the actual bug reports tracing back to AI:

https://github.com/RsyncProject/rsync/issues/929#issuecommen...


"antisemitism" loses it's meaning if you are using it this way.

Literally the only thing the guy posted was a copy-pasted "Israel" location. What other possible interpretation is there than "this guy is from Israel, so you can't trust him"?

So? Still not antisemitism.

> Then somebody screenshots the geographical location "Israel" to attack another commenter. He gets lots of upvotes for it, too.

And you got downvoted for calling out that crap. A sad state this world is in.


Yep.

When someone does that, he gets rightfully called out.

On the other side, accusations of being Russian trol are pretty common, even here on HN.

Why are people more sensitive to antisemitism than to antislavism?

Double standards, or just a hate induced by decades / centuries of indoctrination?


> Why are people more sensitive to antisemitism than to antislavism?

Calling someone a vatnik or Russian troll is mostly because the statement that provokes such a callout reproduces Russian propaganda talking points, and Russia has been running propaganda campaigns for well over a decade now. Similarly, ordinary Russians aren't called orcs, but Russian soldiers are called that because of their despicable behavior in the war theatre.


Replace Russia with Israel and everything you said also applies.

That's not an answer to the question you quoted.


Nobody was talking about Israel or Israeli policies towards Gaza, and there's no evidence that anyone in the thread is anything to do with the Israeli military.

[flagged]


Your opinion reminds me of the narrative about cheaters in the speedrunning community. The cheaters say they cheat not because they feel superior, but because they feel “they could achieve good results if they put in the time”. They feel entitled to cheating.

Thanks very much for this- I had not seen it.

It does not paint a pretty picture, and I did not know this context.

Perhap the tridge I knew is also of the past, but I hope not.


I have quite an early memory of being on the somewhat remote property in Australia that my mother grew up on (Central NSW, near Condobolin).

My Uncle, who then ran the property, walked over to a rock, whacks it with a hammer or similar, shows me a bit of a trilobite (which are totally different to our sort of bytes). He did this a bunch of times. I still have the rocks. No amazing full horizontal cross-sections, but it certainly got my very young mind excited.

There were fossils RIGHT THERE from before there were dinosaurs!

Oh, and that central Australia used to be an Ocean!

These clear demos to young kids, or adults, are great, and the many other examples here in the comments are a testament to that (Vienna? wtf!).


and in the meantime, just a sweep of the committed code (or the to-be-committed code for lots of us) and the code it interacts with, is increasingly catching lots of problems.


It's just such an old old business school idea. Metrics matter.

The classic example is the Second Fleet of convicts that were sent to Australia - they were paid per convict that boarded the ships. 40% of them died. Unsurprisingly perhaps, slave traders won that contract.

They started paying per live convict landed in Australia after that.

A COO at a ~$150B firm like Uber certainly should know what the outcome will be before the leaderboard goes up. Really makes you wonder if the board was incentivising him to increase AI usage; job done, bonus unlocked.


Or maybe it’s just what it looks like: A good old boys network of incompetent inbreds whose only qualification is that their parents could buy them memberships in the right clubs.


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

Search: