Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Did Claude increase bugs in rsync?

alexispurslane.github.io

Post reply on HN
Print view View original

601–610 of 611 posts

Re: Did Claude increase bugs in rsync?

#601

Post by pushcx »

logicprog wrote: post #99
pushcx wrote: post #64

What followed was extraordinary: 329 comments and counting, ranging from thoughtful concern to outright harassment. The thread did not stop at words. One user posted My Little Pony drawings of themselves strangling the "project janitor that pushed vibecoded commits": It spread to Hacker News and Lobsters, generating hundreds more comments. This is false, it did not appear on Lobsters. Here is the function in the code…

I have done so! that was a misremembering on my part. first mention of Lobsters is now here: > On Lobste.rs, in response to the Medium essay Tridge himself posted in response, finally some users like boramalper begin to actually ask for evidence one way or another:

Thanks, I appreciate you sorting out the timeline on such a heated issue.

Re: Did Claude increase bugs in rsync?

#603

Post by JetSetIlly »

Barrin92 wrote: post #427
JetSetIlly wrote: post #422

Earlier quoted context omitted.

> This is not okay. Agreed. The way to address it though, is through calm analysis and reason. The emotional language from both groups is not helping. If there's one problem with Claude et al, it's that it's all happened way too quickly for people to keep up. We're all at different stages of acceptance and I think that's what we're seeing manifest in the various discussions.

>We're all at different stages of acceptance I do hope you see the irony of accusing people of armchair psychology and then hitting us with the five stages of grief. I trust rsync (which handles critical data on my system) because I know a veteran of 40 years wrote the code it runs. If I see code like the one above posted by the OP, that the author wouldn't have written, I start to pay attention. When I then read the…

> The problem isn't this weird gaslighting attempt that we just haven't let Claude in our hearts and souls yet which you seem to have determined is inevitable (spoiler alert, it is not),

I don't believe it's inevitable and in fact, I'm thoroughly against the use of tools like Claude.

My reference to "different stages of acceptance" was only to indicate that people have embraced these things to varying degrees, and it that it seems to be this difference which is causing conflict in discussions like this. (I doubt I will ever fully accept it. A lot needs to change for that to happen).

I didn't really have the "five stages of grief" in mind when I wrote it.

Re: Did Claude increase bugs in rsync?

#604

Post by Laurel1234 »

rickydroll wrote: post #596
Laurel1234 wrote: post #590

Earlier quoted context omitted.

It takes something far more hollow and soulless even than a clanker to type a thing like this.

Beep boop. It comes from recognizing that, yet again, humans are way too full of themselves and are not as unique as they think they are.

Sloppers can't even pretend they don't despise humanity, par for the course.

Re: Did Claude increase bugs in rsync?

#605

Post by cindyllm »

Laurel1234 wrote: post #604
rickydroll wrote: post #596

Earlier quoted context omitted.

Beep boop. It comes from recognizing that, yet again, humans are way too full of themselves and are not as unique as they think they are.

Sloppers can't even pretend they don't despise humanity, par for the course.

[dead]

Re: Did Claude increase bugs in rsync?

#609

Post by UebVar »

eschaton wrote: post #579
UebVar wrote: post #490

Earlier quoted context omitted.

No, you are wrong. In German and French (roman) legal systems this is a "Vermögensdelikt", and explicitly about material damage and gain. Yes, common law can be more broad (in canada it isn't really, it just also includes service, btw.), and yet it clearly does not meet the definition, as there is a damaged/defraued party and fraudulent/gaining party. We are not talking about somebody usurping somebody else reputatio…

No, you are wrong. You are either willfully misunderstanding what I’m calling fraud, or you are misinformed as to what “material gain” means in many legal systems. With respect to the former, “fraud” is a shorthand for “fraudulent misrepresentation,” which is what you’re doing when you take someone else’s IP and try to contribute it to a project without securing the right to do so. It can be read as implicit in the a…

You are misinformed, I suspect you have no idea what you are talking about.

As I said, I do not know all legal systems in the world. If there one where "material gain" matches your idea, please cite the law or a case that includes LLM usage. As I explained in the canadian law even includes services and yet it is so much very much not matching the defintion for reasons explained.

I do understand very well what you mean by "fraud", I do not miss represent it - your opinion on what it should be is plain and simple wrong. I explained why in my previous posts.

You are under the impression that legal science is some kind of folk etymology. It absolutely is not. Fraud is §263 StGB, Art. 313-1 Code penal or §380 of the canadian criminal code. (They all are remarkably similar, because they share a millennia old tradition. Making them IMHO fascinating cultural artifacts.) Here [0] is a structured version of on of these texts. Think of it as a symbolic execution of the law. You can see there is structural mismatch with your "case". Nobody ubsurbs anything from somebody else, and all three laws incude that in their defintion. That was my original claim.

You think you somehow can make up your own private definitions, develop your own private theories about them, apply them and argue about the semantics your made up terms. That is the opposite of how jurisprudence works. It rigorous, with well established scientific and scholastic methods. It operates on term defined by the law. In the case of "fraud" the previous citations, especially in criminal law, and nothing else. German legal science has its own theory what counts as "nothing else" under the name "Wortlautgrenze". These terms and methods vary from jurisdiction to jurisdiction, but by surprisingly little.

Dont call your code a decorator pattern, because you think it is decorative. Different pattern libraries have definitions for that and you need to be able to argue it fits. Like wise, if you feel something involves some kind of misrepresentation its probably not fraud. If things have different names, that probably for a good reasons, especially in legal science.

[0] https://www.iurastudent.de/schemata/schema-zum-betrug-263-i-...

Re: Did Claude increase bugs in rsync?

#610

Post by zzo38computer »

I agree that the bug report is not very good. I also agree that having commits written by Claude is not necessarily what caused the bug, although it might be (it is also possible that some of them introduced bugs and others didn't); whether or not it is in this case, is I don't know (some people think it is, but some think not). (Software without code written by generative AI will still have bugs too.)

However, the claim that "the original post was [...] no bug report" seems wrong; it does have a bug report, although not a very good one. It says that incremental backups using multiple --compare-dest arguments do not work, so it is a bug report. But, it should have been written differently, including by putting the text directly instead of a screenshot, giving a proper title, better details about the bug being reported, etc.

Their claims that they introduced deliberate bugs, are unlikely to be accurate, and not worth making those claims nor the violence that they involve.

I do have reasons for not wanting LLMs to commit code, so I agree with their opinion about that, but that does not justify making a bad bug report and the other stuff that they did. If it is FOSS, someone who disagrees with the project can fork it and make their own version, as has been done with other FOSS projects as well.

I think it is good that they are making statistical analysis. However, they used a language model to classify bug reports. They mentioned some things that might be missed, and they could be missed whether or not you are using a language model to classify bug reports, although there are some other possibilities e.g. whether or not a single report should count as multiple bugs in some cases, and mistakes in marking reports as duplicate.

Post reply on HN