I am curious about how the companies that claim their devs never wrote or reviewed a line of code in the last 6+ months react to this... they basically would have to just judge how the candidate writes a prompt or gathers the information to write it
I think that's a reasonable part of the assessment but the focus is primarily on the candidates understanding of the output and essentially their ability to QA the output.
I noticed a few weeks ago it started being very bad at explaining things (even things itself was doing) and started committing absurd errors (like reading a test of 5 lines and not noticing there was an explicit mock created in one of those, then saying that the test was failing while it was not)
I fear this is just the classic "nerf the model just before we release a new version of it"
This seems a worsening that seems to have started at Opus 4.8.
When I (was forced to) switched from 4.7 to 4.7, I noticed that 4.8 had a very strong bias against viewing tool output or reading files. It ALWAYS wanted to 'grep' things. So for example, it would run the test suite and grep for 'PASS', thus completely missing the 4 tests that FAILed. Its bias is so strong, that it does this despite explicit instructions to the contrary, and continues to do so even after having noticed its own mistake, in the same (claude code) session.
For editing files, it again seems to prefer writing small python scripts, rather than read the file and rewrite it (even small files, like notes). This again causes it to constantly miss duplications, inconsistencies, etc.
And from 4.8 to 5, it doubled down on this. At this point not moderately-long-horizon task is getting done consistently, due to its pigeon-hole view of the workdir.
The thing I like about console compared to pc (which I use the most for gaming) is the total absence of cheaters except maybe a super minority.
To make an example try to play a multiplayer famous fps (like battlefield6 or cod) on console and then on pc, it's not even the same experience. On pc it is basically full of aimbotters, wallhackers, nodamagers, etc every single damn game, on console you get the actual safe experience. Sadly this is one of the reasons why I play mostly only pve games on pc
How they enforce it in every other project with the same policy, if the reviewer/maintainer suspects the pr is ai slope he closes it. That's it, it works fantastically well, I do the same even in my job
These things are done not to block pr done with the help of ai, but to block the ai slops pr, aka those never reviewed, fully vibecoded, and with the submitter that didn't understand anything about the problem or the code, just trusted claude.
The rule is general on purpose to allow the maintainer to freely remove whatever is very evident is just ai slop
Any PR will get the same quality review. It's just that they now have a fast lane so they don't have to invest that much time to review a PR they won't fix properly and/or support.
exactly, I for example don't care usually about false positives, in the very uncommon case it happens the pr creator can discuss and prove he actually understood the problem, the codebase, the project policy and how to explain his solution actually could work.
How do they enforce that contributions aren’t copied and pasted from AGPLv3 or proprietary codebases? The honor system and intuition and occasionally flat out asking people.
Do you think sociopaths willing to lie about how they came up with a contribution are really that common?
Let me tell you a story: I use agents at work. Not to add more useless features, not in a let-loose way, not so that I can slack all day but to produce better output. Think performance optimizations, security fixes, better use of underlying frameworks, produce better documentation, get useful hints from the code base, that kind of thing. In a word, to produce objectively better software. that is my honor: giving the best I can.
Because AI is frowned upon where I work, I kept that under the radar. And I got nothing but praises for the good work, everybody loving it. Later I had the opportunity to port a piece of software from one language to another. LLMs are great for porting stuff when steered well. Again, nothing but praises for the amazing work done in a very short time. And this is where I tried to open a debate about AI by disclosing my tools and methods. And boom, suddenly I was evil. Praises gone. But still, none had the guts to throw the port to trash, because it's still very good. Hypocrisy.
So yes, call me sociopath if you like, but I will lie, produce better software and get the praises if I have to work in an environment that values tools and methods over results.
Don't compromise your professional integrity by lying about how you work. Rather find a job where you don't need to lie about your use of AI if you can.
25 years ago I had to lie about using free software since it was also frowned upon. I even lied to my CTO at the time who wanted me to work completely in a Microsoft environment, with ASP and what not. Being kinda stubborn I did not comply and used only free software. When boss saw the result and the cost I got promoted. My integrity is fine.
You should have been fired for cause, because the way you chose to work exposed your employer to substantial liability due to the “copyright laundering” nature of LLMs. But you prioritized feeling good over everything else. You’re a sociopath and you should be treated as one—a bad-faith actor only out for themselves, who cannot be trusted to collaborate within agreed-upon limits because you don’t think rules apply to you.
There are no agreed-upon limits, everything is new there. And I don't prioritize feeling good over everything else, otherwise I just wouldn't even start the debate. Far from being fired, I was promoted, because the real world is not your fantasy, results matter more than theorical threats. Also keep your name-calling to yourself.
You literally said as much, and I am not calling you names, I am accurately describing your behavior in a way you yourself appear to recognize:
> So yes, call me sociopath if you like, but I will lie,
> produce better software and get the praises if I
> have to work in an environment that values tools
> and methods over results.
You specifically said you would lie about how you work to get accolades if your employer attempted to set rules against working the way you prefer.
The appropriate reaction to your employer setting rules you don’t agree with is to try to convince them to change those rules, engage in a labor action alongside your coworkers to force them to change those rules, or to leave.
What you’re calling a “high horse” is having morals and ethics. Your employer said not to do what you’re doing, so you either convince them to change their mind without doing something they explicitly said they don’t want you to do, or you go elsewhere. If you take the route of being a filthy fucking liar you aren’t in a position to accuse anyone of being on a high horse.
easy to deploy, latest versions of the language are ok, has some of the best web frameworks to work with, absurd amount of experienced programmers can be found on the market
I never understood this. We build in Europe, over earthquake-risk zones, with bricks and steel and we follow rules to make them earthquake resistant. It is not a problem anymore since like the 1980. We now have also methods to make old and very old brick buildings earthquake resistant without demolishing them
It works fine for commercial buildings and multi-family structures here too , there’s even a ton of brick buildings in Oregon (which are currently being retrofitted), but not as well for single family homes because of the cost.
There’s a lot of historical context to understand here. The neighborhood that just burned down in the Eaton fire (Altadena), was built up by African Americans and Latinos who were redlined out of Pasadena even after desegregation. Some of them built their houses on land that they bought for under $100 in the 1950s and 60s. They wouldn’t have been able to afford the kind of construction they’d need to be both earthquake and fire resistant. Their choice was between owning an old tinderbox or renting from slumlords.
What? What earthquake zone in Europe is similar to the fault lines in California? We are talking about entire cities wiped out by earthquakes just 120 years ago.
There’s a plate boundary running under Morocco and across the Mediterranean, but it’s not nearly as active as the Pacific Rim, and it’s quite a long way from Northern Europe.
The main problem is they tried to be a middle way between a forum and a wiki with only correct answer (even if as some other have said, they never wanted to acknowledge that especially in programming the answer is correct or good or valid only for a certain amount of time most often than not)
What we actually need:
- the owner of the question must be the only one capable to say if an answer is good or not for his question
- no other user must be able to modify the question of the owner even if they think it is badly explained. They can always suggest modifies but the owner can decide if to accept them or not
- the moderators can try to suggest "hey we think this is a duplicate of that" but the owner must be the one to accept or reject it. Even if he says "I just want a more current answer, the one you linked is 4 yo" the moderators must accept his decision and keep the question open
- moderators must act aggressively against people (and other moderators) saying things like "this sound like some homework" or "you better use technology y instead of x" or similar
- answers older than a few years should have a visible flag on them signaling that probably that answer is not anymore the best possible one and to be careful using it and, if it doesn't work anymore, to open a new question tagging the old one
