"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.
Dev:
Quite honestly, if that's the level of bugs we're dealing with due to our use of AI tools, that's a pretty good deal.
Response:
Out of the 4 commits I've looked at, I found 2 bugs (potentially). That doesn't worry you? Do half of your commits normally have difficult to spot bugs?
At least the comedic value of this project is pretty high.
The good ol "guns don't kill people, people kill people!" argument
Cool.
"Don't worry about the code being shit, it was already shit BEFORE the AI."
So, anyone know any Lutris alternatives?
Heroic seems like to be better imo. Bottles always bugging out with some bullshit.
For games yes, for in general applications no
I switched to Heroic because of the integration of so many storefronts, but if you are looking for something lightweight, i took a liking to Faugus Launcher
Faugus has been working well for me as well
using wine directly - while sometimes a bit complicated - can work quite well with many not-brand-new games.
I've had a grand success with 99% of games with steam. Lutris is the only thing that can reliably run Battlenet. But that's good honestly, the more of obstacles between me and wow the better
I've never had a problem with battlenet in steam, just add the installer as a non steam game, run under proton, change the target and environment to the new installed location (don't forget the " marks around the target), and play.
Lutris dev keeps pushing me away from Lutris day by day.
My god, I only read a few comments and it's already one of the most toxic threads I've read in a good while. 🤮
I honestly haven't used Lutris this year or last year. I just keep Proton up-to-date with Protonup-qt and launch stuff through Steam, everything works. Gaming on Linux has come such a long way 🥹
Seconded. ProtonUp-QT and GE-Proton are the tits.
I don't like the direction this is taking but I have Lutris and some old games work with Lutris out of the box while they don't work at all with Heroic. As long as it gets the job done, I see no reason to jump ship for those old games. I do use Heroic for newer titles.
Yeah and the dev didn't say he's switching over to a 100% vibe coded model, just that he's ok with it as long as the bugs are manageable.
More specifically, what he's saying is that in his experience so far the AI generated code is just as buggy as human generated code. Which seems perfectly reasonable to me. The point is that all code needs to be carefully reviewed.
The problem with vibe coding is that people who know nothing about coding get an AI to output code that they can't even read, let alone debug. Then they throw it straight into production. Using AI to quickly output blocks of code for software that is being designed, assembled and reviewed by an experienced programmer, using proper review and testing processes, is a very different beast.
Obviously if you regard any use of LLMs as immoral then it doesn't matter. But if that's why people are unhappy about this then they need to say so. If their concern is actually with the results, not some broader immorality of the process, then the dev is absolutely right; they need to actually look at the results.
Yes thank you elaborating. I do this this situation is different than (for example) what was going on with Booklore.
I am not a programmer but isn't the problem with LLM code, that even its errors usually look more plausible and correct than many human coding mistakes? So, it is more work and harder to spot coding errors. Not sure how much there is to it but given that LLMs show that behaviour also with other tasks I would not be surprised if it also translates to coding.
Sure, but the thing is that bad human code also looks plausible and correct if you're not taking the time to carefully analyze it. Bad code can be something as small as a missed comma. It can be writing the correct statements, but putting them in the wrong order. It can be an incorrect indent. It can be 100% correct code that doesn't work because your project is using an older or newer version of a library.
The problem, a lot of the time, with LLMs is that they introduce the necessity of a review step - the ubiquitous "Always double check the output" - that is so time consuming, or so thoroughly invalidates the need for the original output, that you might as well just skip the LLM and go straight to the double checking stage. If you're asking an LLM for information about Brazilian visa policies, but you can't trust that information unless you check against the Brazilian government website, then you should just check the website and not bother asking the LLM.
But with coding, the review stage is already baked in. All code, human or machine, requires careful review. And all bad code can look like good code if you don't know what you're looking for (and a lot of the time it looks like good code even if you do know what you're looking for. That's why a second set of eyes is so important). So as long as the LLM isn't producing significantly more issues than a human coder would, there's no real downside.
There are still dangers to be aware of, of course. But it's a very different scenario from, say, dispensing medical advice.
There's also the question of how LLMs are used in a project. There's a big difference between firing up Claude and saying "Write a program that will make Windows games run on Linux" vs saying "Write a function that checks if an instance of BattleNet is already running." And both the scope of your prompt and the completeness matter. If you are an experienced coder you will know what information you should supply to the LLM to correctly construct the output. If you don't, it'll just fill in the blanks.
People with more substantial coding knowledge have the ability to be more specific about both what they want and how they want it done, so they will get much more consistent results back. And, of course, they have the skills needed to identify bad results for themselves, as long as they are taking the time to do so.
Lutris still is the only launcher i can get battlenet to run on. Steam, Bottles. Just hours of fighting agent has fallen asleep error code. Lutris? Didnt even have to tweak any settings. Battlenet just ran out of the box in the Lutris.
Faugus launcher handled battle.net no problem
Came here to post this. Faugus works the first time every time with battlenet for me. Lutris always took a few attempts.
Tbh if it's working it's working, just leave it
have you tried Heroic? I had more success with it launching some games (SIGNALIS, mewgenics)
I have heroic for gog and epic and it works fine. Havent tried Battlenet in heroic but Lutris just works so Im fine with it.
Can't imagine a more dishonest interpretation of the linked block than this post's title
Maybe you should, you know…read?
Thanks for the bug reports!! Quite honestly, if that's the level of bugs we're dealing with due to our use of AI tools, that's a pretty good deal. I've seen much worse, so much worse in code that we actually shipped in releases. And no AI was used to create those critical bugs. (Emphasis mine)