- cross-posted to:
- linux@programming.dev
- linux@lemmy.ml
- cross-posted to:
- linux@programming.dev
- linux@lemmy.ml
You must log in or register to comment.
So what’s a good alternative? LXDE? lxqt? mate? I feel so disappointed
🤢
So yes, I use Claude quite a bit while developing xfwl4. I don’t vibe-code
Those two sentences are mutually exclusive.
I’m sad to hear that I might have to drop XFCE. Or, at least, stay in the X11 version of it.
Vibe coding would be having the LLM generate most of the code for him.
In this case, he’s not doing that. This is mostly him using the LLM in ways to write up reports on xfwm4, scan for issues, and help with pseudo-code.
It’s easily saved me months of time, not in writing code (for non-trivial things, I can usually write it just as fast or faster), but in tracking down problems and helping me figure out solutions, whether that’s in Claude reading and reasoning about my code to find something that’s wrong, or in adding tons of targeted debugging prints to gather the information we need, and then sifting through the log file for me to do analysis.
I do believe he is in his right to explore new technologies.
If you have been trusting these people all these years, there is not reasons to stop to trust them because the use AI/LLM.
These guys already demonstrate what they know and what are capable of, I do trust them.
The aid of Claude can help to get his achievement faster than ever. This is the truth, and for a project with limited manpower, like Xfce4, I think that is clever leverage this technology.
@freezr @lambalicious @maggotbrain people have outsourced their thinking to megacorporations.
If you have been trusting these people all these years, there is not reasons to stop to trust them because the use AI/LLM.
Trust is not an inherent attribute of a system. It’s a resource, and it can be spent – as well as misspent. This is an example of them choosing to spend that trust, thus giving me the reason.
Delving into AI means cashing out on all the trust they had gotten from me. From this point onwards, at least in what respects to me, they’re starting from scratch… technically they are starting with negative points, but that’s besides the point; the trust is spent.
Research “thermocline of trust”.
@lambalicious @freezr
I might get you wrong, but if the developer would not have disclosed, what kind of assistance systems he is emplyoing, you would still be happy with XFCE4?
Developers also employ libraries for their work, which they usually have no control over how those are geneated.
Like it or not, AI tools are here to stay.
AI systems are heavily employed in any kind of development, especially OS’es.
Knowing that now, what OS will you use from now on?
Is it OK to correct bugs discovered by AI tools or rather not?I understand the topic is divisive, for me is not a blocking point anymore. These tools are evolved and many people are trying to use them the right way.
There is no “right way” to use AI, only “less bad enough for you to shut up your conscience”. It was designed to undervalue humans and give wealth privileged access to skill, built with intellectual and wage theft, and is still powered by the death of the planet and those, sapient or not, who live in it.
If the developer would not have disclosed, I would not have known, so whether I’m happy or not is not relevant. We can’t and don’t (currently) trace the bill of materials of everything we use at home, either. I don’t know how many previous owners does my toaster have.
Like it or not, AI tools are here to stay
Oh look it’s the accelerationism thought-terminating cliché!
AI usage is viral. AI is only “here to stay” in that because the weight of the “”“content”“” generated by people who use it, it forces other people to turn to AI itself to deal with the sheer amount of the sloppage. But we don’t have to accept that, and the clearer our posture regarding that, the better. We can commit to process things at human speeds, on human schedules. After all, in Open Source and in Free Software we don’t owe anyone instantaneousness.
@freezr @lambalicious @maggotbrain
The more you use a tool the less you double check its results, you become complacent with what it does, this is fine for actual tools that produce deterministic results, the slop machine is the opposite of that.@konato @lambalicious @maggotbrain
This is a bias, at certain point there is a trade off, and you must deal with compromises. This applies to almost everything.
@lambalicious @maggotbrain so that’s why he chose Rust?
