This is super helpful and precisely what I was looking for. I’ll study if some of it will fit in my workflow / needs but overall I think it is a great way of managing and building packages with the tools available.
Just as you, I’m too used to inheritance. I’m trying to rely more and more in composition.
Most of my personal tools don’t have interdependencies and are isolated, so that simplifies things a lot. I think i’ll try to isolate them as much as possible and copy paste or adapt the code that couples them for each project.
Thanks a lot again. This was really clarifying and super detailed
why you call every possible solution LLM? like, ok, you don’t like ai, but not everything is ai, guess what - people can also make mistakes and just want to help…
Saying “I just wanted to help” is not an appropriate response to being corrected on a bad suggestion, aside from the fact that the assumption that you (who regularly use LLMs) didn’t make that mistake yourself but that it was made by an LLM there’s nothing wrong here IMO
Just accept that you had a bad idea and move on, and if people assume you make bad mistakes because you rely on LLMs that might be a sign…
But most importantly: Try to unlearn that reflex to say “I just wanted to help”, it’s not useful and it just dismisses the feedback, obviously the reason to reply was to help, why else do it? But intention of being helpful doesn’t remove being wrong. Just try to learn from it instead and move on
@NOTIdealDev I apologize. My words were harsher than necessary, and I should not have assumed you used an LLM to get the answer. That was unfair of me.
what is the idea is use LLMs? and that i will just use them to help? what’s the reason to use AI to help others?
without “I just wanted to help” i guess there wouldn’t be much people who will help, cuz every mistake people make will be bad, it’s a suggestion
also with all respect - i understand you all don’t like ai as i’ve said before, but using it as a “root of all evil” argument is just boring at this point
i just wanted to say - people make mistakes, you don’t know if you’re right, and that’s normal, without mistakes there isn’t learning
That is completely ignoring everything I said, please read again, and stop reading things into what I’m saying and accusing me (again) of being just biased, I said nothing about LLMs, good or bad
I don’t want to engage in an anti ai argument so I’ll abstain to comment on the rest of your post but as for using llms to answer forum questions…
I go to a forum precisely looking for human advice and knowledge based on experience and that can be debated. Copy pasting llm answers or even basing advice on them defeats the purpose of a forum in my opinion. If I wanted an llm answer i would have used ai myself (and trust me, I won’t)
That’s what i said - there’s no reason to copy paste LLm to forum, so why would i need to use it? i say that people also make mistakes, and thus saying someone’s answer is bad because it’s not most efficient, and saying it’s all AI is just rude at this point
FYI, saying something intentionally offensive and prefacing it by saying “with respect” does not make it unoffensive. It just makes you look disingenuous.
Learning to see things from other people’s perspective gets you a lot farther than calling the validity of their perspective into question does. For example, when you told me that you did not use an LLM for your answer, I apologized for assuming so. I trusted that you were telling the truth and were trying to be helpful. I did not argue with you about why I perceived it as such to begin with.
As someone who uses LLMs professionally, and knows how to build them, I’d like to tell you that watching you post about them is painful. You do not understand how they work, what damage they cause, or how to use them responsibly. You have no idea about when they are appropriate to use, and when they are not. My perspective is different from yours because I was learning how to build them before you and the rest of the world didn’t know they existed. (I have also been forced to use them against my better judgment in places where I would prefer not to.)
I find what you have to say about LLMs extremely uninformed. I’m not sure if that is because you do not care to inform yourself or are incapable of doing so. But if you wish to be taken seriously, I suggest you do a lot of research.
I also think @wchc should probably move this whole discussion somewhere else as it has nothing to do with the original question.
I won’t be moving this discussion, as it’ll lose the context, but if you want to continue, please move over to DMs or one of the already existing AI topics.
Any further replies here that don’t touch on the original subject I’ll be removing as off-topic.