4 ms·
One nice thing about ChatGPT, they won't judge you as a fool for asking an easier question.
by Salgat 2y ago
One nice thing about ChatGPT, they won't judge you as a fool for asking an easier question.
- ryandrake 2y agoIt also doesn’t answer your question as if you asked a different one. So many stackoverflow conversations: Q: I am looking for a way to do X in language Y. A: You shouldn’t need to do that. What do you really want to do? Argh. What I really want to do is exactly what I asked. If you don’t know, fine, but don’t try to steer me towards a question you do know the answer to. Just let someone else answer. Not everything is an XY problem[1]. 1: https://en.m.wikipedia.org/wiki/XY_problem https://en.m.wikipedia.org/wiki/XY_problem
- emptiestplace 2y ago> Not everything is an XY problem Pretty close, though.
- michaelcampbell 2y agoSure, but I find that this phenomenon is just different people thinking at different levels on the abstraction continuum.
- emptiestplace 2y agoI had to think about this for a minute because the abstraction levels idea is very appealing, but ultimately I think this comparison is specious and misses the core issue entirely. XY problem is a communication and problem-solving anti-pattern where someone has already decided on an approach and is just asking about implementation details, rather than being open to potentially better solutions that might emerge if they just explained wtf they are trying to do first. Framing it as different abstraction levels obscures this real issue and incorrectly suggests it's just a natural variation in thinking styles rather than a specific mistake in problem-solving approach.
- solarkraft 2y agoI do find those suggestions valuable 80-90% of the time. But in the fringes, on less popular questions, it’s really frustrating because I can’t just substitute it in this way because the situation is just different.
- ithkuil 2y agoOTOH I'd love to have a model trained to address XY problems. Of course ideally you should be able to ignore that suggestion and force an actual response
- smitelli 2y agoIt's a case-by-case thing. Sometimes in this big world you see people asking things like "how do I write a firewall wildcard rule that blocks all IPv6 addresses starting with the number '2'?" and you just want to reach through the screen and shake them by the shoulders and ask them if everything is okay.
- ryandrake 2y agoI'd rather a Q&A site have a specific and correct A for every specific Q, no matter how ridiculous some user thinks the Q is. If I was open to re-thinking my design I would have asked something like "Please help me re-think my firewall rules..."
- kbolino 2y agoAnd, even when it is an X-Y problem, it may still be better to answer the question as asked, if only to then explain that maybe there is another problem that really needs to be solved. Looking at it from a pedagogical lens, it seems some learners (question askers) will fixate on X and be unreceptive to Y, stemming perhaps from some fundamental differences in conceptual understanding between the asker and the answerer. Trying to answer for Y will confound and frustrate; closing the gaps in understanding (and there may be gaps on both sides) is usually a lot more involved than just shifting the conversation from X to Y.
- corytheboyd 2y agoOne one annoying thing about using LLMs to answer technical questions is when you provide context like “I have tried X, Y already, and Z won’t work because…” it does what LLMs do— uses all the words in the prompt to string together more words into an answer. It would make LLMs so much more intuitive to use for technical questions if this problem could be solved… sounds so obvious that maybe it does exist?