[resonious]

Yeah the whole premise of the "XY" problem seems like it was written up by someone who regulars Stack Overflow. It's pretty frustrating when the highest-upvotes answer to a question is "actually you don't want to do that. do something else instead".

[pie_flavor]

The goal of asking a question is not to get the exact answer to the words you spoke, it is to solve the problem the question is inspired by. It's a funny psychological thing, though, where people perfectly comfortable with the idea they don't know the answer to a question are easily offended by the implication that they don't know what question would lead to a solution, even though the foundational knowledge to answer the question is often the same as that required to ask it.

The reason people assume it's the wrong question is because they've seen a thousand other people ask the wrong question. This assumption is not a reflex that comes naturally, it is learned.

Experienced people who really have the question, understand XY just fine and start off with the context, but are happy to provide more; experienced people who say they are offended at common assumptions of XY tend to be in complete denial about legitimate cases of XY.

[tzs]

> Experienced people who really have the question, understand XY just fine and start off with the context, but are happy to provide more; experienced people who say they are offended at common assumptions of XY tend to be in complete denial about legitimate cases of XY.

That's not how it works on Stackoverflow and similar sites. Here's what happens when an experienced person who really does have the question, understands XY just fine, provides context, and is willing and able to provide more context is needed tries to get an answer on Stackoverflow.

1. They do a search and find that people have asked X before on Stackoverflow.

2. They read those questions and answers and find that in all of them the answerers decided that it was an XY problem and answered Y. Maybe the answers were right or maybe they were wrong, but in either case nobody answered X.

3. They post a new question asking X, explaining that the existing questions and answers do not apply because the person actually needs X and provides sufficient context to show this.

4. The question quickly gets closed as a duplicate of one or more of the others.

[hgsgm]

People are OK with what they know they don't know, as no one knows all the "trivial details", but not OK with what they know being wrong.

[JdeBP]

It wasn't. Mark-Jason Dominus was the cause of that name.

I, Raymond Chen, Charles Cazabon, and Eric S. Raymond didn't use that name at all.

I, in particular, chose a name that reinforced the an important part of the concept, missed in this discussion here, that the problem was nonsensical as posed.

* https://jdebp.uk/FGA/put-down-the-chocolate-covered-banana.h...

[mjd]

Thanks! But it was actually Greg Bacon (responding to my original article) who thought of calling the phenomenon "XY problem".

[JdeBP]

That's why I went with "cause" not "coiner". (-: People took your "X" and "Y" and ran with them, and the fact that "X" and "Y" are not blatantly nonsensical together caused some of the meaning to be lost. I've watched this happen over the years on the likes of Stack Exchange.

If Greg Bacon had gone with "Bacon and Hand Drill Problem" instead, perhaps this wouldn't have become so distorted. Although I know a group of Wikipedia writers who would no doubt insist that bacon logically goes with everything. (-:

[arcanemachiner]

I think this is because people Googling a question often end up at a SO post that answers a slightly different question. But since the keywords blazed a trail to a certain answer, that's the one that gets the upvotes, even though it's not the "correct" answer.

This has happened to me many times.

[travisjungroth]

That’s incredibly common and so frustrating. The title will be “How do you delete a column from a dataframe?” and then they’ll include a bunch of super specific code and paragraphs of context. The only answer will be “You actually don’t ever use that column. Just don’t add it in the first place.”, which is not an answer to the title question at all.

[ilyt]

It does happen very often when person trying to fix the issue is not an expert in the domain they are tinkering with and it is just some extra needed for their job and not main part of it.

We see it all the time with developers for example:

"I need sudo access to do XYZ"

"Why?"

"Because I can't access this directory"

"Ok, I fixed permissions and put it in CM, test it without sudo"

"Oh, it works, thanks".

But, of course, on SO you don't have that context.

[einpoklum]

That's why a good question gives some context regarding the motivation, or says something like "I have to do Y for reasons. Please don't answer about problems X or X' or X'', I just need to do Y."