XY Problem | Hacker News Reader

XY Problem

xyproblem.info

35 pointslentil_soup20 comments

There's also the XY problem problem, where some people expect that every of a complex or unusual question is an instance of the XY problem, even when the requester really does just need a solution to Y.

The YXY problem?

Not to be confused with the ABA problem*

Or famous Swedish band ABBA

This is rampant on corporate support forums, where a "power user" with a gold star will answer this and shut down your question.

It's also common on Stack Overflow, but at least there, others will answer the actual question you're asking.

Both are avoided with the exact same approach: when you're asking a question, give the full context. It makes everyone's life easier.

I often find the full context gets into privacy issues, to put it one way, if I'm reluctant to get into the full context.

I feel like this is one example of a general second guessing phenomenon. Sometimes that second guessing is accurate and sometimes it makes too many assumptions about the other person, or otherwise fails to consider other explanations for their behavior.

Listen. If someone on stack overflow asks how to do X, do not tell them they don't want to do X unless you explain how they could do X. Maybe that poster would be better off not doing X, but someone else will come along some day who wants X for real.

I’ve also seen a similar phenomenon where I really want to try Y and people just keep telling me a different way to do X that I’ve already tried.

This is a good check on how good your help request is. Hopefully the problem statement includes a brief summary of the approaches already tried and why those were dismissed.

Or maybe people could simply answer the question as asked instead of answering other questions that the person didn't ask or demanding to know why the person is asking. The expectation that a person needs to tell you their life story to get an answer to a question is absolutely bonkers.

> person needs to tell you their life story to get an answer to a question is absolutely bonkers.

The problem is that the answer, for a given question, in any reasonably complex solution space, is entirely dependent on the very local problem space that you're forcing them to guess where it is in a much larger problem space area. If you're asking someone that's familiar with the broader view of each, they're probably asking because their guess of where you are is close to some "trivial" type solution. Or, they have a bunch of answers that depends on what you're doing, and you're forcing them to guess.

The goal is very clear in your head. The way to achieve that might be a web of possibilities. You asking suggests they can see the web and you can't.

This is something I have known (but not by this name) since the early 90’s. A proposed feature a customer brings to you is almost never what they need. You must iterate with them to discover their real problem and move on from there.

I left a project once because the sponsor wasn’t letting us interact with the real customer to find the real need. That path only leads to doom and despair.

This is why I love building things for internal users. I can just ping them on slack, and not have to go through layers of communication, or worry that I'll be over-promising something to a paying customer.

https://portal.mozz.us/gopher/hoi.st/0/posts/2023-04-30-the-...

I've seen this before and it has always rubbed me the wrong way, and now I think I know why. This framing of answering questions as "wasted time and energy" is assholeish and contemptible.

It is also not an appropriate framework for a responder to follow, only for an asker to think about what they should be asking. And far far far too many people read it and take away from it a belief that they need to find out the real reason a person is asking a question before answering.

When a person asks a question, you do not need to understand why they are asking it. It does not matter that you cannot guess why they want to do or know something. Assuming and digging for an XY problem is just looking for an excuse to not answer the question.

If the question has an answer and you know the answer, answer it and be done. The knowledge is valuable even for its own sake. If not to the person asking, then to someone else who has the same question for another reason. If you want to probe further, do it after answering the question.

Because, goddamn, if you think it's annoying to be asked about something that you think is the wrong thing to do, it's really fucking annoying for someone to demand to know your backstory before deigning to answer a question as asked.

I have learned a lot from people questioning my motives, and from third-party discussions where someone with the same problem had their motives questioned. It often set me on a much better course. Thorough answers benefit the audience too.

A good example of an XY problem is people asking where to buy a suicide cord. Is the helpful answer to show them how to make one, or to hint at the fact that it's called a suicide cord?

If you desperately want to shoot yourself in the foot, don't mind people asking why you want to borrow their gun.

> Is the helpful answer to show them how to make one, or to hint at the fact that it's called a suicide cord?

You create a false dilemma. The helpful answer is the one that answers the question, and if you also want to suggest alternatives, feel free to do more than answering the question, but don't just question their motives instead of answering, because questioning someone's motives instead of answering is an antagonistic behavior. You're basically saying "I don't also want that thing, therefore you must be an idiot." That's an obnoxious framework to operate under.

I once went to the doctor complaining about spleen pain. The doctor's first question was "where is your spleen?" and it was not where I pointed.

The helpful answer is sometimes to ask more questions.

Asking about Y (or Z, or some other problem a few layers down) is common when yak shaving. Aka doing the thing that's needed to do the thing that's needed to do X. Not to be confused with the also-present problem of ADHD sequential distraction by some other unrelated problem (possibly one sighted along the way to eventually get X done).

It's a gross idealization that every problem can be directly solved, or is "shovel ready." In my world there are often oodles of blockers, dependencies, and preparations that have to be put in place to even start to solve X. Asking about Y and Z along the way? Par for the course.

Reply on news.ycombinator.com