1

I have recently noticed that very often answers to not actually answer the question but try to explain that the question is wrong somehow.

Here are some examples:

If we have a look at the first example of the list, the question is basically "how can we introduce a stand-up meeting in my 20-people/11-simultaneous-project team?". Of course we are tempted to answer that 11 projects for a unique big team might be the underlying problem, but should we? I mean, I'm sure that SFreebairn is aware of that.

In the third example the question was "Am I wrong when I think that my product owner should not deal with task assignments?" But I could not help trying to figure out why the product owner behaved this way.

When you ask a question here, you surely are not searching for a definition list or a lecture on how you don't understand what you are talking about. On the other hand those answers often bring up some useful and interesting points. I personally upvoted the second example of the list.

How should we handle this kind of answers? Should we keep on answering this way?

UPDATE

Answer the question

Read the question carefully. What, specifically, is the question asking for? Make sure your answer provides that – or a viable alternative. The answer can be “don’t do that”, but it should also include “try this instead”. Any answer that gets the asker going in the right direction is helpful, but do try to mention any limitations, assumptions or simplifications in your answer. Brevity is acceptable, but fuller explanations are better.

Reading this I think I might have been a bit hard, since the answers I'm talking about match this paragraph somehow. So the discussion should be different : Should we keep allowing this kind of answers? Do they really help the one who asked it? Aren't those answers a bit too condescending by nature?

  • Hi Matthias, I'll take a deeper look at the answers myself, once I get my bike towed to the shop, but take a look at How To Answer and consider including your interpratation of this document in your meta post, assuming it doesn't answer your question. Thanks for bringing this issue to meta! :) – jmort253 Dec 8 '12 at 17:49
  • Hi Matthias, thanks for the clarification. Interesting perspective. I can honestly say that I've never interpreted anyone's answer as being condescending. In fact, people put a lot of time into answers that provide sound reasoning for why they should do X instead of Y. As for "do these help", it depends on the question. – jmort253 Dec 8 '12 at 18:41
  • The issue I see with answers is that some are extremely long. There's a balance I think we need to attain between providing good explanation as well as conciseness. If an entire book can be written on the subject, then perhaps we should comment on the question and seek clarification instead of writing said book. :) – jmort253 Dec 8 '12 at 18:47
  • 1
    @jmort253 Um, didn't you know that Stack Exchange pays us all by the word? :) – Todd A. Jacobs Dec 9 '12 at 1:28
  • @CodeGnome - Lol, in that case, you're a pretty wealthy individual! ;) – jmort253 Dec 9 '12 at 1:33
  • So Matthias, I think I am starting to understand what you mean by "condescending". My interpretation of that word is "we're looking down our noses at question askers", but I think what you meant was that the answers feel like we assume every question-asker is ignorant to all PM processes, because we're always nice to people. ;) If the latter is the correct interpretation, then I can definitely see your point. :) Treating askers like they're clueless could alienate not only askers, but the future visitors and readers who may actually be real, practicing PM's. – jmort253 Dec 9 '12 at 7:29
  • Also, you might want to take a look at the Area 51 Stats. Now I know I'm the first guy to say "don't focus on the stats", but I do watch the visits per day stat pretty closely, and it's been steadily increasing since Summer, so that could be interpreted as we're doing something right. Just don't let the questions per day scare you. That one will eventually increase organically as more and more people find the site and become members. ;) – jmort253 Dec 9 '12 at 7:33
1

When determining how to evaluate an answer on PMSE, since our community hasn't come together to define any guidelines of our own, I use the information from the How to Answer section of the FAQ. The "Answer the Question" section seems most applicable in this case:

Answer the question

Read the question carefully. What, specifically, is the question asking for? Make sure your answer provides that – or a viable alternative. The answer can be “don’t do that”, but it should also include “try this instead”. Any answer that gets the asker going in the right direction is helpful, but do try to mention any limitations, assumptions or simplifications in your answer. Brevity is acceptable, but fuller explanations are better.

All of the answers seem to be providing viable alternatives and also explaining why the alternative is sound. Even answers that say "don't do that" provide alternatives.

On David's answer, we actually have a meta discussion here about this very issue, but as David points out, questions like these point out flaws in new PM knowledge that's important to set straight, which makes the question and answers useful to future visitors. I also believe it's our role to point people with flawed understanding of processes in the right direction.

With that said, if you encounter answers that just point out flaws in the process without providing alternatives, you could consider a downvote, but first, be sure the information pointing out flaws isn't more important posted than not posted. If I was going to electrocute myself, I'd most certainly hope someone would try to stop me, even if that person didn't know how to safely work with electricity herself. :)

  • You must be right. Sometimes those answers look like the classical "-how do I fix this foobar problem in windows? -install Linux!" This is just not the question. But your comment on David's answer on the other meta discussion completely makes sense. A googler might be in the same situation as the person who asked the original question and be interested in what you call the ignorance part. This is a deeper analysis that must be kept as long as the actual question is also treated. – Matthias Jouan Dec 9 '12 at 0:09
  • The main problem with the Windows/Linux example is there's no good reason given to uninstall Windows and switch to Linux. In general, those are trolling answers. ;) If David had just said "You're measuring just to measure, don't do that", then the answer would leave a lot to be desired. This is precisely why I feel it's so important that we, as answerers, always provide an explanation in our answers as to why it's correct. Depending on the problem, "don't do X, do Y instead" could very well be the best solution to the problem, even if it's not really what's asked. – jmort253 Dec 9 '12 at 0:40
  • This particular discussion may help us edit the What is an acceptable answer post so that it fits the PMSE context. As the site grows, it's helpful to have some clearer guidelines. – jmort253 Dec 9 '12 at 0:42
4

X/Y Problems

While people do sometimes misunderstand questions, in many cases people are attempting to help with an X/Y Problem. What is the XY problem? explains:

The XY problem is asking about your attempted solution rather than your actual problem.

That is, you are trying to solve problem X, and you think solution Y would work, but instead of asking about X when you run into trouble, you ask about Y.

Avoiding X/Y Answers

Answers on purely technical stacks are fairly straightforward, in that an answer either works or it doesn't. Answers on PMSE are less straightforward since both the questions and the answers are more subjective, often involve soft skills that aren't subject to mathematical proofs, and are generally about people and process which means that the answers are rarely orthogonal.

With all of that said, if you want to avoid answers that suggest that you look at a process other than the one that you're currently pursuing, then you should probably note in your question what processes or solutions you've already examined and discarded, and why. Without that information, answerers will generally assume that you haven't considered the alternatives.

"Too Localized" or "Not a Real Question"

It may also be useful to consider that answering a given question without consideration of the underlying issues or possible X/Y problems leads to answers that aren't useful to future visitors, or that aren't really answerable without edits that destroy the author's original intent.

For example, if someone asks:

I am constrained to a single FTE resource capped at 34 hours per week. How can I deliver a successful project that requires 4,000 man-hours within 5 business days?

Questions like that beg for an X/Y answer if they are to be answered at all. Addressing the preconceptions in the question, or the embedded assumptions about the underlying process, are essential to providing a meaningful answer.

If we decide (as a community) to disallow X/Y answers then all that will do is increase the number of questions closed, questions downvoted, and questioners auto-banned for asking low-quality questions. This doesn't seem desirable to me, since many of the X/Y questions elicit valuable answers that remind us all that good project management is about examining underlying process and assumptions, and making them as explicit as possible.

  • You made a very good point with xy problems IMO. The problem is then for the answerer to figure out what should be said and what is obvious enough and should not be said. On the standup meeting question for example, is it necessary to say that 11 projects is too much? On the bug question - which is a bit special I must admit - is it necessary to define what a bug is? when writing an answer that contains an "ignorance" part, what level of ignorance should we assume? – Matthias Jouan Dec 9 '12 at 2:06
  • 1
    Great answer CodeGnome. Wish I could upvote the "Avoiding X/Y Answers" section more than once. @MatthiasJouan, I don't think we should assume the person is so ignorant that we should write War and Peace, but somewhat generalized answers are okay. I feel like we should write for the average visitor. So out of 1000 visitors that land on that bug post, how many will already know what a bug is, for instance? Will most of those 1000 people laugh when they see someone try to hold an 11 project standup meeting, or will they too need an explanation to clear up their misunderstandings? – jmort253 Dec 9 '12 at 7:14
  • Here are some stats for you guys to look at that are public, so anyone can see them: Quantcast PM Education Level. We have 1.78 times more graduate and post graduate level visitors than the general Internet, and almost 48% of our visitors are said to have college degrees. Avg age is between 25 and 44, followed by the 45-54 age group. I'd say this means most visitors should have a clue about most of this stuff, but this is subject to interpretation. – jmort253 Dec 9 '12 at 7:21
  • What's more, it just hit me that Stack Exchange always beats into our heads that we're building sites for experts. They tell us that if we build a great resource for the pros, we'll attract both pros and enthusiasts/newbies alike. But if our content assumes everyone is a n00b, then it's possible we could scare away the pros. I'm not sure if we're overdoing it, but this is definitely something for us to think about. ;) – jmort253 Dec 9 '12 at 7:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .