I am wondering what was the reason to close my question Underfull \hbox and Underfull \vbox as being too localized. The question asks what for ways to fix the warning "Underfull \hbox and Underfull \vbox".
Also, I do not understand the first comment by David Carlisle to that question, since it is not clear to me how posting a small document will help.
Here is perhaps a sequence of events that lead to the closure, as well as some motivation behind it:
It has been decided by the TeX.SE community to host "Answer the unanswered" sessions (see Regular “Let's deal with unanswered questions” chat meetings?) in an attempt to reduce the number of unanswered or open questions. These "janitorial sessions" are there to refocus from the perpetual incentive to answer the next, new (perhaps easy) question, while older (perhaps more difficult) questions tend to be left unanswered and turn stale.
Another reason here is that some answers may never receive successful answers, while others are just too general to provide an effective answer. Even a recent discussion on meta.TeX.SE (see Should I self answer my question?) discussed whether specific answers to a general question should be answered in "combined form" to provide a definitive "answer to the original question".
While yours most likely did not fall in any of these categories, comments suggest that there was not enough detail to adequately answer the question. As such, it was brought up during the most recent "Answer the unanswered" sessions in chat, where your question was mentioned as perhaps being too localized/TL.
Your comment rebuttals suggest there was no need to supply a so-called minimal working example (MWE). However, this is a general misconception when it comes to answering questions on the Stack Exchange network. Not only are there a myriad of solutions to the warning "Underfull \hbox" (see No line here to end for one cause) and/or "Underfull \vbox", you are after solving a specific instance of this warning. This could depend on the code that you use, the document class that you use, the structure of the code that precedes that actual warning, or perhaps something different altogether (like TeX's line-breaking routine).
As a questioner you should try to put yourself in the shoes of the possible answerer in order to have the best shot at finding an optimal solution. That is why we usually push questioners to "provide an MWE that reproduces the problem." This requires you (the questioner) to provide as much information as possible so that an answerer can reproduce the problem (independent of the questioner), isolate what causes it, address possible alternatives and provide a solution.
Of course, if confidentiality is at play, 100% reproduction is very difficult. However, if the question is left in this general or even specific, open-ended setting, with no reproducibility, and it turns stale, it is most likely destined for closure as being "too localized" by definition:
This question is unlikely to help any future visitors; it is only relevant to a small geographic area, a specific moment in time, or an extraordinarily narrow situation that is not generally applicable to the worldwide audience of the internet.
As an example, the generality of your question may just as well have been solved by (to name a few when searching for "underfull vbox"):
\hboxes and \vboxes and how can I get rid of them?\vbox\vbox (badness 10000) … while \output is active”?But without knowing more detail, a solution provided at the above links might have helped, or not.
Werner has covered a lot of the reasoning in his answer, but as my name is on the 'closed' statement I think I should also chip in.
Some questions are unlikely to gain answers, not because they are general per se but because they are general in a way that seems to require a specific answer. Th standard approach to these questions is first to ask for clarification, and often this is best done using an MWE (although there are of course other approaches). We then normally allow a period for the original questioner to reply, and hopefully for an answer to then be given. After a couple of weeks, if there is no change to the question then it may be considered for closing, for the reasons Werner has outlined.
On this specific case, there were comments asking for further detail, but the question is written such that the problem is very general but the requirement (to fix them in a specific document) is for a very focussed solution. That fits the general criteria we use for closing as 'too localized', so I asked as part of the Answer the Unanswered session. There was support there for closing, so I voted to close. As a moderator, if I vote to close a question it goes through without needing to accumulate the standard number of votes (currently 5), but this almost always involves checking first.
The 'Answer the Unanswered' sessions of course do not imply that all questions should have an answer or be closed. There are many good questions to which there is currently no good answer, and these are left open. The aim is to keep on top of questions which are unanswered due to way the way the question has been posed or due to lack in attention.
Since I'm mentioned in the question....
It always helps to supply an example. If nothing else it saves the person trying to help from having to manufacture a document for testing.
In this case without an example it was really impossible to offer any advice. You showed a warning message and asked
What is the way to fix it?
Well the answer is:
Do not do whatever you did (or do what you did and ignore the warning and tell TeX that that is good enough).
Which isn't very helpful, but without knowing what you did it is impossible to be helpful.
As the question was unanswerable and had remained so for some time, it got closed as noted in Joseph's answer.
