Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 153 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I made this question: Why C++ specs say that int and short are both at least 16 bits?

It is in the same style as this question: int a[] = {1,2,}; Weird comma allowed. Any particular reason?

Why did my question get closed, while the other question (that is highly visible, so it is not like people just missed it) is not?

share|improve this question
I can't find a link, but this is a common question, the standard response to this is the presence of a question that is not closed does not mean that such questions are acceptable... usually means the older question was not found by someone to delete it. – psubsee2003 Feb 20 '13 at 14:29
Except like I mentioned, the older question is not THAT old, and I deriberately clicked on the most voted question on the tag (meaning, it has seen lots of eyeballs). To me the criteria is looking like if someone with more than 20k of reputation is asking the question, or writing in the comments against the question. (as soon as Raymond wrote a negative comment, the question got 1 downvote and the close votes). – speeder Feb 20 '13 at 14:30
I never said it was old.... I said older, which is true, the question is older than yours. – psubsee2003 Feb 20 '13 at 14:31
But as @ShaWizDowArd said... people were more forgiving even just a year and a half ago. A good percentage of some of the most highly upvoted questions on the site might not have made it beyond the first few hours if they were asked now. – psubsee2003 Feb 20 '13 at 14:34
Or they are more forgiving to anyone that has high reputation because that person has high reputation, thus whatever that person do must be more reputable... I had not asked a question in SO in many time (because of how negative SO recent behaviour has been), and now that I ask a very serious question that I had genuine interest to know, it gets closed because Raymond complained. Where I should have asked then? This question noone can prove that it was badly written or anything like that. – speeder Feb 20 '13 at 14:37
"they are more forgiving to anyone that has high reputation because that person has high reputation"...nah, not really. "it gets closed because Raymond complained"..again, not really. Raymond was just one of 5 users who thought the question should be closed. – Bart Feb 20 '13 at 14:40
He was the first, and people were upvoting it until he complained, then immediately (about 2 minutes later when I checked) it already had 3 close votes and one downvote. – speeder Feb 20 '13 at 14:41
Ironically, this is an older answer that predates the non-closed question you are referring too, but here a good explanation/reference: Why do some off-topic questions get closed, and some don't? – psubsee2003 Feb 20 '13 at 14:42
One difference is that the C standards committee has explicitly documented the reason for allowing the trailing comma. See Rationale for the C99 standard. That removes the "not constructive" close reason, because the one correct answer is known. – Bo Persson Feb 20 '13 at 17:09
And what about all other questions in language-lawyer? most of them are not closed... – speeder Feb 20 '13 at 17:32
but @BoPersson to be fair, until there are answers no-one would know that... – Kate Gregory Feb 20 '13 at 18:18
@Kate - Some of the language-lawyers would know. :-) There are quite a few of the regulars in the C and C++ tags that have read most of the background material published by the respective ISO committee. Some have even written a few of those docs! – Bo Persson Feb 20 '13 at 18:27
@BoPersson I should have written more clearly. The majority of possible question-closers wouldn't know, in the absence of answers, that the question was answerable. I have seen C# questions gain a comment that "unless Eric Lippert answers we will never know" followed by an answer from Eric. Closers may underestimate the community – Kate Gregory Feb 20 '13 at 18:34
up vote 5 down vote accepted

The tag is a lightning rod for questions that don't belong on Stack Overflow, for exactly the reason that Raymond mentions in his comment:

Questions about why a language is designed a particular way are not usually constructive. What is your practical programming problem? If you need types of a specific size, then use the types designed for that purpose like int32_t. If you want to know how many bits are in a particular type on a particular implementation, use <limits.h>.

I personally think the tag should be abolished, but the C++ crowd is rather vocal about it.

Your question was closed because of the reason that Raymond Chen stated. It's a solid reason that could apply to most of the language-lawyer questions. It's primarily because they've fought for their tag that it's been kept.

share|improve this answer
Then what should be done? This is allowed, or not? Right now the status is a limbo, some questions get closed, while others not, without clear criteria, because although they don't fit the whole of SO, they fit a subset (C++ coders, including me). – speeder Feb 20 '13 at 15:35
The "why" questions generally don't belong, but "is something well-defined/implementation-defined/undefined behaviour" tend to be legitimate, IMO. – Daniel Fischer Feb 20 '13 at 15:36
@speeder not everything is black and white. Some things require judgement. And some things are just luck about which people happen to see your question. Ask questions about real problems you face and things will work out well. If you face a real problem related to type sizes, ask for help with that problem rather than background on language design, and see what happens. – Kate Gregory Feb 20 '13 at 15:45

Questions that ask "why" are almost always closed. It's a rare treasure that survives. The "extra comma at the end" one is intriguing. Notice that the asker:

  • read and quoted from the spec
  • anticipated the most common response and refuted it
  • left people wondering "hm, good question, why is that?"

Most why questions don't show anywhere near that level of care. They ask hypotheticals with almost no background work: why is true 1? why can't it be 42? why aren't ints 12 bits long? why is bool four letters long, boo would work better? why was the word main used for the entry point? and either nobody cares, or everybody already knows, or it just doesn't matter. The question doesn't engage the interest of the handful of people who might have some information to provide.

Most people with high reputation got that reputation by knowing a lot about their field of interest, being curious about their field of interest, and being generous with that knowledge and curiosity. It's hardly surprising that such a person could, occasionally, ask a "why" question that didn't deserve near-instant closure. I find that far more likely than a herd mentality that lets high-rep people do what they like.

And as for going along with what Raymond Chen says, find me a time when he was wrong on a technical matter and I'll stop going along with him on technical matters. I don't take his advice on what movies to watch or what shoes to buy, but his point about just using the language as written (and the tip to embrace limits.h instead of trying to memorize it all yourself) is a very good one.

share|improve this answer
I disagree with your first sentence. Questions that are just 'curiousity' questions are more likely to be closed, but otherwise, Why questions seem to do really well:… – George Stocker Feb 20 '13 at 15:27
Interesting thread. I especially like "What good is knowing the answer?" I think it's possible for "most why questions get closed" and "why questions that meet certain criteria are good questions" to both be true. "Almost always" might have been too strong - but might not have; numbers would be great here. – Kate Gregory Feb 20 '13 at 15:39

On quick glance:

  1. That other question contains code.
  2. It appears that other question explains the issue better.
  3. That other question was posted year and half ago when people were more "forgiving".
share|improve this answer
If you visit the language-lawyer tag, it is filled with similar questions, many without code, or more recent. Some are closed, some are not, without clear criteria. Why not just outright ban all questions in the tag instead? – speeder Feb 20 '13 at 14:29
I understand your frustration, but from quick look in the tag questions I really don't see many of them closed. – Shadow Wizard Feb 20 '13 at 14:30
Then why my question was closed? Someone thought it fit perfectly the tag (that I did not knew that existed, someone else that added the tag), and the tag exist, and most questions on it are not closed... why my question was? – speeder Feb 20 '13 at 14:32
Sorry, I can't read the other users mind... apart of what I said here I really can't know. – Shadow Wizard Feb 20 '13 at 14:34

You must log in to answer this question.

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