I'm sure others must have hit this problem after being on EL&U for a while. With some questions which crop up repeatedly, it can be quite difficult to find an earlier related or exact duplicate, even if you know it exists.

As an example, I know there are a lot of duplicate/overlapping questions covering the subtleties of conditional tenses (could/should/would, might/ought, etc.). But it's a waste of time searching for any/all of those words because just about anything returns thousands of results.

It's pointless sorting any results by "most votes" because many of the "definitive" answers are quite old, but have few votes. Often they effectively "lie dormant" because more recent new users never come across them to upvote. There may be other ways to improve the situation, but I suggest...

Increment a count against existing question when a new one is closed as "exact duplicate" thereof, and make that count available as a "sort results" key in searches.

It seems to me the questions I'm interested in should rapidly float to the top with such a system. Added to which I'd expect a significant improvement in Answer quality over time. Later contributors would have a vested interest in vying for upvotes on pages they'd expect to be visited often, well into the future. And all visitors would be likely to upvote good answers.

I'm saddened by the fact that the most upvoted Questions of all time are a somewhat anal discussion about nested [sic]'s, and a closed Dalai Lama joke. This hardly reflects what I like to think of as EL&U's laudable aim of becoming a consultable reference source for future visitors. Anything that might raise the profile of genuinely important / repeatedly relevant questions over these "viral" distractions must surely be a good thing.

share
add comment

2 Answers

I believe this is already what the FAQ tab of the Questions page does. Or at least, the tooltip for the "FAQ" tab says "questions with the most links", which certainly sounds like what you're describing.

Of course, this is no help when the duplicate you know exists is not actually a frequently asked question, merely one that has already been asked at least once. For that, we need a better search function, with a use/mention distinction of some sort.

share
1  
I don't know what "Questions with the most links" is supposed to mean in the first place. Most internal links to/from other EL&U pages? Most links out to other pages on the Net? Probably not "Most links from previously-closed questions", since it includes many that have actually been closed. Anyway, there are currently 1500 of them, and I honestly cannot see what possible use that particular list would be to anyone - there's no way to sort/search it, and it doesn't even seem biased towards "quality" questions/answers in any way. –  FumbleFingers Aug 24 '11 at 19:04
    
I take it from your comment that you never seriously looked at that page, and that you don't see my problem as significant for the general case (since you didn't even upvote the question). I won't deny I am disappointed. –  FumbleFingers Aug 24 '11 at 19:06
    
@Fumble: I don't know how you came to the conclusion that I think your problem is insignificant. I've had the same problem, and I'm willing to bet so has everyone who cares even a little bit about duplicates. I'm saying that (1) part of the functionality you want already exists, and (2) it wouldn't always help. –  Marthaª Aug 24 '11 at 20:50
    
Well, the first entry under Questions->FAQ is a trivial one Origin of the term "by the way", and the last is "Grizzly status" [Closed]. There's no mechanism to sort, and the list content seems meaningless anyway. In what way might this list feasibly useful for anything, let alone provide part of the functionality I seek? You say you've had the same problem, yet clearly don't have a solution. So why would you not at least upvote my question as being worth asking? And "wouldn't always help" is irrelevant - if my idea works, over time it would help more and more. And improve quality, –  FumbleFingers Aug 24 '11 at 21:16
    
@fumble Martha is correct, the /faq tab does this and it's measured by internal links. If the question is linked from another question, comment, or answer, that makes it linked, and it will show up in the "linked questions" sidebar. –  Jeff Atwood Aug 26 '11 at 3:22
    
@Jeff Atwood: Looking at the question currently second (What is an alternative way to say "I am done"?) in the faq list, I see no links to anything there. Why would there be many links to such a trivial question? Are you sure it actually works? That list doesn't seem to reflect anything meaningful to me, and it changes radically from day to day. –  FumbleFingers Aug 26 '11 at 3:36
    
...the "related questions" sidebar doesn't seem to be about links. It just looks like high-scorers on pattern matching some words which may not be relevant in the first place. –  FumbleFingers Aug 26 '11 at 3:38
    
@Fumble: "related questions sidebar" <> "linked questions sidebar". The former is, indeed, based on pattern matching, but the latter is based only on links. Also, what FAQ list are you looking at? I don't see that "I am done" question on it anywhere, let alone in second place. –  Marthaª Aug 26 '11 at 13:38
    
@Jeff Atwood: ok, so can the existing FAQ functionality be applied to searches somehow? I.e. "sort by frequently-askedness" or something? –  Marthaª Aug 26 '11 at 13:43
    
@Martha: I click on Questions, then the "faq" tab which has the "hover text" questions with the most links. As I said, this list seems to change radically at least every day. When I posted my earlier comment, the one I linked to was indeed second in the list. It's not even in the first 300 today, and I got bored looking after that. I don't make these observations up, you know. –  FumbleFingers Aug 26 '11 at 15:17
    
@Fumble: there's something screwy going on with your FAQ list. Is it maybe taking you to the last page or something? For me, the top questions are: english.stackexchange.com/questions/2964/… english.stackexchange.com/questions/15780/… english.stackexchange.com/questions/1482/… –  Marthaª Aug 26 '11 at 18:11
1  
@fumble are you looking at english.stackexchange.com/questions/tagged/word-choice?sort=faq ? If not, you might be in the wrong place. Basically that faq tab, when browsing by tag, means show me questions with the word-choice tag that are commonly cross-linked to other questions. –  Jeff Atwood Aug 26 '11 at 18:24
    
@Jeff Atwood: I am looking at that list, but I hadn't filtered using any particular tag. The one giving that "trivial/irrelevant" question second in the list didn't have any search criteria at all. On my first "real-world" attempt I had searched for various combinations of could/should/would/etc., but never found the good answer I knew was around somewhere. I've just looked at this "faq" list after filtering for[single-word-requests], where the second entry is Is there a word or phrase for the feeling you get after looking at a word for too long?. I don't see why that should be up there! –  FumbleFingers Aug 26 '11 at 19:07
    
@Fumble: The semantic satiation question is up there because every time someone finds it, it's a revelation to them that they then have to share. :) –  Marthaª Aug 27 '11 at 13:04
add comment

For info, we do now make this data available when selecting a duplicate (i.e. the "close as duplicate of" popup); see http://meta.stackoverflow.com/questions/104096/i-want-to-know-how-duplicious-a-question-is-and-if-it-is-the-best-example-of-suc

It is not currently available separately (i.e. for use as a sort); I'm also a little dubious how much real use that sort would see. It might be possible to get this from SEDE, though.

share
    
My response is on meta.se –  FumbleFingers Sep 1 '11 at 12:35
add comment

You must log in to answer this question.

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