What is meta? ×
Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 130 Stack Exchange communities.

Too many times I just know something's a dupe but can't for the life of me find the original question that is duplicated. I search the sites from Google to find things, but in coming up short there I usually leave it and rely on someone else to clean it up [sigh]. What are other ways to do it better?

(or maybe this is one that gets closed employing said techniques...)

share|improve this question

4 Answers 4

up vote 3 down vote accepted

After Google (using: site:stackoverflow.com <search phrase>), I find that the best way to check for duplicates is to go the the 'Ask Question' page and type in a title for the question, as you would ask it. The 'Related Questions' suggestions contain a really good list of candidates to look through. If you don't see it, change up the wording a bit and try again.

share|improve this answer
    
can i find google here? i must vote him up :) –  Umer Jan 17 '11 at 7:29
    
This doesn't always turn up all duplicates. –  Mark Cidade Nov 3 '11 at 19:17

More often than not, they're simply listed in the "Related" column...

(And obviously, if you think you participated in a duplicate, search with user:me.)

share|improve this answer
    
Ah, yes. Thanks for pointing out the user:me bit. I do do that, just forgot to mention. –  squillman May 10 '10 at 18:54
  1. You can use the regular search tool for what you're about to ask and see if similar questions pop up.
  2. There's an automated suggestion tool that pops up when you enter a title in the Ask Question page. This uses a different algorithm than the one in the regular search.
  3. You can use Google with the string "site:stackoverflow.com" and your question to see if their
  4. You can think of synonyms for key words/phrases in your question, and try steps 1-3 again after using them to modify your initial query.
  5. If you can find questions on a related topic, whether through a search page or the tags list, you might be able to find a dupe of your question in its "Linked" or "Related" lists.

It sounds like you already used methods 1 and 2. 5 is a longer shot and isn't really efficient as a first-line tool. In my personal experience, 4 is the most critical one because different people think differently and there's nothing you can do about that; your mileage may vary depending on how closely you think to "the norm," insofar as a "norm" exists.

share|improve this answer

Despite lamingtons and lamentations, the site search works pretty well in finding the many dozens of duplicates and their masters.

The search page gives the main point when doing these lookups:

to ensure that the words appear in the results
+apples +oranges

Add that with a [tag] in the search and you're on your way to voting to close as a duplicate and watching that vote age away as six more answers are posted in the time it took to bend your knees.

People from all sorts of backgrounds may word things in their own special broken dictionary ways, but in the end, there are a few unique keywords in the post that will always lead you home.

And if there aren't, chances are good you're looking at a question that's done no research at all and a candidate primed for closing.

But it's leagues less work duplicating an answer across five different questions with different ways to say "thanks in advance" than it is to plug in a few keywords into the search. And that's why users have a lot of reputation.

share|improve this answer

You must log in to answer this question.

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