Finding duplicates for a question, if they exist, is not always easy. What techniques can I use to find question duplicates? How should duplicates be handled?
2 Answers
First of all, understand that having duplicates on a site is not a bad thing. They provide multiple landing points for visitors seeking a very similar solution. That's correct: Different questions leading to the same solution.
Perhaps someone is asking for the sum 12+15 while someone else is interested in finding the product 3×9. The example may seem trivial, but the principle carries through to a large number of possible duplicates. To rephrase while voting to close Q1 as a duplicate of Q2 doesn't necessarily mean that Q1 is asking the same as Q2. They may be asking for entirely different things.
Based on the above idea, I think about what would be required for an answer, and search for that. Additionally, in order to reduce the search results, adding appropriate tags help. Here are some examples:
Source: Highlight text in verbatim environment
Duplicate search approach: I know fancyvrb
allows for escaping to change colour, so before quickly writing up and answer, lets search for fancyvrb [color] is:answer
, sorted by votes. Note the search specifically includes a reference to answers only. Searching through the first page reveals Changing color of text snippet in a verbatim
environment which provides a viable solution.
Keys to finding duplicates include:
Being patient (not just writing up a solution because you know the answer);
Searching and perhaps refining this search using some ninja search options;
Switching to Google to search for solutions specific to TeX.SE by adding the search term
site:tex.stackexchange.com
. Google's search has often reveals far better hits than the search engine provided by Stack Exchange.
Meta references:
- How should duplicate questions be handled?
- Give an incentive for finding duplicate questions
- What constitutes a duplicate question? In particular, is a question a duplicate if the answer is contained in another question?
- “Close as duplicate” - what if only the answer is a duplicate?
- Should I mark as duplicate if the question is answered elsewhere, but it's not the same question?
- When is a duplicate question not a duplicate?
in addition to the suggestions provided by werner, there's also a very nice meta question
which can be checked for possible duplicates.
if something is there, good. if it's not, and you find a really great duplicate in your further search, add it to the list, to make the next person's job (or your own next search) easier.
and if a question you've found is nearly "perfect" (i.e. completely covers all possibilities), don't feel shy about adding another answer, or editing an existing one, to provide that missing bit. or add a comment, asking the person who gave the best answer to improve it by adding what's missing.