I try to avoid creating duplicates but I always fail.

Consider my recent question.

I was searching the finite subgroup criterion.

So I typed into the search function "Finite subgroup criterion" and found nothing of relation

So I asked my question, write my post titled "finite subgroup criteria"

and somehow after a day I get flagged for duplicate and I find the post "Finite subgroup test" to be precisely the question I was looking for.

Is this a case of not using the words generally enough?

Are there better search techniques?

Thanks for assistance

  • 1
    $\begingroup$ I am not sure whether it helps in this specific case, but you might have a look at: How to search on this site? Some other questions linked there might be of interest, too. (This post is linked in the help center, too.) $\endgroup$ Commented Feb 12 at 5:38
  • 10
    $\begingroup$ Searching, in general, is hard work and smart work. Try various keywords in google, the math search engine Approach0.xyz, try searching just this site using google search with site:www.math.stackexchange.com added in. Also, "finite subgroup criteria" is way too general. Typing the entire criterion out on google could work, and then you could shorten unnecessary words where necessary. $\endgroup$ Commented Feb 12 at 5:47
  • 1
    $\begingroup$ Also, if you type up a question or specific parts of a question, M.SE has more criteria to use to compare it with and tends to give better search results. Also, In general, searching for something specific and distinct to your question ior topic works better on any search engine. $\endgroup$
    – nickalh
    Commented Feb 12 at 12:26
  • 10
    $\begingroup$ One should also keep in mind that duplicates can be useful. Unless your question is deleted, when somebody searches for finite subgroup criteria or finite subgroup criterion, they will find your question - and that will lead them to the duplicate (or duplicates) linked there. $\endgroup$ Commented Feb 12 at 13:21
  • 3
    $\begingroup$ @MartinSleziak If it is necessary to assist poor search tools then any such assistance should be added to the canonical dupe target(s). Otherwise we have no hope of stopping rampant duplication. $\endgroup$ Commented Feb 12 at 17:13
  • 5
    $\begingroup$ Don't feel too bad about having trouble with search - I still have trouble with it. But if you are having frequent issues, then it's probably worth taking @SarveshRavichandranIyer's suggestions and work on improving your skills. Just realize that it is a separate skill, and don't expect it to work as easily as Google. $\endgroup$
    – JonathanZ
    Commented Feb 12 at 17:57
  • 3
    $\begingroup$ Perhaps I should have mentioned the searching chatroom - which can be used to discuss various things related to searching and even asking for help with searching for some specific question. $\endgroup$ Commented Feb 12 at 17:59
  • 2
    $\begingroup$ Let me also add some other things that have helped me (1) I ask people to search in groups, sit together on your respective machines and plan out searches for mathematical content. If you have a group, brainstorm for 5 minutes for keywords and stuff and parallelize your search. (2) Use different search engines, like Google, Bing, DuckDuckgo and search.brave.com with the same keywords. They have different algorithms and one could work (3) Sometimes, try opening up these browsers in incognito mode so they don't have you search history in mind, and then search the keywords. $\endgroup$ Commented Feb 14 at 5:36
  • 1
    $\begingroup$ It's been a long problem that Stack's default search isn't that great. Most recommendations point to using another search tool. I think Stack Corporate is trying to use recent machine learning to make search better, so we'll see how that turns out. $\endgroup$
    – Teepeemm
    Commented Feb 17 at 23:23

1 Answer 1


Stack Exchange's search function is notoriously bad. Try Google, Bing, etc.

If you searched diligently, and you did a good job of writing your question in a general form, and your question turns out to be a duplicate, that is not a terrible outcome. It means that you will help others find the answer via search, because now anyone else who searches on "finite group criterion" instead of "finite group test" will readily find your post and then be able to follow the link to the answer.

Check a textbook. Find the concept in a textbook. Check what phrasing they use to describe it, then search using that phrasing. Also, sometimes you will find the answer in the textbook, and if not, it provides some context you can use to improve the question.

Sometimes, it is helpful to use subsets of the words when searching.

When you write a post, make sure to be clear about what you are asking. We are a question-and-answer site, so we require you to articulate a specific question. A question usually ends with "?". Don't force us to guess what your question is. In your post, I can't tell what your question is. Your post contains only statements, but no question. You mark a section as "Question:" but then what follows it is a declarative statement, not a question. So I find it painful to try to imagine what you might be trying to ask.

Also, if you are going to ask about "the converse", and it is not 100% clear what you mean by that, I suggest that you state explicitly the result you are trying to prove. Textbooks always show a careful statement of the theorem before trying to present a proof. Try to do the same. You need to be very clear about what you are trying to prove, before you can have any hope of proving it. Help us out, by providing all of the parts you can do yourself.

  • $\begingroup$ Hi thank you for the instructive comment! $\endgroup$ Commented Mar 30 at 4:02

You must log in to answer this question.

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