4

There are a lot of questions on Sec.SE that are duplicates, or, at least: near duplicates.

For example, this question has a lot of attention, and no mention of duplicates. Although it's similar to this much older question and I expect many other similar questions have been asked. A common theme is also seen in this question, where the question has some particular twist that makes it not-quite-a-dupe.

I think the problem is a bit deeper than people not bothering to look for duplicates. It's hard to find duplicates using search, because people don't always use search-friendly keywords. e.g. a search for "responsible disclosure" does not match "how do i report a vulnerability?" And when you do find duplicates, often the answers are only moderate quality, that doesn't answer all the nuances of the subject.

If we really wanted to fix this, we could work harder to create canonical questions with great answers, and tags that help them show up in searches. We could then more aggressively mark near-dupes as duplicates, potentially even editing the answers to cover the near-dupe questions. This would greatly improve the mission of letting people find good answers in web searches.

Maybe this is something that people don't mind. It's not necessarily a bad thing to have duplicate questions on the site. And you get more rep from a couple of upvotes on a dupe answer than you do for a close vote. But general Stack Exchange policy is to close duplicates, and we should really follow that.

7

The best way to make canonical questions easy to find is to list them in the tag wiki for the most relevant tag. I haven't seen it done (or done it) much here; I try to do it on other sites, for example ssh and sudo over at U&L.

Making it easy to find duplicates is good because it saves reinventing the wheel. It also ensures that the good answers will be all in one place, instead of having interesting tidbits spread over many near-identical questions.

I encourage everyone to do a sweep through tags and identify commonly-asked question. If there's a thread with a good set of generic answers, link it in the tag wiki. If there isn't, write a good Q&A pair or call on meta or chat for someone to write one.

  • I hadn't realised that was done - which probably means either I don't spend enough time on sites where it is done, or we don't let new folks know the value of the tag wikis. Excellent suggestion. – Rory Alsop Dec 2 '14 at 9:47
5

I support this. Where I know a dupe exists I try and find it, or where someone points out the dupe I try to close as dupe (and even merge if necessary).

I think there may well be a small piece of work to improve the questions so that it can indeed be considered 'canonical' but this will help clean the site up.

You must log in to answer this question.

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