No.2 on http://blog.stackoverflow.com/2010/07/the-7-essential-meta-questions-of-every-beta/


Should we put a caution on the longevity of answers? There are questions already about encryption and about legislation, both of which change with time (for instance, only a few years ago SHA1 and MD5 didn't have any known weaknesses; a couple of years before that SOx didn't exist). However, the problem is mainly going to affect "drive-by" visitors who google something like "best algorithm and key size to encrypt a document" so the FAQ may go unread by exactly those people. I don't know.

  • Perhaps include something in the FAQ that directs authors of such answers to add notation of the time-sensitivity in their posts. Or, maybe higher-rep users could edit the questions to include a comment to the same effect so that not every post has to have the same note. – Iszi Jan 5 '11 at 17:11

With the increasing use of social media mining tools such as Maltego we should be stressing that people should think carefully about what they are asking questions on. Especially if they are asking questions about their companies security posture or configuration.

If I was going to do a profile of a company there could be some rich pickings in this site as it develops.


We've got a question on here about asking discussion questions, with the opinion that discussions are OK. However, the site FAQ still has the boilerplate statement about subjectivity:

What kind of questions should I not ask here?

Avoid asking subjective or argumentative questions. If you must ask a subjective question, make sure it meets the six guidelines for great subjective questions, or it will be closed.

This would be better put the other way around, I think (with thanks to AviD for suggestions in the comments):

What kind of questions should I ask here?

[...] As many aspects of IT security boil down to interpreting policy or choosing an "appropriate" level of protection/risk, discussion-type questions are acceptable here. They must be on-topic and meet the six guidelines for good subjective questions. Subjective questions may be edited by other users to yield questions that can be answered objectively; those that cannot may be closed.

  • I agree. So in effect you're saying we should explicitly allow discussion / subjective questions? How should these be tagged? – AviD Dec 1 '10 at 11:13
  • 1
    On second thought, I agree about discussion questions - this is a type of peer research. Since we might not yet have all the answers, and there are a lot of factors to consider... But I don't like subjective questions here, as it is our field is perceived as black magic, when proper analysis can very much lead to objective answers. It might depend on various factors, which could be different in different situations, but for a given situation it should be objective. What I'm trying to say is discussion:yes, subjective:no. – AviD Dec 1 '10 at 11:16
  • 1
    @AviD: your comment makes me think that there are questions the asker would think are subjective but yield analytical answers, in which case it's hard to turn down subjective questions upfront. They just need to be moderated well, and users need to understand that this will happen. – user185 Dec 1 '10 at 11:27
  • @Avid: Regarding tagging, I don't think that we need a special tag for discussion questions. Tags like "watercooler" or "chat" imply bad questions, and a discussion might evolve out of answers to the question. Visitors will be able to tell from the answers what kind of answers there are to a question ;) – user185 Dec 1 '10 at 11:29
  • 1
    Hmm, @Graham, that is a good point. But what seems to be working on SO is if a question is asked in a subjective manner, it gets rewritten to be non-subjective in order to stay open/ontopic. Otherwise, if it cannot be rewritten thusly, it should get closed as subjective. – AviD Dec 1 '10 at 11:30
  • @Avid: I like that suggestion. I'm going to incorporate it into the answer. – user185 Dec 1 '10 at 11:31
  • re tagging, okay I agree with that - but what about a research type tag? Though that is too generic and soft... – AviD Dec 1 '10 at 11:31
  • @AviD: agreed, too generic. And research doesn't stay research for long; it becomes trade secrets, patents, standard practice, commodity, then legacy. And we can't have tags that decay over time ;-) – user185 Dec 1 '10 at 11:43
  • Hehe. But I do think there should be some kind of taggulary distinction between questions that are asking for known answers, as opposed to questions that are seeking to discover new information/analysis/perspective/etc. – AviD Dec 1 '10 at 11:49
  • @AviD: an analysis tag makes sense to me. It sounds more focussed than discussion or perspective, and perhaps we add to the FAQ that analysis answers should be justified by evidence where possible to promote useful answers. – user185 Dec 1 '10 at 11:53
  • No, I think analysis is something different: taking familiar rules/formulas/principles/etc, and applying them to a specific situation. Or, trying to "decipher" a complex bit of whatever. This is different from breaking new ground. Maybe research still is the best for that... – AviD Dec 1 '10 at 11:58

Adding this here too so it isn't missed. Questions about virus removal of specified viruses: Is it a right place to ask about viruses?


I think we should have something in there specifically around not posting requests for ways to hack a device/app/database/OS. Haven't yet thought of appropriate wording (It is 1am and it has been a long day) but after having to close a couple of questions on this topic I think it is worth having a standard FAQ answer.

You must log in to answer this question.

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