8
$\begingroup$

I was waiting to get some examples of questions that might illustrate a lower bound on the kinds of questions we want to field. Here are some examples:

In each case, the question has an answer that can be found almost immediately on Wikipedia, or enough of an answer can be found that the question should be made much more specific.

Should we vote to close such questions, after supplying a comment stating where to look ? There is also the risk of such questions being homework questions, though in this particular case I think it's more a matter of initial seeding of the site.

$\endgroup$
  • 1
    $\begingroup$ If Wikipedia has a decent article that answers the question fairly directly, then the question is not adding real value. $\endgroup$ – András Salamon Aug 17 '10 at 1:28
  • $\begingroup$ how would you classify the following question: "what's the best algorithm to this day for finding a Sperner coloring?" this you wouldn't find it on wikipedia, and if you try google there is no useful information. $\endgroup$ – Marcos Villagra Aug 17 '10 at 3:54
  • $\begingroup$ this would definitely be within scope in my estimation. $\endgroup$ – Suresh Venkat Aug 17 '10 at 4:25
  • $\begingroup$ Why are these questions closed when they are within the scope? This community is not very considerate!!! $\endgroup$ – txwikinger Aug 17 '10 at 14:36
  • 2
    $\begingroup$ I think that if "google it" is a correct answer, the question doesn't belong. There should be an implicit assumption of "I have googled it" in any question asked. $\endgroup$ – Evgenij Thorstensen Aug 17 '10 at 15:19
  • $\begingroup$ @txwikinger I think the reason that basic/googleable questions are voted closed is because we want this site to be useful to the community. The community, as far as I understand, is researchers and people (w/ a reasonable cs background) interested in TCS. If this site becomes filled with questions like "What is an MST?" and "How do I sort?" or on concepts in wikipedia, it stops being useful, at least to the intended community. Many participants here are probably not so interested in spending their time answering homework questions, and if this is what the site becomes, people might leave... $\endgroup$ – Lev Reyzin Aug 19 '10 at 16:27
  • 1
    $\begingroup$ yes. See for example the Market for Lemons problem: en.wikipedia.org/wiki/The_Market_for_Lemons Ignoring the pejorative connotations, the principle is that a forum for experts can draw in amateurs, but not vice versa. And I think most of us here intend to use this as a forum to draw on expertise. For example, one of the most famous researchers in algorithm game theory is answering questions on this site as we speak. $\endgroup$ – Suresh Venkat Aug 19 '10 at 20:53
  • $\begingroup$ This is not the issue. The issue is that according to Jeff Atwood, you cannot change the definition of the proposal after people have already committed to it. If this would have been clear, before the commitment, people could have made different decisions. The proposal is something like a contract, if you change it after the commitment, you are in breach of contract. $\endgroup$ – txwikinger Aug 21 '10 at 1:32
  • $\begingroup$ but what is the definition of the proposal ? if you look at the list of off-topic questions, you'll see one of the questions listed as 'not a good example of off-topic' because it's easily googlable and is in wikipedia. the intent there was that it wasn't off topic per se (i.e on a different topic), it was just inappropriate. $\endgroup$ – Suresh Venkat Aug 21 '10 at 5:18
2
$\begingroup$

I would agree that the three questions above are examples of the type of question that should be closed due to being too elementary.

Having said that, I would like to argue that the scope of allowable questions should be kept fairly broad. As someone who hasn't done graduate coursework in TCS, I find the CW/big-list questions very helpful. (My own question had 4 votes to close before anyone had a chance to reply, but I think it turned out to be useful to others.) I know that you don't want the mix of questions to become too elementary and put off the serious researchers. It would be nice to be able to segregate the more basic questions from the advanced, but given that the theory community is small, there probably aren't enough people to support two separate StackExchange sites. And hopefully some of the people asking basic questions will eventually be asking (and answering) more advanced stuff.

Here's one little idea that might be workable: ask that anyone posting a question that could be considered undergrad-level material to tag it with something like "intro". Then anyone who doesn't want that stuff cluttering up their view could add the intro tag to their ignore filter.

I think over time the mix of questions is going to include more like Jeff Erickson's, and this will become a less sensitive issue. (By the way, has Bill Gasarch visited here yet? He's posted some questions in the past on the CC blog that would be great for this site.)

$\endgroup$
  • $\begingroup$ given the difference of opinions overall, this might be the best way to do this. let people filter tags that they don't like. $\endgroup$ – Suresh Venkat Aug 27 '10 at 6:39
  • $\begingroup$ I like the option of adding an "intro" (or some such) tag to a question that the poster/moderator(s) might think as introductory since I am essentially self-taught in this field and would have a few (probably a lot many) holes in my understanding and really look forward to posting some of them here so that the experts can step in and clarify them (just like I did for the FNP complexity class). There are many like me who don't have anyone to consult/go to in case of problems in TCS and this site seems to be a blessing in disguise. $\endgroup$ – dhruvbird Sep 11 '10 at 9:42

You must log in to answer this question.

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