[stack] overloaded, crossed and invalidated

For the few past days, I have been monitoring Cross Validated, a forum on statistics that is part of StackExchange (“a fast-growing network of 71 question and answer sites on diverse topics from software programming to cooking to photography and gaming“…) for questions of interest, but I think (hope?) I will stop there my involvement. First, I fear this type of forum is too addictive (at least for me, and I already have my share of web-related addictions, witness this very blog!). Of course, Cross Validated is an interesting site for questions related with statistics and machine learning, with a great LATEX interface that allows to type math formulas in a natural way; however I also find the exercise rather frustrating and to some extend futile, which is another reason why I do not wish pursuing the experience any further. For one thing, some of the questions found there are of the “please do my homework for me” type and I am plagued with enough emails of this sort (connected with my own exercises) to look for further hassle. They are however easy to spot and thus eliminate. For another, I suspect a majority of questions, while honest and deep enough, are often asked at the spur of the moment, i.e. without a preliminary search on a paper or online source, like Wikipedia or a textbook. E.g., a question about Bayes theorem that brought decent answers but not further than the Wikipedia entry on the topic. At one level, I would like to give in to temptation and to answer questions I feel I have a valid and informative answer. However, it does not seem like an efficient use of my time (read my books instead!) And also I am not completely convinced this fundamentally helps the persons who ask the questions in the first place. What may lie at the bottom of my unease with being involved in such forums is the sad fact that most questioners want answers without getting through the necessary steps of learning the bases and the background theory surrounding the question. While not being a teacher at heart, this approach gets against my views on learning (“Give a man a fish, &tc.”). Observations towards this view are that (a) many questioners are “one-shot” occurrences, i.e. are never seen again on the forum and (b) such questioners often fail to acknowledge answers that are not posted immediately, i.e. are not really interested in the debate surrounding the question they asked in the first place, only in someone solving their problem for them…

Anyway, this post is my very personal opinion on why I should not get involved with Q&A sites: it does not aim at criticising people asking or answering questions on Cross Validated, quite clearly, as some questions may lead to interesting research developments and as some answers are well-though, helpful, and informative. Just not my ballpark!

8 Responses to “[stack] overloaded, crossed and invalidated”

  1. […] in the past with superusers’ supervision. Hopefully, this promotion will not induce spending even more time on the forum… And increase my stress at reading often too often questions written by people […]

  2. Just to illustrate how hard I find getting off the site, I have now reached a 1000 level with 28 answers. I am glad my most rewarded answer was about uninformative priors.

  3. Part of the reason I enjoy the site so much is that I learn alot, not just from my own questions but by perusing other questions and answers. I agree it is addictive though, and I’m sure I have more to learn from the content of the site than you do!

  4. […] (still!) looking at questions on Cross Validated on Saturday morning, just before going out for a chilly run […]

  5. […] (rather foolishly) involved myself into providing an answer for Cross Validated: “Can the standard deviation of non-negative […]

  6. Mathoverflow is fortunate enough to have a very good group of moderators which enabled MO to establish a strongly professional culture. (Not that there aren’t some new problems arising from it.)

    But above all, MO, as the first scientific expansion of the SE network, has the luxury of math.SE — specifically designed for student-level questions.

    I think there’s also an upside of you not getting involved too much: you’ll give younger researchers a chance to show their expertise.

  7. Interestingly, Tim Gowers, in a recent blog about getting rid of journals that obviously attracted a lot of traffic, suggests using the grading system of mathoverflow.

  8. I participate at Cross validated and at Stack Overflow (mainly tag r) and my impression is that SO works well, while cross validated doesn’t.
    Maybe is the nature of the subject. Programming questions are more suited to these type of questions, while statistics questions are less suited? I don’t know, but I do like SO.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.