As I stated in the comments, IMHO the policy banning these kinds of questions is not good long term. Apparently this is relatively new policy.

The primary objections are just as true for other questions, and problem questions or answers can be handled just as they are for any other question. This issue was why "community wiki" was first established, and I think "community wiki" dealt with it fairly.

  • If the question is not generating good discussion, then close it as unconstructive.
  • If the question is question is generating answers that won't last, close it as too localized or edit it to make it more general.
  • If it is duplicate, close it as duplicate.
  • If an answer is just an Amazon link, downvote with a comment requesting explanation.
  • If an answer describes multiple items, with different attributes and reason the items are good, upvote it.
  • If a comment is arrognat or insulting flag it.
  • If the issue is one of reputation, then put limits on how much rep one can get from a single question. Or do so for questions with "off topic" or "list" or "subjective" tags.

I think I have made my point.

Beofett closed the "Best Book" question.

I disagree. Many of the SE sites have one (or several) Best Book question, and they uniformly provide great info.

Server Fault - 46 upvotes, 3400 views

Stack Overflow
- JavaScript - 122 upvotes, 60 upvotes for the accepted answer, 19,600 views
- PHP - 95 upvotes, 3 answers with > 10 upvotes, 42000 views

- Java - 7700 views, 195 upvotes, 318 upvotes for the accepted answer

Bicycles - Maintenance - 12 upvotes, 250 views

The policy about avoiding "no right answer" questions is generally sound, but is applied pedantically in this case. The "best books" question is going to improve the site, not detract from it.


2 Answers 2


No Stack site allows "what is the best/your favorite book about...?" questions. It's even in the faq. Real questions have answers and "what is the best book about...?" doesn't have a definitive answer.

You'll see that the trilogy (Stack Overflow / Super User / Server Fault) have a couple of these hanging around with big disclaimers that they are historical and you can't do that here any more. Quite simply, we figured out that these types of questions are really bad for stack sites, and we'd rather try some new mistakes on for size than repeat old ones.

Just because another kid got away with it doesn't mean you should do it.

All "what is your favorite X" or "what is the best book about X" questions detract from the site. Good questions share expertise. Amazon already has book recommendations -- there's absolutely no value in crowdsourcing them here. What we get is a lot of noise, chatty or one-line answers, and in return we've produced a list that is less valuable than a best seller list because at least best seller lists are constantly regenerated in order to remain up-to-date.

That just isn't what stack exchange is all about. We want explanations, not lists. We want "how" and "why", not one-liners. We aren't here to do surveys, we're here to help our fellow parents get things done. Most importantly, we're creating a living reference; the only thing worse than expertise that may go out-of-date quickly is a list that may go out-of-date quickly.

  • 1
    I disagree .. virtually none of the questions on Parenting are going to have a "definitive" answer. For instance, the question "How do you evaluate schools before a cross country move" certainly does not have one single, definitive answer. Yet it is a good question, and can certainly have a "best" answer.
    – tomjedrz
    Aug 17, 2011 at 4:51
  • 3
    Sure there can be a definitive answer to a subjective question. We may disagree on which answer is the definitive answer -- but for a good subjective question, there certainly is one. Stack sites are about sharing expertise. There's no expertise involved in having a favorite book. If it can be answered in one line, it's probably a bad fit for stack. The question about evaluating schools seeks explanation -- it's a "how" question, not one that just seeks to elicit a list -- that makes it a good question.
    – HedgeMage
    Aug 17, 2011 at 15:33
  • 1
    Allow me to re-emphasize .. there is no single definitive answer. Getting lots of bad answers is not an issue, as long as a few good answers get created as well. Every question get some bad, trite, or shallow answers. Downvote answers that don't contribute, upvote answers that do, and any issues go away.
    – tomjedrz
    Aug 17, 2011 at 22:22
  • 1
    Some best books questions even have single, definitive answers. The best book on C programming has it's own wikipedia page ... en.wikipedia.org/wiki/The_C_Programming_Language
    – tomjedrz
    Aug 17, 2011 at 22:25

There have been at least two other questions asked on this site that were similar, if not identical, to this question.

One of them I linked in my comment explaining my reasons for closing, as I felt that it would be a helpful resource to the asker.

The other question, which is practically identical was closed by the SE Community Manager. If you notice, I was one of the users supporting the question, yet it remained closed. Instead, the question I linked in my comment was created to replace it, and stands as an example of what is considered an acceptable format community wiki.

If you look at the examples you listed, most of them are closed for the exact same reason I closed the one on our site.

I believe closing of these types of questions reflects a relatively new viewpoint on policy. Several of those closed questions appear to have been open for years, before being closed in the last month or two (or as recently as two days ago).

While I had originally been in favor of questions of this type, I believe I see the reasoning behind decreeing them "not constructive". When you have a question with 20, 30, 40 or more answers, the answers cease to become useful, concise references, and instead become noise. Sure, there are some great answers in there, but how many people are going to read through all 50 "best Java book" answers? The end result is that the first few good answers will wind up on the top, and stay there. Meanwhile, any new answer starts off with a score of zero, and will wind up at the bottom of the list, and few, if any, people will make it far enough down through the list of answers to ever read the new contributions, let alone vote on them. Thus, potentially good answers are wasted, and useful information is presented in a format that is not useful to anyone.

If there are only a relatively small number of resources, it might be possible to maintain lists like this. However, it is my belief that there are far more books on parenting, and related to parenting, than there are on Java, JavaScript, or any other programming language or environment. The signal to noise ratio for any list of parenting books on our site would quickly cease to be useful (in my opinion).

  • I agree that it is a new viewpoint on policy, and I don't like it.
    – tomjedrz
    Aug 17, 2011 at 4:30
  • The fact that there are tons of books on parenting INCREASES the need for some "vetted" lists. The reason those questions are great is that, by looking at upvote counts, you can make an overwhelming list small. If I wanted a book on Java programming, I could search on Amazon and find a 100. But the search on programming.SE, and I get one.
    – tomjedrz
    Aug 17, 2011 at 4:34
  • @Tom, the spike you mention can be reduced by marking the question community wiki although the system would eventually do that automatically. Aug 17, 2011 at 5:32
  • 3
    @Tom The search on programming.SE does not return one book. It returns roughly 50. There is one that is much more highly voted than the others, but it was also published 3 years ago, and there may be better, more up-to-date books available now. In fact, it is possible that some of the books on page 2 are better now, but we'd never know because few people ever look at those answers.
    – user420
    Aug 17, 2011 at 11:57
  • @beofett ... yes, but there is a one, recommended by a whole bunch of people with some knowledge of the field. There may have been a better book published since then, but I can be pretty certain that the highly voted book will be good.
    – tomjedrz
    Aug 19, 2011 at 5:41
  • The "expiration" problem is a problem for any answer, because times change and the world changes. Salt, anyone?
    – tomjedrz
    Aug 19, 2011 at 5:46

You must log in to answer this question.

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