Do we want "literature recommendations" and similar "list/subjective questions"?

Question

Our very first question was https://crypto.stackexchange.com/questions/1/crypto-class-in-college-but-looking-to-take-it-to-the-next-level, asking about book recommendations.

By design, this is not a question which permits a unique correct answer, like said in the FAQ:

You should only ask practical, answerable questions based on actual problems that you face. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

[...] To prevent your question from being flagged and possibly removed, avoid asking subjective questions where …

• every answer is equally valid: “What’s your favorite ______?”

This is a typical What's your favorite book?-question. These are not necessarily bad, but we should think about whether we want these here. (We can change the FAQ.)

For comparison, User Experience seems to have decided that open-ended questions are allowed there.

Answer 1

You don't want this here.

Cryptography is such an awesome fit for a Stack Exchange site. You really don't want to reach for this low-hanging fruit because of something "user interface design" allows — apples-and-oranges. UX's opinion-discussions are held within the context of evaluating aesthetics. That's not in the same scope of your garden-variety "what books should I read" questions. And frankly… their "what do you think about…" discussions are the low-hanging fruit of their site; It's certainly nothing to aspire to or emulate.

I just closed this question, but closed is not deleted. Please continue this discussion. If the community truly feels this is a good fit for this site, posts can be reopened. But I hope the community sees otherwise, so please read on —

It's difficult to explain why, exactly, these "let's make a list" question lessens the value of a site until you've seen it over-and-over again on 57 sites. There's nothing inherently wrong with asking an "I'd like to hear what everyone thinks about…" question; It's just that we specifically forgo asking these types of questions because they're simply not good fit for this type of Q&A site.

Asking a good question is hard. Cryptography is hard. But that's what makes this site so special. With discipline, you can keep this site focused on those long-tailed question of specific expertise that a Cryptography community will find so intriguing… instead of launching this site to the same uninspired questions that have been asked 100 times before on every other site on that subject.

This Cryptography site is targeted at experts. When a top expert arrives on your site, you want them to see a list of questions that says "Wow! this the site for me!"

Big List^™ questions, on the other hand, have little to do with expertise. Asking everyone to contribute to a large bucket of answers means that it stops being a question of specific expertise and becomes a "poll" of the community. It doesn't even matter what the answer are. Answers start piling on; a disproportionate amount of voting accumulates; There's not even an expectation that any one answer will be better than any other. People vote on what they recognize. That's where the so-called "expertise" breaks down.

People who visit this site will emulate what they see. You're home page is your design. Questions about your "favorite website" will inspire yet another "favorite book" and "favorite blog" post, followed quickly by "favorite podcast" then "must-have software", "best tool", "most egregious security laps", "funniest joke", "most annoying coworker" … on and on. It's best just to avoid this variety of question altogether.

Don't aspire to the same trash you see on every other phpBB forum on the Internet. Don't aspire to the low-hanging fruit. You don't need it.

Answer 2

I'm going to say no, however, I could see the argument for community wiki.

I'm a user of Stack Overflow since the existence of actual community wiki question boxes, which was some time a go now in StackYears. Whilst I'm fully glad that the feature is mostly disabled except at the discretion of the moderators, the “list of X” book questions on Stack Overflow are valuable resources.

On the counter side - earning reputation on these easy questions distorts the value of reputation. Also, consider that it is highly likely everyone will upvote these sorts of questions, pushing them to the top of the highest voted question pile. Do we really want that visible to prospective members? I'm not sure we want that - so another question might be would CW prevent them ending up there? Also, allowing a single list of X question will add additional close-work as other list of questions become debated - it's hard to be clear exactly what list ofs are allowed and not.

Finally, “list of X” is part of the not constructive close reason. I think the SE network wants to discourage these questions as a whole.

So overall, I can see the arguments for wanting these questions and the value they provide, but on balance I think the disadvantages outweigh the positives.

Answer 3

I think it's a mistake to close questions of this nature…

1. Closing questions alienates a (typically) new user, and
2. there are literally hundreds of questions like this on other SE sites, often some of the most upvoted and (to me) some of the most informative.

This site is supposed to be Q&A, I know, but someone people would simply like to see a list of free online books and no other site really aggregates them in one place.

Answer 4

I think questions like this should be allowed but structured.

There is a problem: Textbooks are expensive; not just the cost of one but the time invested in reading one. There are lots and lots of cryptography textbooks. And some are frankly not that good.

We have a solution: The crypto.se community is in a place to offer good advice and help steer people toward a suitable match. It is hard to find advice elsewhere: on Amazon, most crypto books have under 5 ratings.

What I propose is one question asked by a moderator for a list of textbooks: each answer lists only one book, anyone can contribute to describing that book (pros/cons, who the audience is, if its didactic or a reference, etc.) (I think we can do this with a community wiki for each answer?), and users can express their preferences by upvoting/downvoting each book. Any future questions will be closed and referred to this one question.

This proposal is essentially what was done in the closed thread discussed here. I am not claiming novelty for this structure of Q&A, just proposing that we implement it.

If the model works, it could be applied sparingly to other requests for preferences that come up (academic conferences, univeristy programs, magazines, blogs/podcasts, etc.). Sparingly can be defined as only after a discussion on meta if we want such a Q&A for a particular topic.

Answer 5

Eventually we're going to want them, and eventually they'll be useful. But in the beta period, I'm going to say that they're not a good idea, nor will they be useful with the initial set of people. When we have a larger audience, there will be folks who can point out some obscure text, but for now, those would end up at the bottom of the list and would be ignored.