Meta Stack Exchange is where users like you discuss bugs, features, and support issues that affect the software powering all 158 Stack Exchange communities.

What is meta?
Here's how it works:
  1. Any Stack Exchange user can ask a question
  2. The community provides support, votes on ideas, and reports bugs
  3. Your voice helps shape the way Stack Exchange operates

I recently had a question closed due to a bad fit for Q&A

How does one request book recommendations for formal treatments of subject matter?

share|improve this question

migrated from Nov 26 '11 at 3:50

This question came from our site for professional and enthusiast programmers.

Thanks Peter. Unfortunately, I don't have an account. – noloader Nov 26 '11 at 3:49
Thanks random. From the link, I can't tell if asking for a recommendation is on-topic or off-topic. Next (for me) is the small leap to citations for answers. If book recommendations are subjective and not worthy, then neither are citations using the same book. It does not make sense to me. – jww Nov 26 '11 at 4:40
I'm giving you a sympathy upvote. Surely you got so many downvotes because people were saying "Don't ask about shopping recommendations, it's offtopic". Downvotes on meta mean disagreement. To downvoters I remind you this sort of question is ontopic in chat – The Unhandled Exception Nov 26 '11 at 13:54
The Unhandled Exception - thanks for the sympathy, but there's no need. Folks can keep subtracting points if they like - I really don't care. If a book recommendation is off topic, perhaps it should be stated clearly somewhere. Its not in the FAQ (I did read it before posting). Hence the reason I asked for the recommendation in the first place, and then asked 'how to request a recommendation' after the first request was closed. Jeff. – jww Nov 28 '11 at 14:37
It's right there in the FAQ, second entry, "What kind of questions should I not ask here?": 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 ... avoid asking subjective questions where every answer is equally valid: “What’s your favorite ______?” [or] there is no actual problem to be solved – The Unhandled Exception Nov 28 '11 at 16:11
I find it a bit ridiculous that it's considered 'subjective' to recommend books for serious situations. There are many resources out there, and it's extremely time consuming to filter through them all. I'd consider the question objective simply because it provides a target focus and has the possible effect of saving time/money/efforts. The discussions don't particularly last that long, as good books are usually agreed upon. If the site in general is built upon assisting developers, I'm sure that this is quite a big topic to be discussed in general. – TheCodingArt Apr 19 '13 at 3:47
"ask practical, answerable questions based on actual problems that you face" Practical problem: I don't know much about X, and must become an expert on X for work. Practical solution: A set of book / tutorial / blog recommendations from individual(s) who have a reviewable record of expertise on a technical site which I frequent. – Parthian Shot Jun 10 '15 at 16:27

What's to say a good book today is a good book tomorrow? Or the next day, or the next day? In 2007, The "ASP.NET tips and Tricks book" (that Jeff Atwood and others published) was a great ASP.NET book, but today, there could be better choices.

So that's one problem.

Secondly, it's somewhat subjective. There's no real 'fact' here. I may think Fooing widgets, 3rd edition is the BESTEST EVAR. You may think the 2nd edition of 'Baring Foos' is awesome-sauce. Either way, what happens when the OP picks one or the other? It's all opinion.

That's problem #2.

Finally, the answers could go on forever. There are 2,948 titles on Amazon for books about C# (according to an Amazon search on books for... wait for it... C#), so you could conceivably have 2,948 answers on the 'best book for C#'. Who's to say that one book is better than the others?

If you want a book recommendation, look at blogs or Amazon reviews. Stack Exchange is not the place for it.

share|improve this answer
"Secondly, it's somewhat subjective. There's no real 'fact' here. I may think Fooing widgets... " Good books should generally get upvotes, and bad books should get down votes? In the big picture, the voting system appears to be well adapted to this sort of thing. Perhaps I'm missing something. – jww Nov 28 '11 at 14:41
@noloader you are, people rarely downvote. It costs reputation, and that keeps people from downvoting. Also, Who's read even 1% of the books on C# out there? How can you give upvotes without knowing whether it's a good book? Finally, the 'one book per answer' means that after the first page, barely anything will get read, resulting in massive duplication, which then makes that question like every other forum post in the world: undecipherable and unhelpful. – George Stocker Nov 28 '11 at 14:52
Keep book reputation to Amazon, where it sells (and searched for). – Independent Jan 3 '12 at 11:35
I don't like the future proof requirement as it would render many answers invalid. Over the last decade we have seen memory latency grow larger than computation, render some precompute methods obsolete. Before FMA were added to x86, vectorization involved a loss in precision, now it can be the other way around. Likewise at one time recursion was preferred for code maintenance now its difficult to parallelize. In 2000 it was better to write a device SDK for Windows and then port it to other platforms, now its best to start with Linux. Things change and each post comes with a timestamp. – Mikhail Jan 6 '13 at 1:28
K&R, "The C++ Programming Language," Effective C++ all tend to remain good books... Many of these domains have agreed-upon best books, I'm skeptical we can't find "constructive" room for some questions. – djechlin May 17 '13 at 18:44
The whole point of asking the question is that, yes, there are thousands of books in Amazon on any given tech subject. However, of those books, the overwhelming majority of them are probably entirely useless, or at least poorly written, for their subject matter. Additionally, a list on SO can aggregate sources. Say you don't run across the industry-standard book on Amazon, but it can be found as a PDF on a .edu, or it's an O'Reilly book available online. If there were only one book on a given subject, there wouldn't be any point to asking the question to begin with. SO is the perfect place. – Parthian Shot Jun 10 '15 at 16:22
@ParthianShot The reason why we don't work well is that we don't handle subjective opinions well. We handle objective fact really well. Lists of 'good books' are by definition, subjective opinion. There are book lists on Stack Overflow that have done this; but their success is so few and far between it's far easier to say "don't do Book recommendations" than try to thread the camel through they eye of the needle. – George Stocker Jun 10 '15 at 17:59
@GeorgeStocker While it is generally true that, past a certain threshold what counts as a good book is subjective, what counts as a terrible or definitive / authoritative books tends not to be. The softwarerecs SE site seems to be doing pretty well. I feel like one for books would also. Every book's review calls it authoritative, but that won't last long on SO if everyone can edit. So, on SO, you'll know that CLRS is authoritative for algorithms, Gang of Four is the book for design patterns, K&R for C, C++ Primer for C++, etc. – Parthian Shot Jun 10 '15 at 18:51
@GeorgeStocker I suppose my counterpoint would be that, while some things are objective opinion, if everyone (or a significant enough percentage) in aggregate shares the same opinion, their concurrence of opinion is an objective reality which is useful to know, and probably reflects an underlying objective measure. It may not be measuring what they believe it's measuring, but (probabilistically speaking) if 90% of people find a book immensely useful and relatable to learn a given discipline, 9 out of 10 people selected at random from that same distribution should (on average) agree. – Parthian Shot Jun 10 '15 at 18:55
@GeorgeStocker Or, even shorter... Subjectivity at scale is objectivity. – Parthian Shot Jun 10 '15 at 18:55
@ParthianShot If you want to sway the larger Meta community, your best bet is to pose this as a question with your points. You and I disagree, but that doesn't mean you can't try to sway the larger community. – George Stocker Jun 10 '15 at 19:10
Which answer should the OP choose? I'd argue that a which-book(s) question should not have an answer chosen, but rather remain open. This is also true of some other questions, where multiple good solutions are given--choosing one answer in such a case does not, IMO, enrich the global body of knowledge. – Mathieu K. Jul 9 at 6:05
@MathieuK. For reasons that are completely historical now, we don't do well with questions where there is no objectively right answer. – George Stocker Jul 9 at 12:45

You can ask for recommendations in chat!

Your best bet is to first see if anyone has created a chat room for what you're looking for and then if not, create a new chat room first, something like Recommend me some books. Then stay in your room, and possibly draw some people in from active rooms (like the tavern, the lounge, casual chat) with a polite, discrete post, something like:

Hey everyone. I am looking for some book recommendations. If you have them, join me in [link to your room](http://url.of/your/room)

Remember that while chat is realtime it's also archived, has RSS feeds, and can act like a forum as well. So You can create a chat room and as long as you keep it active, you can return to it days later and see what was said when you were gone. So you don't have to worry about missing any recommendations.

FYI, This is what what Jeff Atwood himself did when he wanted recommendations for a new router.

Do not ask about shopping recommendations on the main site. They will be closed. It is offtopic.

share|improve this answer
More on the shopping recommedation. Good lord.... – jww Nov 28 '11 at 14:43
@noloader I don't understand that comment... what do you mean? – The Unhandled Exception Nov 28 '11 at 16:08

I don't know that you can. Shopping recommendations "In your opinion what should I get" questions are off-topic on the entirety of the Stack Exchange network.

share|improve this answer
"Shopping recommendations are off-topic...." - not necessarily true. There are a number of books which are available for free, with no shopping involved. For example, Handbook of Applied Cryptography ( – jww Nov 26 '11 at 4:01
Money doesn't have to change hands for it to still be a shopping request @nol – random Nov 26 '11 at 4:03
Random - its clearly not a shopping request. I'm not sure how you and Al Everett are arriving at shopping request. Its a request for a book(s) on specific topics of a subject matter. – jww Nov 26 '11 at 4:04
@noloader You're getting hung up on the word "shopping". Questions asking for recommendations for the best book on a topic are generally considered off topic. – joran Nov 26 '11 at 4:07
Joran - not to be argumentative, but I did not even use the word 'shopping'. It was brought up by others.… – jww Nov 26 '11 at 4:10
@noloader And they're often closed, as over half of the front-page results of that query were. "Shopping" isn't the point; replace the word with "looking". – Dave Newton Nov 26 '11 at 4:23
Dave - there's a big difference between looking and shopping. The first responder used the word shopping, not me. Anyway, I'm not clear why only half were closed; and how one would use citations in answers (since a citation is an implicit recommendation by a subject matter expert). Should answers with citations be removed? – jww Nov 26 '11 at 4:43
It's an infinite list problem, for one. For another, you've already chosen your solution: a book. Now you're just polling to see what other people think you should get. Fine, don't call it "shopping". "In your opinion what should I get?" questions are off-topic for the entirety of the Stack Exchange network. – Al E. Nov 26 '11 at 4:59
They are quite common in math, usually converted by moderators to community wiki (and I think they should be on-topic personally: one of the best ways to benefit from more experienced coders). – gnometorule Dec 5 '13 at 13:29

I started out believing that SE could support a recommend X, or Suggestions for a Building Architecture Book covering the User Experience Perspective style request, but after reading this and several of the linked questions, I have to conclude that the SE framework fails in this regard.

The reason I find the idea of such questions compelling, is my perception that there is a set of relevant canon for many fields. If you are in medicine then you've likely at least heard of Gray's Anatomy, the original, not the other one. In Architecture, then The Timeless Way of Building is something you've likely bumped into. This plays out across the majority of subjects I come across. The more I read about a field, the more material I see that references some subset of what came before. We even have the seminal work of Cyberpunk.

But canon is ultimately an illusion. Only this moment I learned of Bruce Bethke. No matter what I, nor even my group of wildly diverse and widely read friends have encountered and consider to be canon, and no matter how long ago a work was published or even how many times it has been referenced, someone could reasonably argue that the canon of X was merely founded on some W that came before. History has deep roots.

Additionally new books get written, this is not a problem only of having to add it to the list or even superseding the strengths of the existing works though. This means that canon changes. I'd easily add, as suggested in the question, How Buildings Learn to the list in the Architecture category. But why choose that one and not The Not So Big House? So now in answering these questions I have to consider when and where I should stop. Do I only include the first work I read on a subject? Should I include the possibly relevant books from the appendix of works I trust? What if a book in a completely unrelated field jogged some fundamental understanding in me but maybe not the next guy?* How do I answer if the book I consider definitive on subject X does not even acknowledge the existence of some concept Y that the OP mentioned?

To add to @GeorgeStocker's point about a good book today but maybe not tomorrow, it gets worse than just better material being in print. What happens when the 333rd edition of How to Win Friends and Influence People devotes 90% of its pages to how to use Facebook to pickup dates? What happens when the book I, or even the majority of the community, remember(s) is not the book being published when the next user clicks to see the Bestest of the Best of the Best reads on Underwater Basket Weaving?

In the end these questions can only ever be considered subjective, and therefore not a good fit for the SE format.

*See A Pattern Language for an Architecture book that made a significant impact on Software Engineering and to an extent Computer Science.

share|improve this answer
Book suggestions can still be passed on in answers to questions, as you can quote from books / articles and link to the books in question. Hell, StackExchange even has an Amazon affiliate link whenever a book link is posted which means you're doing your bit for the network with each link to amazon. However questions requesting book recommendations don't benefit the network. – JonW May 4 '12 at 21:55
"I started out believing that SE could support a recommend X ... I have to conclude that the SE framework fails in this regard.". Not entirely correct. While recommendations are off-topic for most sites, there is one that is specifically for Software Recommendations; see However, my feeling is that that "experiment" is not going that well. It only has 25 users with >1k reputation, which suggests that not many people are making recommendations. So may be that supports your "fail" assessment. – Stephen C Aug 17 '14 at 2:10
@StephenC: I suspect that most people with the skills to make a few recommendations on that site are, like me, not possessors of the breadth of knowledge required to contribute frequently. I would suggest that interpreting the stats of that site requires treating it as a special case. – Mathieu K. Jul 9 at 6:13

For other items, you can ask about features and specific scenarios to solicits facts to help you form your own opinion.

Instead of asking "Which sander should I buy?", you can ask What's the significance of orbit diameter in a random orbit sander?. Instead of asking "Which backpack should I buy?", you can ask What is the difference between external and internal-frame packs?

I can't see how you could do anything equivalent for books though. In the end it would still be a poll of peoples' opinions. Honestly the best thing to do would be search on Amazon and read the reviews.

share|improve this answer

You must log in to answer this question.

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