My question was closed as too localized:


I'm not sure that "too localized" is a bad thing in this case. This question, in my mind, would have been useful for people who already own, have read, or have considered buying many of the "classics" (GoF, SICP, Dragon, Pragmatic Programmer, etc, etc, etc). I'm at fault for not stating my rational in my question, but I asked the question precisely because this would be useful for me. It's hard to sift through all the past very good books to find the good new books. Given the number of upvotes and stars I think it was useful to others too.

Furthermore, it can be useful for people who want to keep on top of new developments in the field.




Remember that the purpose of the Stack Exchange family of websites isn't to be a closed eco-system for the sole benefit of regular visitors or the question-asker, but to be a resource for people searching for answers. That is, to out-Experts-Exchange Experts Exchange.

If I'm searching for best programming books in, say, 2012, finding questions that only applied to 2010 isn't helpful and reinforces the problem people have with searching for answers: tons of out-of-date and irrelevant answers.

Because of this, questions need to be as timeless and universal as possible. When asking a question, you should be thinking "Can this question help others in a year from now?" Books like Code Complete are good recommendations because they have been proven to be relevant for years. It's not clear at all what books of 2010 are going to be all that useful even six months from now.

  • There are already generalist book threads, this one is not meant as a substitute to the best programming books, it's just meant as an answer to a different, more targeted question. – JDelage Dec 14 '10 at 8:58
  • @JDelage: I think you missed my point; I understand you're asking a different, more targeted question. I'm saying your question has been closed as being too localized because this site is meant for generalized questions that are not time limited. If you want to ask ephemeral questions like "best books of this year", you're in the wrong place. – user8 Dec 14 '10 at 9:08
  • Fair enough - I guess my point is that there doesn't seem to be an obvious reason to apply such a guiding principle with absolute consistency. I understand the ambition of not becoming a repository of obsolete knowledge, but I think a 1 yr time frame of book publishing isn't at risk of becoming irrelevant for a while. I also think that if a top 10 book was published in 2010, it would take several years to get to its "right" spot on a general list. That's a weakness of the SO engine IMHO. – JDelage Dec 14 '10 at 9:54
  • If I look at the "Most influential book..." thread on SO, and look at the pub dates, I see for the top 10 books: Dragon (2006 but initially published in 1977), Art of CP (1997 for 1st volume), GoF (1994), Mythical MM (1995), Refactoring (1999), C prog language (1988 but 2nd edition), Code Complete (2004, 1993 for 1st ed), Pragmatic Prog (1999). This makes it impossible to use such a list to discover new books, because even masterpieces cannot match the accumulated readership of those older books. – JDelage Dec 14 '10 at 10:07
  • @JDelage: I think what Mark is trying to say is that SE just isn't well-suited for discovering new, not-yet-widely-recognized books - until they're already well-known, they just won't have the votes! Try Amazon... – Shog9 Dec 14 '10 at 16:00
  • Maybe to ask about non classical books could be OK, but books released just for one year is too localized. Rarely books get the status of good reading in his release year. That question just get biased, unreliable and irrelevant answers. – Maniero Dec 14 '10 at 19:50
  • Ok, I'll get off my horse then... If anyone has any idea on how to phrase a question that would get to the good recent books, please go ahead. :-) – JDelage Dec 14 '10 at 20:12
  • @JDelage I still think you're approaching this site the wrong way. Answers that are only relevant for a specific moment in time are not useful for this site, regardless of how the question is asked. If you want to ask about books every programmer should read, that's fine (although I'm sure it's already been asked), but there is no question that's going to be acceptable here that would only provide "good recent books" as answers. – user8 Dec 14 '10 at 21:58
  • @Mark - isn't it a question of degree? What isn't time limited in the long run? – JDelage Dec 15 '10 at 10:04

You must log in to answer this question.

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