The plural of anecdote is not data (![[info]](http://l-stat.livejournal.com/img/userinfo.gif){kind=small} q_through_t) wrote,@ 2006-01-02 15:09:00 |
|
notable publications in computer science
I have been looking for some time now for a "greatest hits" list in computer science of influential publications and notable books.
And of course, Wikipedia's started in on an answer
This is part of a longer meditation that I've been thinking about a lot -- how do you answer the question of what the core texts of a field, of any field, are? I see this as one of the greatest challenges for learning a new subject, and perhaps one of the biggest challenges for both librarianship and for endeavors like Wikipedia. How do you provide a concise, complete introduction to a subject? How do you determine the truly core texts? Citation analysis is of limited applicability, and there aren't good enough tools to do it, anyway (Web of Science will never be able to measure the influence of something like The Mythical Man-Month or Godel, Escher, Bach). How, as a non-subject expert (a librarian, say) do you determine what the central sources for a field are? Finding more specialized information is in many ways easier; I can get you cutting edge research no problem, but how do I select the very best introductory book for, oh, say, object-oriented programming? Whole forests have been decimated for the subject -- how do I pick the best? In the end, if you're looking for introductory information, people's choices of what to check out will likely be determined more by factors like what's on the shelf and size of the book and interface-friendliness (or perhaps professorial recommendations) than truly by content -- but there are applications where this is an unacceptable approach ... such as writing Wikipedia articles, or making recommendations to undergrads. For Wikipedia in particular, I believe that the difficulty of finding central sources for non-specialists is one of the biggest issues for the site. In recent conversation with a friend who's also in the business, he expressed a criticism of Wikipedia that the articles tended to be not so much inaccurate as incomplete -- a much more insidious and intractable problem. I believe that this is but another expression of what could be termed the introductory-texts problem.
I have been looking for some time now for a "greatest hits" list in computer science of influential publications and notable books.
And of course, Wikipedia's started in on an answer
This is part of a longer meditation that I've been thinking about a lot -- how do you answer the question of what the core texts of a field, of any field, are? I see this as one of the greatest challenges for learning a new subject, and perhaps one of the biggest challenges for both librarianship and for endeavors like Wikipedia. How do you provide a concise, complete introduction to a subject? How do you determine the truly core texts? Citation analysis is of limited applicability, and there aren't good enough tools to do it, anyway (Web of Science will never be able to measure the influence of something like The Mythical Man-Month or Godel, Escher, Bach). How, as a non-subject expert (a librarian, say) do you determine what the central sources for a field are? Finding more specialized information is in many ways easier; I can get you cutting edge research no problem, but how do I select the very best introductory book for, oh, say, object-oriented programming? Whole forests have been decimated for the subject -- how do I pick the best? In the end, if you're looking for introductory information, people's choices of what to check out will likely be determined more by factors like what's on the shelf and size of the book and interface-friendliness (or perhaps professorial recommendations) than truly by content -- but there are applications where this is an unacceptable approach ... such as writing Wikipedia articles, or making recommendations to undergrads. For Wikipedia in particular, I believe that the difficulty of finding central sources for non-specialists is one of the biggest issues for the site. In recent conversation with a friend who's also in the business, he expressed a criticism of Wikipedia that the articles tended to be not so much inaccurate as incomplete -- a much more insidious and intractable problem. I believe that this is but another expression of what could be termed the introductory-texts problem.
