Blog

4Mar
2004
The Value of "Teach Yourself" Books

Sean Corfield blogged a reference to a rant against "Teach Yourself X in N Minutes|Hours|Days" books (see http://www.corfield.org/blog/past/2004_03.html). Sean (and Brian LeRoux) seem to agree with the rant's basic premise, and there is some validity to their concerns (I agree, some "Teach Yourself" books really are fundamentally flawed). But, as the author of a couple of these books I think they have missed the real purpose and value proposition of crash-course simplified just-the-basics getting-started type titles, and I state as such in a comment on Sean's entry.

Comments (2)



  • Steven Ringo

    I agree with you Ben. Whilst the article has merit and yes it does take years to master programming. But if you are already say an accomplished C++ programmer, then it may take you only 21 days to become quite proficient in Java. Also you gotta start somewhere! It if it weren't for all these "dummies/x days" et al books out there I would still be reading awful help manuals!

    Bit strange though they way he says: "Learn at least a half dozen programming languages. Include one language that supports class abstractions (like Java or C++), one that supports functional abstraction (like Lisp or ML), one that supports syntactic abstraction (like Lisp), one that supports declarative specifications (like Prolog or C++ templates), one that supports coroutines (like Icon or Scheme), and one that supports parallelism (like Sisal)."

    I mean - who's ever even heard of Sisal, Icon, Scheme? Didn't Lisp (as a mainstreamer) die a long time ago?

    I would far rather recommend HTML, XSL, Java, Javascript, and get the same benefit. I think this is one geek thats taken his geekiness just a step too far... :-)

    #1Posted by Steven Ringo | Mar 4, 2004, 03:47 PM
  • seancorfield

    Well, Lisp and Scheme are still in solid use. Sisal and Icon I'm not familiar with - I'd point people toward Occam as a parallel language.

    Of the languages you advocate, HTML isn't really a language, XSL is a good example of declarative programming, Java is already on his list and JavaScript isn't really different enough to contribute.

    A broad spectrum of languages really does help expand your problem-solving ability because it gives you different perspectives.

    I think Ben and I are in broad agreement here: some of the TY books are fundamentally flawed because, in my opinion, a number of authors are just trying to make a fast buck (witness all the data structures / algorithms books for C++ and Java that were weak rewrites of C books). Not all TY books are bad but it is a much-maligned genre for a reason...

    #2Posted by seancorfield | Mar 5, 2004, 02:38 AM