Dec
5
A Sermon on Programming Languages, from Jason Schroeder
December 5, 2006 |
Programming languages are cool. Love them. Like them. The most important languages are Prolog, ML, and Haskell. The best ideas are represented there. The problem being that people imagine that somehow a language makes programming easier. It does not. But the hope remains. The major abstractions are function calls, garbage collection, exceptions and objects. Once one leaves the safety of the fire, there are language features like monitors, first order types, first order functions, functors/module systems, dynamic scoping, unification (Prolog only), annotations and then one gets into the really rare toy features. Way back when, garbage collection was considered a toy feature. (The chip provides the number, operation and memory abstractions.) The language exists for the writer, not the computer. So many language features are written with the expectation that somehow the author's task is made easier. In order to do this, the features have to be "different" (orthogonal) rather than just syntactic sugar. Reducing keystrokes is nice in the beginning but the real power is doing stuff that simple keystroke reduction cannot do! Like creating functions at runtime. Or creating interconnected packages of classes and objects at runtime by inserting meta-parameters of types and data. (Macros on steroids.) Features like that permit one to not have to create all the support code to make it happen in your own way. One can look at the source code and go "oh, this is what is happening" as the language demarcates the parts that truly vary and the parts that are simply different! A handy library can do all this work. But then the library better be a standard library so that it is not just some other pile-o-junk with suggestive function names. So the difference between a well developed library and a language feature is minimal. In fact, the only language feature that cannot be implemented as a function is short-circuiting AND and OR… C++ has too many language features. (Arguments ensue). And C has so few it is amazing that is all one really needs. But without their standard libraries for string manipulation, I/O, POSIX compliance, etc., these languages would be nothing but curios. I really want to spend my time writing in ML, the funkiest. It is has great libraries but not enough to keep me from having to write library wrappers. Haskell provides the highest meta-abstractions I know of but it is not widely used so its practicality is lessened (but oh, one can show how smart one is by orchestrating amazing meta-programming abstractions to make the compiler write the program for you). Prolog is prolog, if the problem is expressible in prolog there is no reason not to use it, the debate rests on whether anything useful is expressible in Prolog… In conclusion, the best advice find tight library functionality and stay close to C/C++/Java. C# is more marketing than stable. Behemoth libraries get spooky. If you are trying to write web services, then Ruby-On-Rails is way cool. But if you are hankering for a real server, then C/C++/Java are mandatory. Java servers are freakish mounds of code to the uninitiated. A good old C server was demonstrable in 200 lines but the expectation of servers has increased infinitely since the Internet Bubble!
Sam Humbert adds:
A good, practical intro to Haskell has been rolling out on Mark Chu-Carroll's blog. The tone of it isn't "for dummies," but it's clear and direct. Even I can make some sense of it. Haskell is arguably the most important of the newer (i.e., non-Lisp) "functional languages". It's kinda-sorta like Python visually, and has some similarities to R, which has itself been accused of being a functional language.
Comments
Archives
- May 2013
- April 2013
- March 2013
- February 2013
- January 2013
- December 2012
- November 2012
- October 2012
- September 2012
- August 2012
- July 2012
- June 2012
- May 2012
- April 2012
- March 2012
- February 2012
- January 2012
- December 2011
- November 2011
- October 2011
- September 2011
- August 2011
- July 2011
- June 2011
- May 2011
- April 2011
- March 2011
- February 2011
- January 2011
- December 2010
- November 2010
- October 2010
- September 2010
- August 2010
- July 2010
- June 2010
- May 2010
- April 2010
- March 2010
- February 2010
- January 2010
- December 2009
- November 2009
- October 2009
- September 2009
- August 2009
- July 2009
- June 2009
- May 2009
- April 2009
- March 2009
- February 2009
- January 2009
- December 2008
- November 2008
- October 2008
- September 2008
- August 2008
- July 2008
- June 2008
- May 2008
- April 2008
- March 2008
- February 2008
- January 2008
- December 2007
- November 2007
- October 2007
- September 2007
- August 2007
- July 2007
- June 2007
- May 2007
- April 2007
- March 2007
- February 2007
- January 2007
- December 2006
- November 2006
- October 2006
- September 2006
- August 2006
- Older Archives
Resources & Links
- The Letters Prize
- Pre-2007 Victor Niederhoffer Posts
- Vic’s NYC Junto
- Reading List
- Programming in 60 Seconds
- The Objectivist Center
- Foundation for Economic Education
- Tigerchess
- Dick Sears' G.T. Index
- Pre-2007 Daily Speculations
- Laurel & Vics' Worldly Investor Articles