Computer Languages Overflow

The unprecedent improvement in hardware processing speed not only spawns a lot of new programming languages, but also resurrects a lot of dead ones. Needless to say, without those dual, quad or upcoming octo cores which offer higher and higer processing power, these languages are basically useless 5 years ago.

Well, maybe a little bit exaggerated but recently there are a hell lot of programming languages out there. Here I count distinct languages (not the implementations though) instead of family of language, ergo even OCaml is a dialect of ML, I count them separately.

Here is a list that I can think of:

• C
• C++
• Common Lisp
• Scheme
• OCaml
• Haskell
• Python
• Perl
• ML
• Ruby
• Visual Basic
  
• Parrot
• C#
• Ada
• Cobol
• Smalltalk
  
• Erlang
• Java
• Fortran
  

The list can go on and on, but these are what I have read about recently.

For some reason, a lot of people tend to take a particular language (usually it is the language they love) and hail it as _the_ language by having language shoot out, functional comparison, or comparing on the ground of aesthetic beauty. I have seen some paper heralding a particular language X and to show the comparison is 'fair', language Y and language Z are included.

Some points to ponder upon:

1. Each language is designed with a certain domain in mind, Fortran is mainly for numerical computing, while C is more of a low-level, system programming. No language can excel in everything.
2. When you have a hammer, all problems will look like nails. If you are used to programming in C with the weakly typed system, the strongly typed system in ML will most likely drive you to smash your monitor and chew on the debris. Each language has its own caveats.
   
3. When writing test programs in different languages, not many people can fully utilize the power of each language in question. Even if you invite so called experts to join the fun, the problem definition most probably will already give some languages an unfair advantage. Having the unfair advantage usually doesn't imply that language is superior, just the problem on hand matches its design domain (cf #1 above).
   

Having this rant after reading a paper on a multi-threading model built on top of scheme and was compared against the god-daddy of reliable and multithreading language, Erlang.

If you force a dog to meow, you get a lousy cat, and a very unhappy dog.

Luckily I am a cat person. Meow... Wait, someone calls me a wolf??? Hmm...