Is computer science progressive?

[bascule]

I see CS as something which is certainly "nichefying" (to use a Chris Anderson neologism)

 

In the niches, I see immense progress! I also see minimal adoption rates.

 

Perhaps the two best languages to grace the near-mainstream of late are Python and Ruby. Recently I've been working with another, more niche language known as Erlang which deals with the increasingly prevalent problems of concurrency and fault-tolerant distribution. I consider all of these languages to be immensely progressive.

 

However, I've been reading Steve Yegge's blog. He's renowned among the computer science community, works for Google, and has an excellent blog. Recently he "ported" the Ruby on Rails framework to JavaScript, soon after blogging about the requirements of what he thought would be the "Next Big Language" (NBL), a list of requirements which seemed to stipulate JavaScript/ECMAscript.

 

I see other parties tending towards ECMAscript as "NBL" too, most notably Adobe with their Flex and AIR platforms. Adobe has gone so far as to license their Tamarin JIT compiler for ECMAscript under an open source license.

 

Are niche languages really drawing people towards progressive programming patterns? Or will a new form of "blub" (to use a Paul Graham neologism) remain the dominant paradigm?

[pcollins]

You want to be a bit more clear on what you mean by 'progressive?'

[kanzure]

Computer science, in the sense of the study of computational abstraction and the mathematics of computation, is indeed an active area of development. However, I would be careful when saying "CS is programming languages," because technically that's not true unless you were wanting to make abstractions about all sorts of Turing-compatible languages, but I see that this is not what you are talking about.

 

I agree with pcollins, "What?"

 

- Bryan

[bascule]

You want to be a bit more clear on what you mean by 'progressive?'

 

The easiest way to define it is an overall trend of incorporating features into mainstream languages which make them less like Turing machines and more like Alonzo Church's Lambda Calculus.

 

In general I believe that imperative approaches are being increasingly augmented with functional ornamentation. The mass adoption of closures and other types of anonymous and higher order functions is a big win.

 

However, I would be careful when saying "CS is programming languages," because technically that's not true unless you were wanting to make abstractions about all sorts of Turing-compatible languages, but I see that this is not what you are talking about.

 

Not all languages are recursively enumerable. And even so... what? I understand the argument that the scope of CS extends beyond languages, namely algorithms which are language agnostic. But even so, this is far from the scope of the discussion.

[Fred56]

Your argument has so far also completely avoided the fact that CS also studies architecture of real-world machines and is partly an engineering discipline. The effect CS studies have had on chip makers is one example of this (not to mention all those communications equipment makers).