views:

88

answers:

3

Well, I know it's not all about speed and memory usage.

But I would like to know what you think will happen to most of the high-level programming languages. As far as I know, Java is much faster than it was in the past, what about python, php etc.

+2  A: 

Speed has more to with Moore's law than the language itself. So if you are looking in absolute terms, you'll get more bangs for more buck by just upgrading your machine on a regular basis.

In terms of memory footprint, I expect most languages to continue gathering functionality thus increasing their footprint.

jldupont
Depends if speed is defined as latency or throughput. Moore's law isn't going to do anything about the speed of light which dictates minimum latency.You touch on this in the second point; we trade off memory (capacity) for time whenever possible because time is more finite than memory.That being said, +1, I agree with both your points.
Gazzonyx
@Gazzonyx: I like your thinking i.e. the physics angle. I agree 100% with you. I was thinking more in terms of *throughout*, dimension in which Moore's Law does have an influence: more processing units in parallel pack in ever denser packages.
jldupont
+1  A: 

High level programming languages will continue to get more abstractions that make it easier for developers to specificy what they want a computer to do, without having to get their hands dirty with difficult underlying details that a compiler and/or runtime system is better at optimizing anyway than any developer might be able to do a priori.

Think about:

  • support for multi-threaded execution (like Parallel Extentions in latest .NET)
  • specifying structure and functional outcome instead of manually telling computer exactly how and in what order to shuffle which sets of bits around

Those kinds of things.

peSHIr
A: 

Parallelism, given that increasing the number of processing units (cores) is the principal way of gaining speed nowadays. To make it manageable to humans, software transactional memory seems to be one of the most promising real-world solutions.

Joonas Pulakka