Re: F2003 standard: Can Class(*) be used in generic proogramming?

Gary Scott wrote:
Simulate wrote:
The meaning of me is that if CLASS(*) can be used in generic
programming in a perfect way,why "Intelligent macros" is planed to
contained in original Fortran 2008 standard?

Looks like everything we knew is now obsolete...

DSLs Lead Development Paradigm Shift
eWeek (03/26/08) Taft, Darryl K.

The software development community needs to move beyond its use of static, procedural languages and frameworks and start using language-oriented programming.

"Language-oriented programming" ??

For the irony impaired amongst us, I'll state up front that this is completely tongue-in cheek, but aren't we doing language-oriented programming now? I mean, I use a language oriented towards my problem (which includes my knowledge of language) and I program in it.

OTOH, why use an existing language when you can create a new one every time you start a project? (And here's me thinking following OOP principles religiously led to code bloat. Hoo boy! :o)

So, when are DSL capabilities going to be added to Fortran? 2012? 2015? By the time the features have been agreed upon and compilers are available, some new paradigm(s) du jour will have supplanted this decades' set.

Actually, now I think about it, the next generation of "paradigm shifts" will probably be OOP as the young raised-on-DSLs hotshots rediscover it! ha ha.

So, it's official. I have graduated into the ranks of old-codger-dom. Sigh.

cheers,

paulv


ThoughtWorks senior application architect Neal Ford, speaking at TheServerSide Java Symposium on March 26, says domain-specific languages (DSL) are designed for specific tasks. Ford says ThoughtWorks colleague Ola Bini envisions a future stack of basic programming tools consisting of a "stable language" at the bottom level, with dynamic languages built on top of that, and DSLs added at the top layer. Ford says that DSLs improve the software development process by "eliminating noise," and that programmers experienced in dynamic languages tend to build DSLs on top of their low-level language. "Using DSLs evolves the way we build and use frameworks, escalating our abstraction levels closer to the problem domains and farther from implementation details," Ford says.

.

Relevant Pages

  • Re: LISP diploma project idea
    ... rather than solving problems in general-purpose programming languages, ... "If there are no appropriate DSLs for your problem, ... The man knows nothing of Lisp. ... to the language, it becomes increasingly difficult to add ...
    (comp.lang.lisp)
  • Re: Suggestions on which language to "try" learning.
    ... I'm just expecting that DSL designers will usually do even worse ... than programming language designers. ... substrate for implementing most DSLs in the form of libraries. ... programming language environment is lexical conventions. ...
    (comp.lang.functional)
  • Re: F2003 standard: Can Class(*) be used in generic proogramming?
    ... The software development community needs to move beyond its use of static, procedural languages and frameworks and start using language-oriented programming. ... ThoughtWorks senior application architect Neal Ford, speaking at TheServerSide Java Symposium on March 26, says domain-specific languages are designed for specific tasks. ... Ford says ThoughtWorks colleague Ola Bini envisions a future stack of basic programming tools consisting of a "stable language" at the bottom level, with dynamic languages built on top of that, and DSLs added at the top layer. ...
    (comp.lang.fortran)
  • Re: DirectX in HLA
    ... I guess that you have a great knowledge of DirectX ... > understanding by looking at them in assembly language... ... > actually represents, really, is a means to "undo" the OOP so ... > is NOT an "OOPL" (object-orientated programming language), ...
    (comp.lang.asm.x86)
  • Re: DirectX in HLA
    ... I guess that you have a great knowledge of DirectX ... > understanding by looking at them in assembly language... ... > actually represents, really, is a means to "undo" the OOP so ... > is NOT an "OOPL" (object-orientated programming language), ...
    (alt.lang.asm)