Another .NET Language 34
Wankers Anonymous writes "In an interview with David Simmons, CTO of SmallScript Corp., Learn about a new .NET language about to debut, the ins and outs of its creation, as well as some insider history behind the genesis of the .NET platform. "
well done... (Score:1)
Re:well done... (Score:5, Interesting)
The next version of the CLR adds support for generics and some other stuff.. language support will only get better. (I don't see Sun working on making the JVM better suited to languages other than Java...)
- Steve
Re:i smell a shill (Score:2)
Re:i smell a shill (Score:2)
The GPL is cool but I prefer a BSD style license - I've never put any restrictions on any of the code I've personally released.
Shareware, well, the term's not so common these days is it? I still like the idea of "pay for this software if you like it".
A model I've been toying with is one that works like shareware except that once a predefined amount of $$ is donated, soure code is released. Ransomware, I guess - sort of like what happened with Blender, but designed that way from the start.
A well documented API is better than access to source code in most cases anyway.
- Steve
Re:i smell a shill (Score:2)
Re:i smell a shill (Score:2)
I'm sure if I said something stupid about Linux I'd get a lot of people responding correcting me.. does that make them Linux shills?
- Steve
Re:i smell a shill (Score:2)
Re:well done... (Score:2, Interesting)
At the very least, they didn't even bother looking at Common Lisp - Franz [franz.com] took a look at the CLR several years ago and decided that it wasn't even worth the trouble. It is impractical to get CLOS (the Common Lisp Object System - multiple dispatch, multiple inheritance, generic functions, and completely dynamic (you can re-define a class at runtime, and all the instances, and subclasses, etc. will be converted according to either a default or user specified method the next time they are accessed)) implemented with any sort of efficiency. Closures and dynamically generated lambdas (anonymous functions that capture a lexical environment and plain vanilla anonymous functions, created at runtime, respectively) seem to present a similar sort of problem from what I know of the CLR (I understand that they'd have to be represented as objects, please correct me if I'm wrong).
I've heard other similar objections to the inadequacy of the CLR when it comes to dynamics languages, and overall I'm not terribly impressed with what Microsoft is doing or how it is going about it (the seemingly primary reason why .Net has/will have Scheme implementations from both Northwestern and PLT is because of rather large grants).
Re:well done... (Score:1, Insightful)
Effectively however these are criticisms that IL isn't high-level enough. Kind of missing the point; IL isn't designed to be high level - it's effectively an assembler code. Guess what? In assembler, you have to represent closures as objects. In assembler you can't implement CLOS primitives with much "efficiency" (number of instructions). CLR + IL define a platform which supports LISP + CLOS just as well as a raw CPU does; it JIT compiles so the speed of code is asymptotically more-or-less the same. And Allegro Common LISP is hardly a poster child for efficiency anyway.
One issue with all this is that the JIT phase can slow down certain very dynamic programs. Python.NET has issues in places because of this. But that's the price you pay for using a bytecode. Java would have the same problem if it attempted to be as powerful and dynamic as
However, I must agree that MS seem to have dropped the ball somewhat in their support of lexical closure. It is possible, but it's highly awkward, AIUI. However, these are not mainline features that Microsoft's market are demanding. What Microsoft is offering is a unified programming paradigm for Visual Basic, C++ and C#. Other languages are "nice-to-have", but hardly essential for 99% of the world's programming needs.
If I were going to design a language... (Score:2)
However, the genuine interest Sun has taken in making the Java runtime available for lots of *platforms* is pretty attractive, too. If the Mono Project [go-mono.org] doesn't make it, I'd have to go with Java, but I think Mono will eventually have pretty good coverage of the platforms of interest to me.
Re:well done... (Score:2)
On a Java VM simply any language, except perhaps C++ compiles and runs.
See http://grunge.cs.tu-berlin.de/~tolk/vmlanguages.h
The only limitation the Java VM has is that it supports only single inheritance. So if you like to map C++ to JBC you need to invent a new object format and work with delegations.
I find it interesting that suns VM is nearly unchanged since 8 years while Microsoft tried to beat SUN and tried to make it better in some way but need to change their own VM just a year after it was published.
angel'o'sphere
Re: well done... (Score:1)
> that was only posted two days ago....
I wonder how many people are deliberately submitting stories that have already been posted, to see if they can get a dupe.
I also wonder whether the editors are doing it deliberately, as some kind of joke (or attempt thereat).
new name (Score:1)
In MicroSoft Marketing speek shouldn't that be:
"The Fail.NET Initiative"
good work
Re:Lots of reasons why I want .NET to fail (Score:1)
Okay, this is really freaking me out... (Score:2, Funny)
David Simmons
(the one not associated with SmallScript Corp.)
Don't comment here (Score:1)
In case article gets /.ed... (Score:1)
dupe.NET [slashdot.org]