How Nested Functions Work, part 2

language_fan foo at bar.com.invalid
Mon Sep 21 13:37:09 PDT 2009


Mon, 21 Sep 2009 15:29:13 -0400, Jeremie Pelletier thusly wrote:

> language_fan wrote:
>> Mon, 21 Sep 2009 14:01:27 -0400, bearophile thusly wrote:
>> 
>>> Justin Johansson:
>>>
>>>> I'd be interested to know how good D is for implementing
>>>> scripting/dynamic languages .. maybe that could change the odds?<
>>> You can surely implement Ruby or JS or other dynamic languages with
>>> D1. But I don't know how this can change the diffusion of D a lot.
>>>
>>> A possible way to use (and spread) D on the web is to compile D (with
>>> LDC) for NaCl: http://code.google.com/p/nativeclient/
>> 
>> The native client has a rather static policy model. I suppose the
>> future of mobile code is in dynamic languages like JVM & .NET based
>> ones, Actionscript and Javascript. Of course you all disagree, you are
>> free to do so. It just makes sense to me to use e.g. Java since it is
>> not only faster than D in many cases, but it has a nice security model
>> and a nice class loader (both of which are not perfect, though)
> 
> It really just comes down to what the language allows you to do. You can
> easily write portable code in any language given the right platform
> abstraction, some of them just have that abstraction in the language
> itself at the cost of losing most of your freedoms.
> 
> Java is mostly popular in academic contexts, it may have nice features
> but I don't see it getting popular among systems programmers anytime
> soon. C# and .NET have some nice features but just like Java they lack
> what systems languages provide: liberty.

If you look at the job markets in Europe, most jobs require knowledge of 
Java, since maybe 10% of programming is with systems programming 
languages (50% of that in the embedded market), 10% uses scripting 
languages, and the rest, 80%, uses Java/C#/ASP. So to you it might look 
like the commercial world is divided between systems programming 
languages and scripting languages. In reality the "academic toy 
languages" like Java and C# dominate the markets.

> That's what most people I met who praised CS around failed to grasp:
> there are no "wrong" languages, and no "better" languages. But when you
> spent 3+ years of your life studying something your ego can get the best
> of you when you're given something else :)

State of the art CS is not using Java anymore. Praising Java is not 
academic wistful dreaming.



More information about the Digitalmars-d mailing list