Re: Declarative constraints in practical terms

From: Bob Hairgrove <invalid_at_bigfoot.com>
Date: Sun, 26 Feb 2006 21:25:07 +0100
Message-ID: <f834021p5br3acj94ntuabocv32303c243_at_4ax.com>


On 26 Feb 2006 12:16:04 -0800, "Marshall Spight" <marshall.spight_at_gmail.com> wrote:

>Bob Hairgrove wrote:
>>
>> I don't see why you see C++ as "the downside"? ;) But I really don't
>> want this thread to degenerate into a Java vs. C++ advocacy war ... we
>> drift enough here as it is!
>
>I used C++ for about ten years, and Java for close to that. Both
>of those languages have excellent features, and I was a strong
>advocate for each in turn. I'm less of an advocate now.
>
>C++ lacks a module system, first class functions, sum types,
>and pattern matching. It lacks concurrency primitives,
>garbage collection, and, grievously, any kind of relation
>type. :-) Worst of all, it lacks any kind of safety guarantees.
>
>In the above, there are some features that could be
>added as libraries, but there are also some that can't,
>and there are some that ought to be part of the language
>proper.

The next release of the C++ standard is supposed to address most of these issues, e.g.:

What do you mean by "first class functions"?

>C++ has pointers and .h files, and #define. The
>metaprogramming language is completely unrelated
>to the regular language; it doesn't even use the same
>programming paradigm.

This is because C++ has to remain compatible with C in key areas. However, that will eventually be superceded as well, IMHO.

>OTOH, C++ has what seems to be the most powerful
>metaprogramming facilities ever. And it dominates
>software development, with only Java as a serious rival.
>
>Good points, bad points. Like everything else.
>
>http://www.deftcode.com/archives/every_language_war_ever.html

Funny! Thanks for the link.

>Marshall

--
Bob Hairgrove
NoSpamPlease_at_Home.com
Received on Sun Feb 26 2006 - 21:25:07 CET

Original text of this message