On Mon, 29 Oct 2007, Rainer Joswig wrote:
> In article <························@k79g2000hse.googlegroups.com>,
> llothar <·······@web.de> wrote:
>
>> Was looking for an implementation of a "trie" datastructure and run
>> over this code:
>>
>> http://www.accesscom.com/~darius/software/scheme-data-structures/trie.scm
>>
>> Wow, thats impressive close to the perl line noise. Isn't it?
>
> It's not that bad. But it is easily to see that Common Lisp
> code should look better by default. Scheme has some obfuscating
> built in.
>
Sure, if you look at RxRS only, but almost no scheme implementation
provides only whats specified in RxRS.
> Scheme code automatically looks a bit more ugly than
> necessary because it has only a primitive parameter
> list and no keyword parameters as in Common Lisp.
Well, I use Scheme a lot, but I would never in a larger project use a
scheme implementation which made it impossible to use
keyword parameters as in Common Lisp.
(many others scheme programmers doesn't seem to mind lack
of keyword arguments though, strangely enough, but there's
very little which prevents you from using keyword arguments)
> I also prefer to have objects (and structures) in
> the standard, instead of using local variables in closures.
>
> With R6RS is could be written differently, but then R6RS does
> not really look as an improvement.
>
No reason to use R6RS to write things differently.
> CL, ugly as it is sometimes, allows you to write some decent
> looking software.
>
So does scheme.