--00151757357cf18b880463b2100c
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

On Tue, Feb 24, 2009 at 9:14 AM, Pascal J. Bourguignon <
pjb / informatimago.com> wrote:

> If you were using CL (if more people were using CL), and Ruby were not
> invented, more bosses would let us use CL as a dynamic language, and
> indeed we wouldn't be forced to use Ruby (or worse, perl or python).
>
> Of course, you share the culpability with perl, PHP, Python, Java and
> so on.


So your argument is Common Lisp would win by default if it were the only
game in town?  That's not exactly a confidence boosting argument.

But I'm mostly justified in thinking this, by the authors of most of
> these languages themselves.  Some of these authors are just poor chaps
> not knowing better, but some other including Java and Ruby knew very
> well Lisp, and instead of choosing to help develop and spread lisp,
> they voluntarily chosed to create different and inferior languages,
> picking only parts of lisp, and grafting over their abortive creation
> some monstruous syntax.


Guy Steele, famous for his "drag them halfway to Lisp" quote about Java,
also spent his time making a better Lisp... in his case Scheme.

But this further demonstrates their pragmatism: since it's clear that Lisps
will not see widespread popularity, the next best thing is to look to Lisp
for good ideas and use those in the creation of new languages.

That's exactly what Matz did with Ruby.

The fact that these monstruous syntaxes view good in the eyes of the
> unwashed masses
>

The quality of a grammar isn't not inversely proportional to its size, and
indeed despite the complexity of the Ruby grammar people find it not only
easy to read, but easy to transform.

You may wish to place emphasis on the latter as Lisp's homoiconicity makes
it easier to conceptualize things like macros, but I prefer to place
emphasis on the former as I spend much more time reading code than
transforming it.

Your "unwashed masses" find it easier to read the code of languages with
complex grammars than they do to read Lisp.  While you seem to attribute
this to some failing on their part, I'd attribute it to Lisp's homoiconicity
leading to an excess of tokens, making the language syntactically noisy.
Perhaps the "unwashed masses" simply prefer their languages use a more
complex grammar to eliminate some of that syntactic noise.

This is perhaps the good that will come from these evils.  A lot of
> programmers are now exposed on the goodness of dynamic languages, and
> they may notice (or if not by themselves, *I* will try to make them
> notice in threads like this), that their fad language of the day, is
> actually some incomplete subset of Lisp, and then why not use the full
> and authentic thing?  Common Lisp is this way!
>

On the contrary, I think programmers who decide to learn a Lisp and try out
CL as their first are far more likely to be turned off to Lisp in general,
as opposed to those who try saner Lisps like Scheme.  That was certainly the
case with myself, and I didn't really come to like Lisp until I tried out
Scheme.

I think the failure of Lisp to gain widespread popularity can largely be
attributed to Common Lisp itself.

-- 
Tony Arcieri

--00151757357cf18b880463b2100c--