From: Siegfried Gonzi
Subject: Re: Bigloo success story: PHP compiler written based on Bigloo
Date: 
Message-ID: <40FCBDA1.BCDC11FE@kfunigraz.ac.at>
[do not take it with hostility but I think such things are also interesting for Common Lisp people:
the following thread also to comp.lang.lisp:

http://groups.google.com/groups?dq=&hl=en&lr=&ie=UTF-8&threadm=2a56f6a3.0407190916.c68386%40posting.google.com&prev=/groups%3Fq%3Dcomp.lang.scheme%26ie%3DUTF-8%26hl%3Den

]

Johnny wrote:

> ··········@hotmail.com wrote in message news:<····························@posting.google.com>...
> atever reasons, he chosed Bigloo over Common Lisp:
> >
> > ==
> >  I would actually have preferred to use Common
> > Lisp for this project, but Bigloo does give you maybe half of what you him> get with Common
> Lisp, as well as stuff that you usually don't get --
> > good C integration, portability, small executables, pattern matching,
> > and a built in (and extremely elegant!) lexer and parser combo.
> > ==
>
> In the toplevel, does bigloo only compile to bytecode?
> What are the ways in which Bigloo's C integration better than CMUCL's ?



Hi:

I do not know. I am not an expert in CMUCL. I do not think that Bigloo compiles to bytecode first as
the way OCaml for example does it.

I suggested to him that he should write an article covering the usefullness of Bigloo and he could
also work out the differences between Bigloo/Scheme and Common Lisp in that specific task. I think
Linux magazines or software engineering magazines are open for such stories.

However, he said that he is actually very busy but later  he will try to fullfill such an article.

Personally I think such an article will make up for a good reading.

I think the problem with CMUCL actually is that you cannot easily avoid the CMUCL distribution at
the end user site. If you carefully follow the slashdot link you will see that at the original PHP
compiler site there is no mentioning of Bigloo whatsoever (I didn't find any hint to Bigloo except
for the slashdot headline and I know the brain behind the project by accident from the Bigloo
list).  The only thing you need is gcc.

But you can contact him personally and ask  himwhether he will write a small  compilation of his
reasons of not  using Common Lisp; he may then post stuff on comp.lang.lisp.

I congratulate him not so much for his decision of using Bigloo/Scheme. I applaud to his
courageousness of chosing a low value perceived language which does not play any great role  (among
the many millions of C++ and Java programmers) in the theatre of programming languages.

I gather it depends on the individual wahtever tool he chooses.

Fensterbrett