Re: [Axiom-developer] Literate Programming

[Stephen Wilson]

Hello,

I would like to add a few comments:

First, pamphlet files as we use them in the Axiom project, from my
understanding, are subject to possibly radical change.  So even though
I have enthusiasm for the pamphlet concept, I have reservations about
promoting the format for use by other projects -- under the assumption
that a pamphlet file is somehow tied to Axiom and its literate goals.
Of course, that does not preclude the use of a literate document type
for GCL.

Second, the general audience for GCL of course needs to be considered.
I like the idea of embedding a URL into a docstring, as it might be
integrable with Slime or some other development environment in a very
useful way.  IIRC, the ANSI spec gives the implementation a lot of
freedom regarding how docstrings are handled.

I have given only brief thought to what my `ideal' literate
environment might be for the specific task of writing Lisp code.  A
language specific tool offers the possibility of tailoring not only
the typesetting of the source code, but also the possibility of
`weaving' selected components of the literature into a doc string.
Within Lisp and an IDE, one might be able to envision many
interactions between the document on the one hand, and cross
referencing via embedded URLS in doc strings, appropos, who-calls,
etc.

So I guess there might be two notions involved here.  Documenting GCL
the Program, and allowing GCL to function as a member of a powerful
Lisp based literate programming tool chain.


Sincerely,
Steve