RE: [Axiom-developer] Issues with documentation revisions

[C Y]

--- Bill Page <address@hidden> wrote:

> There is at least one better tool than noweb: Leo
> 
> http://webpages.charter.net/edreamleo/front.html
> 
> I believe that many of the design features of Leo would be beneficial
> to the Axiom project.

I took one run at Leo and bounced - I'm willing to try again, though. 
I may have also thought of a specific example which might be a good
test/showcase.

In outputing TeX instead of ASCII art, there are any number of specific
issues and design decisions that might need to be made on a per-topic
basis - the use of the SI style package for correct formatting of units
being once case I am aware of, another is the use of chemical packages
for LaTeX when implementing chemistry related pamphlets. 
Unfortunately, neither a centralized mechanism for all TeX style
outputs nor spreading these things over many pamphlets is ideal - the
centralized mechanism separates the TeX logic from the conceptual work
that it is linked to, and the diluted method makes a specific display
bug hard to track without knowing exactly what part of the codebase it
relates to.

Leo (if I understand it's concepts) could solve both of these problems.
 If the tex output code is in one or several discrete chuncks, and its
corresponding documentation is also organized this way, then it would
be possible using Leo to assemble either a document with all tex output
code in it (for output debugging) or a document with all code
pertaining to (say) the unit concept, which is more logical for
authoring and updating in a non-debugging context.  Both of these
outputs could be consistent, latexable pamphlet files, just generated
based on the focus of interest at the time of the query.

Bill, is that a reasonable use case for Leo+Axiom?  You are far more
fluent with it than I am.

Cheers,
CY

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com