[Axiom-developer] Re: article "standard" header/footer

[Ralf Hemmecke]

Dear Tim,

> well, i'm WAS planning to do a document-and-cleanup cycle on
> axiom.sty so that (a) only the things that are used survive (there is
> cruft from history in there (b) it formats nicely so i can append it
> to stand-alone documents and (c) the macros are documented like
> everything else. i haven't done that yet due to time. i did merge a
> couple macros from other packages so that we could write axiom
> documents that only depend on 1 package and we could include that
> package as part of the .pamphlet file.

This sounds like hardcoding style files into axiom.sty. Who is going to
maintain them??? Even if axiom.sty is a literate program, it should not
be our task to maintain LaTeX stuff. A few commands is OK, but please
not all of those fancy style packages.

> i hate getting documents i can't format because the author hasn't
> included ALL of the tree of .sty files.

That is not a problem. Just put all necessary .sty files in a .zip 
archive and distribute them with Axiom. Whenever there is a new version 
of a package or just a new version of a whole LaTeX distribution, we 
might have to update. But to maintain the packages ourselves is just a 
waste of time. There are other people who can do it better. Our focus is 
on mathematics, right?

> plus noweb (WAS) planned to go away in favor of an axiom 
> implementation that had different features (like support for .booklet
>  files which use an optional URL syntax in chunk names). that may
> still happen but it is open for discussion.

I've once read something in the noweb documentation why Norman Ramsey 
has not introduced an "include" feature. One can put the files on the 
command line anyway and weave and tangle them together. Can I read 
somewhere about your reasons to introduce such extensions to noweb?
I also don't understand exactly what "booklet" should serve.

> sigh. old technology again. ok. propose something different for the 
> bibliography. i started on the bibtex thing (there is a
> bibtex.pamphlet the src/doc directory which would collect annotated
> bibtex references so we could keep them in one place and have a fully
> documented bibliography. if you suggest what you'd like to see we can
> figure out how to connect it to the bibtex.pamphlet and then spread
> the new machinery all over axiom.

I imagine the following here. We have one big bibtex database. If 
someone adds a contribution, he/she should use bibkeys from that 
database. Only if there is no bibentry there, new entries could be 
provided in an accompanying bibtex file to the persons project (which 
would eventually incorporated into the ONE.

So the wrapper script would write

\bibliographystyle{alpha}
\bibliography{axiom} % the ONE bibliography file
\IfFileExists{\projectname.bib}{\bibliography{\projectname}}{}

at the end of the document (followed by \end{document}).

In the above code piece \projectname expands to the name of the person's 
project.

There is still something to do in a Makefile to generate the .bib file 
from its .pamphlet and run bibtex on the whole stuff. But that is simple.

> -> )credits
>
> you'll see a list i maintain of all of the people who seem to have 
> contributed. having just done that now i see your name is not on that
>  list (i'll correct that in the next release). when time permits i
> run back over my email and add new people.

I don't remember that I've yet really contributed to Axiom. I haven't 
modified code at all. It's just taking part in the discussion.

Ciao
Ralf