Re: [Axiom-developer] I would like to work

[Ralf Hemmecke]

> See my previous email about the 
>
> http://wiki.axiom-developer.org/book--main--1
>
> pamphlet files. All of these volumes can be updated directly through
> the web. We can create diffs of these against the versions in the
> current source distribution and vice versa. Maybe we could even
> hack together a good way to make periodic automatic commits from
> one to the other.
>
> Right now the version on the web is about 9 months old and was
> based on the tla archive by the same name. I don't know the status
> of that archive relative to the versions of similar files in the
> axiom--main--1 source distribution.
>
> I guess we should make some kind of effort to coordinate this
> and make sure everything is upto date and in sync before deciding
> how best to proceed with large scale edits. No?

Yes. Now we have Gold=stable, Silver=unstable, and several testing 
branches. See http://wiki.axiom-developer.org/AxiomSources .
The tutorial should be edited on one of the testing branches (either on
http://wiki.axiom-developer.org/axiom--test--1/ or on a specially 
created branch on SVN).

I must say, I would like to see axiom--test--1 kept in sync with a 
branch on our sourceforge archive. I have no idea whether that is easy 
to achive, but it would avoid questions like "Where is the most 
up-to-date source that I should edit?".

The tla branch book--main--1 on http://axiom-developer.org/archive/axiom 
should be merged to the testing branch on SVN and then *REMOVED*.
Look at book--main--1. It does not contain the other axiom sources and 
when I first found out, I was confused about the 
src/doc/bookvol1.pamphlet appearing in axiom--main--1 and the same file 
in book--main--1. Cannot we just have book-main--1 in a subdirectory of 
the ordinary Axiom/Gold/Silver/Testing? Nobody cares if the other 
volumes are not yet written, but it would be easier to know where to 
edit. And merging would proably also be easier.

Any other opinions about this?

We currently have more branches of Axiom than we have developers. And 
even worse, all these branches are quite badly documented what they are 
worth for.

Ralf