[Axiom-developer] Re: patches and fixes

[Gabriel Dos Reis]

root <address@hidden> writes:

[...]

| There are a large number of "fixes" being applied which appear to
| have nothing to do with autotools, such as the sttaylor fix and the
| sage change. Worse yet, it appears that the sage connection won't work 
| with gold so sage users cannot get the latest axiom "gold" release.

Which "Sage connection" patch do you have in mind?

| Careful thought will show that this is going to be a major future problem. 
| 
| Consider what happens when the time comes for build-improvements to be
| merged with silver. There will be hundreds of differences, many of which
| will have nothing to do with the build process. 
| 
| Each difference has to be fully documented (which I don't see anywhere). 
| Each difference has to be tested (and I don't see any test cases anywhere).
| Each difference has to be merged and then tested on many platforms 
| (pty fixes are going to take a lot of time and testing). 
| 
| I'm impressed with the rate of change of fixes and I don't want to 
| slow the motion but I think it would be wise to consider submitting
| documented, tested fixes against gold if they do not require autotools.

I'm quite impressed you're worried about possible forks, as you have
stated at multiple times in the past that "it is open source and
everybody is free to take the directions they want".

That said, I'm not a big fan of forks -- on average, they are wsted efforts
and community fractures.  But I'm not worried too much,
because build-improvements is, well, about improving the current build
process.  There are a number of issues that need to be fixed on
build-improvements.  Patches against silver for such things are less
relevant. (Though I don't think sage connection is one of them).

Also, please keep in mind that people will be tempted to apply fixes
where they see activity.  That is in the nature of open source (and
software development i general).  You can hardly blame them.

-- Gaby