Re: feature freeze -> 3.0 release

[David Bateman]

John W. Eaton wrote:
> Other than bugs that should be fixed and a probable switch to GPLv3,
> are there any major obstacles to a 3.0 release?  Although I'm sure we
> can all point out a few places where things could be improved,
> features added, etc., I think the current sources are in reasonably
> good shape and much better than any 2.1.x version of Octave.  So I'd
> like to propose a feature freeze (though I would consider the addition
> of new independent functions that don't require changes to existing
> code) followed by a release in the next three or four weeks.
> 
> Comments?
> 
> jwe
> 

The only thing I would have liked would have been saving of function
handles to matlab files, as I coded the loading of them. However, as
this part of the file format is not documented its likely to be painful
getting it right, and might break other things in the matlab file
format, so forget for now...

There were the edit.m, textread, etc functions that I also thought to
move from octave-forge, but these are independent pieces of code and so
shouldn't prevent a release...

The only thing I really really want to see fixed before 3.0 is that at
least one distribution is happy with the octave package manager. After
the changes to pkg.m are in can we have at least one test release prior
to a full release? In that case might we have a 2.9.15 is a very short
time if I get a patch together for pkg.m that Orion is happy with? If so
I'd put off an octave-forge release for 2.9.14 to get pkg right and
release octave-forge based on 2.9.15.

D.