[Bioperl-l] BioPerl 1.6 RC1

[Chris Fields]

On Jan 4, 2009, at 9:51 AM, Gabriel Valiente wrote:

>> ...
>> I'm not sure what we should do at this point late in the game, but  
>> I would support pulling it and leaving it in trunk until a decision  
>> is made, then adding it back in a point release at a later point.   
>> I need to know something soon, though.  I already have RC1 out; RC2  
>> is to be tagged in the next day or two, and I would like to get a  
>> final release out in the next few weeks (as well as create 1.6  
>> branches for bioperl-db, bioperl-run, etc).
>
> As authors of the Bio::PhyloNetwork modules, we have made every  
> effort to conform to the Bio::Tree API. Nevertheless, it would be  
> best if active core developers could please have a closer look. In  
> any case, our motivation in publishing these modules as part of the  
> BioPerl distribution was to make them available to the large  
> community of BioPerl users and if you end up deciding to pull them  
> from the core distribution, they won't be that visible anymore.  
> Regarding renaming Bio::PhyloNetwork to Bio::Phylo::Network, I don't  
> like the idea very much because the Bio::PhyloNetwork modules do not  
> have much in common with Rutger's Bio::Phylo modules. Thanks,
>
> Gabriel


The question isn't whether they will be included within BioPerl per  
se, but specifically within the core modules of BioPerl (see below).   
Judging by your response I think we can include them for the 1.6  
release.  It might help if we set up bug reports to be passed on to  
you or others responsible (with the obvious constraint that fixes for  
1.6 won't break any API).  We can set something up on bugzilla for  
that; just let me know what email you want reports sent.

BioPerl 1.6 is to represent the last 'full' or old-style release.   
Regarding all modules automatically being in core: we have extensively  
discussed on the list the problem of code bloat in core.  We plan on  
splitting off specialized modules into subdistributions, similar to  
bioperl-db, etc, post 1.6.  This is something that most (all?) core  
devs appear to agree on.

Based on that I would say that post-1.6, unless the Bio::PhyloNetwork  
modules prove to be inextricably linked to a large portion of BioPerl  
classes, I could see these being included in a specialized bioperl- 
phylo or similar package, kept in a separate subversion repo just like  
the other bioperl-* distributions.  Won't be up to me alone though,  
and anyone is welcome to discuss this further on the list.

chris