[Bioperl-l] A new name for Bio::Moose?
cjm at berkeleybop.org
Tue Jul 28 01:53:09 UTC 2009
On Jul 27, 2009, at 9:23 AM, Chris Fields wrote:
> Herein lies the namespace problem. What namespace would we inhabit
> with Bioperl 2? Bio2::*?
> We can't place anything in Bio w/o running into potential problems
> re: namespace collisions, both locally and on CPAN (yes, I do intend
> on releasing to CPAN at some point). Bio* primary maintainer is
> BIOPERLML, so releasing a Bio::Annotation::Foo on CPAN would
> potentially collide with any BioPerl Bio::Annotation::Foo as an
> unauthorized release. If all code is sequestered in
> Bio::Moose::Annotation::Foo, we don't have a problem beyond it
> requiring more typing, hence the request for a name change
> (preferrably something short) ;>
> Anyway, for all intents and purposes, if everything works out it
> will very likely become bioperl 2.0, and the old Bio::Moose (or
> whatever) will be deprecated on CPAN in favor of this. Until then,
> it is simply a side project to explore the feasibility of moving to
> a Moose-based framework. We can then work on reimplementing the
> various split-off Bio::* we have been discussing elsewhere.
> One main side benefit of doing this: it's incredibly freeing. I'm
> finding numerous places where we could optimize things in BioPerl,
> even in the case should this not pan out. For instance, memoize and
> clone locations instead of creating everything de novo. Annotations
> and AnnotationCollections could be lightened considerably if the
> proper framework were provided, something like Data::Stag or XPath.
As the author of Data::Stag and one-time proponent of less object-y
more xml-y modeling I'd say instead go for something a bit more
moosey. I have a feeling that it is possible to use the meta-level
features of Moose to come up with a solution for annotations that
would turn out to be (a) flexible, extensible and intuitive, with the
right benefits from both strong and weak typing (b) mind-bogglingly
complex. Not sure which. But if Bio::Moose is just for experimentation
just now might be worth trying.
> Also, I have found other instances that will be beneficial to a
> perl6-based implementation (roles vs interfaces primarily). I'll be
> posting here and blogging about these along the way.
> On Jul 27, 2009, at 10:25 AM, Robert Buels wrote:
>> Bioperl 2. If the Moose business doesn't work out, throw away the
>> moose code and do something different. But I think it would be
>> good to christen it as the embryonic bioperl 2 to get some momentum
>> around it.
>> Chris Fields wrote:
>>> Pushed by a recent suggestion by Robert, I am considering changing
>>> the name of the Bio::Moose project to something simpler. I would
>>> like to steer away from naming this directly after the
>>> implementation and have something simpler namespace-wise.
>>> I have thought of 'Alces' (the genus name for moose), which
>>> indicates both the Bio aspect and the implementation in an more
>>> indirect way (and is a bit shorter). However, I would like to
>>> solicit suggestions for alternatives. The shorter the better, and
>>> the 'winner' will receive a free beverage or so (on me) should we
>>> Bioperl-l mailing list
>>> Bioperl-l at lists.open-bio.org
>> Robert Buels
>> Bioinformatics Analyst, Sol Genomics Network
>> Boyce Thompson Institute for Plant Research
>> Tower Rd
>> Ithaca, NY 14853
>> Tel: 503-889-8539
>> rmb32 at cornell.edu
> Bioperl-l mailing list
> Bioperl-l at lists.open-bio.org
More information about the Bioperl-l