[BioRuby] What is the bar for releasing biogems?

[Raoul Bonnal]

Hi Ben,
I agree with Pjotr, post and publish the gems.
>From my experience is better to publish/release on rubygems when the gems is
enough stable to be used by others otherwise you will spend a lot of time on
fixing their issues.
The gem doesn't need a huge documentation, short and clear is better, some
comment in the code and that's it.
Find an effective description for your gem, every time I look for gems I
read/search the short description.
Another rule I'm trying to follow is to define/reuse a namespace which is
pluggable into BioRuby -but this is my personal point of view-.

Don't forget to drop an email here :-)

We need to restart IRC meetings...
 

On 19/01/12 09.23, "Pjotr Prins" <pjotr.public14 at thebird.nl> wrote:

> Hi Ben,
> 
> On Thu, Jan 19, 2012 at 05:58:01PM +1000, Ben Woodcroft wrote:
>> Hi there,
>> 
>> I was hoping for some advice from the list about policy on releasing
>> biogems. I have a few (2 or 3) biogems on my computer which:
>> * Solve a discrete problem (in my case, wrapping around an underlying
>> bioiformatic program and parsing the result)
>> * At least a little bit unit tested
>> * Are bioinformatics-related
>> 
>> However, they are also:
>> * Not fantastic leaps forward - they don't solve big problems
>> * Probably limited to a small audience, since the programs themselves would
>> be of little use outside the (not large) field (bioinformatics of protein
>> sub-cellular localisation in apicomplexan parasites).
>> 
>> I find having a biogem is a convenient mechanism. But I don't want to
>> release code that is of no use to anyone else (or any more of it..),
>> particularly as I cannot know whether I'll continue to use the code once my
>> PhD is done. Should I release the gems?
> 
> Yes. Please post them,
> 
> Do not worry about interest, take up, or quality. That is up to the
> people who ultimately take an interest in your gems. If there are
> issues they may approach you, or become maintainers themselves.
> 
> With the growth of number of gems we will find ways to handle
> presentation and quality issues. First on our list is an automated
> testing frame work - which will show on the site the gems that pass
> their tests.  Next we will create a subsection for 'development'
> and/or 'unstable' gems. That way normal users can feel safe in using
> the tested and stable gems. The meta packages (bio-core etc) already
> have an implicit policy it that way. Anyone should be able to install
> bio-core gems.
> 
> In other words, don't worry. Release early and often. That is the OSS
> adagio, that is what http://biogems.info/ is about.
> 
> Pj.
> _______________________________________________
> BioRuby Project - http://www.bioruby.org/
> BioRuby mailing list
> BioRuby at lists.open-bio.org
> http://lists.open-bio.org/mailman/listinfo/bioruby