[Bioc-devel] splitting a Bioconductor package in two

[Michael Lawrence]

On Wed, Sep 6, 2017 at 9:07 AM, Stephanie M. Gogarten
<sdmorris at u.washington.edu> wrote:
> Hi,
>
> If one wanted to move some code from an existing Bioconductor package into
> its own, separate package for ease of maintenance, would that constitute a
> new package submission with the full review process? Or is there some
> expedited submission available?
>
> We are refactoring the GENESIS package and debating whether to move the
> low-level statistical testing code into a different package (which GENESIS
> would then import). As I see it, pros include easier maintenance and
> testing, and the potential value of the stats package to other users and
> developers who might not want to import the entire GDS framework that
> GENESIS uses. Cons include the overhead of submitting, writing vignettes,
> etc. for a set of functions that are meant to be used internally.
>

There probably is value in separating the stats from the I/O, but it
would probably take some work to make those internal functions
suitable for use by others. Probably enough work that it would benefit
from a review. It all depends on how likely those stats will be
independently useful. Are there any existing use cases?

> What are other pros and cons of this approach (versus just keeping
> everything in one big package)?
>
> thanks,
> Stephanie
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel