[Bioc-devel] Update policy on experiment data and annotation packages

[Marc Carlson]

Hi Julian,

This is a complicated issue for us and we have to choose our next move 
carefully since we don't have unlimited resources. Especially not with 
respect to time.  But I wanted to let you know that we appreciate your 
comment and that we are still thinking about it.

   Marc



On 10/10/2013 03:05 AM, Julian Gehring wrote:
> Hi,
>
> What is the consensus on updating data in experiment data and 
> annotation packages?
>
> The bioc website [1] does not state any differences between the two 
> package types in terms of updating their content.  From the bioc core, 
> I have the information that (a) experimental data packages should 
> represent 'frozen' data and not get updated over release cycles, while 
> (b)  annotation packages should get updated with every release cycle. 
> Should we add this information to the website?
>
> I'm curious what this means for experimental data that accumulates 
> over time, i.e. data from big consortia, as represented by e.g. 
> 'curatedOvarianData', 'SomaticCancerAlterations', and others. Should 
> one create create a new package with each release cycle (indicating 
> the data version in the package name, as the 'SNPlocs*' packages) to 
> ensure reproducibility?  Or update an annotation package with each 
> release, and try to ensure backwards compatibility within the package 
> itself?
>
> Best wishes
> Julian
>
> ----
> [1] http://bioconductor.org/developers/package-guidelines/#package-types
>
> _______________________________________________
> Bioc-devel at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/bioc-devel