[Bioc-devel] Merging or renaming a fork, and appropriate journal for package updates

[Sean Davis]

Hi, Pariksheet.

For #1, I suspect that it might be challenging to publish a paper based on the context you provide, but I do not have a good handle on the specific changes you have proposed. To make your work publishable, you would probably need to at least publish the software and it appears from #2 that piece is not quite done. Bioconductor does not have any recommendations on journals that folks use.

For #2, your best bet is to reach out to the package maintainers via email since not all maintainers use GitHub as religiously as others. Bioconductor team does not generally push changes to Bioconductor software repos; that is the responsibility of the package maintainer. Also, it appears that your issue is from several years ago, so it may have slipped off the radar. To make it easier for package maintainers to accept changes, be sure to provide a formal pull request (not code in an issue) against the current main/master branch on github to minimize maintainer work.

Best,
Sean


From: Bioc-devel <bioc-devel-bounces using r-project.org> on behalf of Pariksheet Nanda <pariksheet.nanda using uconn.edu>
Date: Wednesday, December 1, 2021 at 5:18 AM
To: bioc-devel using r-project.org <bioc-devel using r-project.org>
Subject: [Bioc-devel] Merging or renaming a fork, and appropriate journal for package updates
Hi folks,

I'm wrapping up my dissertation and one of the chapters touches on a
summer of patching a Bioconductor package that currently lives as a
separate GitHub fork (the list of changes is here [1]).  2 of the
questions I've been asked by a member of my committee are whether to:

(1) associate a publication with the work, and
(2) republish the code with the original or as a separate package.

For (1) while I appreciate the traditional unit of research output is
the publication, I'm struggling to think of a suitable journal for what
essentially is discussion about enhancements and some bugfixes.  I've
seen R package updates published in the Journal of Statistical Software
(JSS) which might be appropriate?  I suppose any place that gets indexed
on pubmed would work (yes, JSS is part of the NLM catalog [2]).  What
would you suggest?  Perhaps the Bioconductor project collect metrics for
publication activity about its packages to get more funding and has some
preference?

For (2) I would prefer to merge back with the original Bioconductor
package.  I tried upstreaming an early changeset [3], but besides my
issue being open, there are currently 2 other open GitHub issues with no
response which makes me wonder if upstream is dead.  If that's the case,
would someone from the Bioconductor core team be willing to work with me
to proxy commit to git.bioconductor.org?  I've made some API breaking
changes, so I expect I would need to create at least 2 branches: one
that can be commit with a deprecation warning for upcoming API breaking
changes, and a second branch with API breaking changes to be commit at
the subsequent Bioconductor release.  Or maybe I would need to create a
branch for each feature change; honestly I don't know if that would be
or less work but certainly it would be easier to read the git history.

Pariksheet

[1] https://github.com/coregenomics/groHMM/blob/1.99.x/NEWS
[2] https://www.ncbi.nlm.nih.gov/nlmcatalog/101307056
[3] https://github.com/Kraus-Lab/groHMM/issues/2

_______________________________________________
Bioc-devel using r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/bioc-devel

	[[alternative HTML version deleted]]