[Biopython] Biopython Digest, Vol 156, Issue 19

Ivan Erill ivan.erill at gmail.com
Wed Dec 23 09:54:47 UTC 2015 

>
> More seriously, I'd be much happier with a convention where you'd cite the
> major biopython paper, the specific module papers if there are any, and
> include the hash of the revision you used. This should be included in the
> wiki page where we define 'citation guidelines'. It is simple, helps
> everyone (although occasional contributors might feel robbed), and is fully
> reproducible.

This seems very sensible in terms of ensuring reproducibility and
maintaining impact without added complications.

That being said, and at the risk of opening a can of worms, I find it
> very unfair that many people that have recently contributed hard to the
> project are not on the author list of the 2009 paper (you, for
> example. Bow is another). That would be fixed with a new paper. Deciding
> on the author list would be the can of worms. We all can agree that it
> would be another Cock et al paper, but getting the list of people that
> are the "et al" would be complex.

That requires analyzing whether there is substantial new stuff to warrant a
new publication and whether the new pub could make it to a journal of at
least similar impact. Having an old (2009!) but highly cited publication is
good, but may backfire if people/funders perceive there has been no
(citable) progress in the area. Being able to cite extensions (e.g.
BioPhylo) helps address that problem and enticing extension developers to
publish, rather than publishing updates of Biopython, might be the way to
go. Then again, most major projects in bioinformatics (e.g. GO, RefSeq,
RegulonDB, STRING, eggNOG...) publish frequent updates, so why not
Biopython? Having several Biopython papers might dilute the citation count,
but you can then state that there have been X biopython publications, so
it's a tradeoff, as usual.

Regarding the can of worms, most journals currently accept author "lists"
that may be maintained externally. A paper could be signed by the major
contributors of the current version (and/or whomever actually takes the
time to write the darn thing) followed by a "list" author pointing to the
historical contributor list.

Ivan


On Wed, Dec 23, 2015 at 3:38 AM, <biopython-request at mailman.open-bio.org>
wrote:

> Send Biopython mailing list submissions to
>         biopython at mailman.open-bio.org
>
> To subscribe or unsubscribe via the World Wide Web, visit
>         http://mailman.open-bio.org/mailman/listinfo/biopython
> or, via email, send a message with subject or body 'help' to
>         biopython-request at mailman.open-bio.org
>
> You can reach the person managing the list at
>         biopython-owner at mailman.open-bio.org
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Biopython digest..."
>
>
> Today's Topics:
>
>    1. Re: DOI for BioPython (c.buhtz at posteo.jp)
>    2. Re: DOI for BioPython (Fields, Christopher J)
>    3. Re: DOI for BioPython (Iddo Friedberg)
>    4. Re: DOI for BioPython (Tiago Antao)
>    5. Re: DOI for BioPython (Jo?o Rodrigues)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Tue, 22 Dec 2015 18:35:30 +0100
> From: <c.buhtz at posteo.jp>
> To: biopython at mailman.open-bio.org
> Subject: Re: [Biopython] DOI for BioPython
> Message-ID: <3pQ4bR6MwMz5vNM at dovecot03.posteo.de>
> Content-Type: text/plain; charset=UTF-8
>
> On 2015-12-22 16:38 Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com>
> wrote:
> > What's better: citing the paper that describes the algorithm(s) or
> > the pull request that fixed a typo in the documentation?
>
> The latter because a DOI/Reference is about reproducibility.
> --
> GnuPGP-Key ID 0751A8EC
>
>
>
> ------------------------------
>
> Message: 2
> Date: Tue, 22 Dec 2015 18:30:47 +0000
> From: "Fields, Christopher J" <cjfields at illinois.edu>
> To: Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com>
> Cc: "biopython at mailman.open-bio.org" <biopython at mailman.open-bio.org>
> Subject: Re: [Biopython] DOI for BioPython
> Message-ID: <1E2AA25C-2269-4281-A364-0FEE6F673EE1 at illinois.edu>
> Content-Type: text/plain; charset="utf-8"
>
> I agree re: tracking citations, etc, particularly for funding.  However,
> for attributing credit for developers it?s actually tricky, particularly
> with large projects that have an evolving group of core developers.
>
> My personal example: I wasn?t on the original bioperl citation (joined
> after), yet I am now essentially managing bioperl development when I can.
> My commits and releases are the things I could to point to, so it?s nice to
> have something of a compromise (e.g. an updated DOI via Zenodo).  Not to
> mention it shows that development has continued past the point of
> publication (sometimes with significant changes to the API), which as we
> all know isn?t true for a significant chunk of published software.
>
> So in my opinion it?s never a bad thing to have an updated (officially
> sanctioned!) citation reflecting contributions from new developers or
> significant changes to the codebase.  The original citation will never go
> away.
>
> chris
>
> On Dec 22, 2015, at 10:38 AM, Jo?o Rodrigues <
> j.p.g.l.m.rodrigues at gmail.com<mailto:j.p.g.l.m.rodrigues at gmail.com>>
> wrote:
>
>
> Having all citations concentrated in one paper with all the major
> contributors helps tracking citation data, which helps those major
> contributors and OBF in future funding ventures. Further, a new release
> does not change the entire code base, so if you were to use Bio.PDB in say
> release X and Y, where there weren't any changes to this module, you'd cite
> two different DOIs, which wouldn't make any sense from any point of view.
>
> Also, Peter's paper is not *his* paper about the software. It's the entire
> team, or almost, on that paper and it describes the different modules and
> their purpose and functionality.
>
> What's better: citing the paper that describes the algorithm(s) or the
> pull request that fixed a typo in the documentation?
>
> Cheers,
>
> Jo?o
>
> A ter, 22/12/2015, 16:05,  <c.buhtz at posteo.jp<mailto:c.buhtz at posteo.jp>>
> escreveu:
> On 2015-12-22 15:10 Peter Cock <p.j.a.cock at googlemail.com<mailto:
> p.j.a.cock at googlemail.com>> wrote:
> > You can just say Biopython version 1.66 (for example) and
> > cite the paper Cock et al 2009 and its DOI:
>
> I need to cite the software not your paper about the software.
> It is unscientific to do it like that. And somebody could say I just
> want to say "thank you" in the way to increase your ImpfactFactor.
>
> If there is no DOI I will just name the main website an the release
> identifier (version number).
>
> > Using Zenodo.org<http://zenodo.org> we'd have to register a DOI for each
> > release, which does not seem that useful.
>
> Exactly that would be useful and what DOIs are for - especially for a
> scientific software like BioPython.
> --
> GnuPGP-Key ID 0751A8EC
> _______________________________________________
> Biopython mailing list  -  Biopython at mailman.open-bio.org<mailto:
> Biopython at mailman.open-bio.org>
> http://mailman.open-bio.org/mailman/listinfo/biopython<
> https://urldefense.proofpoint.com/v2/url?u=http-3A__mailman.open-2Dbio.org_mailman_listinfo_biopython&d=BQMFaQ&c=8hUWFZcy2Z-Za5rBPlktOQ&r=fbHa8Njtvh9VmSnzJxiEUTW9NWDwMMwQAzhgZDO41GQ&m=v7VA0sPPM-8wrEQBkLt238WQ-og1ST_QKBqUGcABdgY&s=sgGUzE7f0glVTLTJewIX1aFy5qMG8RzPqMJ4htLwdQc&e=
> >
> _______________________________________________
> Biopython mailing list  -  Biopython at mailman.open-bio.org<mailto:
> Biopython at mailman.open-bio.org>
> http://mailman.open-bio.org/mailman/listinfo/biopython
>
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://mailman.open-bio.org/pipermail/biopython/attachments/20151222/91e78b6c/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 3
> Date: Tue, 22 Dec 2015 13:29:13 -0600
> From: Iddo Friedberg <idoerg at gmail.com>
> To: biopython at mailman.open-bio.org
> Subject: Re: [Biopython] DOI for BioPython
> Message-ID:
>         <
> CABm4-MRaG3jqz1iEgWcz22fxQogHouwynYnpiiEZKysNYs6TxA at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> COI: I am a paper co-author.
>
> My $0.02 is that citing the paper is much more beneficial for the project,
> for the reason Leighton said: a highly cited paper does help justify work
> on the project to funding agencies as well as institutions. Speaking for
> myself, I have cited the paper in two funded grant applications, one which
> has resulted in some code contribution (the GAF parser) and one which
> hopefully will benefit Biopython as well (recently funded to do some work
> on GO, need to talk about that in the dev list). As Leighton said, funders
> & bosses judge importance by citation count. While Biopython is not
> directly funded, some of us manage to spend some time on it using our own
> grants.
>
>
> To facilitate reproducibility, a major version citation is sufficient, IMO,
> as those are archived. If you used some non-major version on github, you
> can fork/clone the version of your liking to your own github,  assign it a
> DOI yourself, and cite that. For reasons mentioned, I would advocate citing
> both the paper and the DOI you have assigned.
>
> On Tue, Dec 22, 2015 at 11:35 AM, <c.buhtz at posteo.jp> wrote:
>
> > On 2015-12-22 16:38 Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com>
> > wrote:
> > > What's better: citing the paper that describes the algorithm(s) or
> > > the pull request that fixed a typo in the documentation?
> >
> > The latter because a DOI/Reference is about reproducibility.
> > --
> > GnuPGP-Key ID 0751A8EC
> >
> > _______________________________________________
> > Biopython mailing list  -  Biopython at mailman.open-bio.org
> > http://mailman.open-bio.org/mailman/listinfo/biopython
> >
>
>
>
> --
> Iddo Friedberg
> http://iddo-friedberg.net/contact.html
> ++++++++++[>+++>++++++>++++++++>++++++++++>+++++++++++<<<<<-]>>>>++++.>
> ++++++..----.<<<<++++++++++++++++++++++++++++.-----------..>>>+.-----.
> .>-.<<<<--.>>>++.>+++.<+++.----.-.<++++++++++++++++++.>+.>.<++.<<<+.>>
> >>----.<--.>++++++.<<<<------------------------------------.
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://mailman.open-bio.org/pipermail/biopython/attachments/20151222/b955e520/attachment-0001.html
> >
>
> ------------------------------
>
> Message: 4
> Date: Tue, 22 Dec 2015 18:03:48 -0700
> From: Tiago Antao <tra at popgen.net>
> To: Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com>
> Cc: biopython at mailman.open-bio.org
> Subject: Re: [Biopython] DOI for BioPython
> Message-ID: <20151222180348.5d044506 at urso>
> Content-Type: text/plain; charset=UTF-8
>
> On Tue, 22 Dec 2015 16:38:37 +0000
> Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com> wrote:
>
> > Also, Peter's paper is not *his* paper about the software. It's the
> > entire team, or almost, on that paper and it describes the different
> > modules and their purpose and functionality.
>
> One can discuss if it is good or bad, but basic pragmatism dictates
> that the proper citing object is the 2009 paper, period. This for the
> reasons stated here by others already. Myself I would not be able to
> have contributed over the years if there was not a high-cited paper to
> put forward as a justification.
>
> Having another (non-paper) citing object will damage the project and
> has my full frontal objection. Things in science work as they work, and
> while I might object to paper citation metrics, that is the
> world we live in. I believe Biopython already lives too much on the
> volunteering of a few individuals, removing the only thing that
> provides "income" would be fatal. We actually need more (not less)
> forms of "income".
>
> That being said, and at the risk of opening a can of worms, I find it
> very unfair that many people that have recently contributed hard to the
> project are not on the author list of the 2009 paper (you, for
> example. Bow is another). That would be fixed with a new paper. Deciding
> on the author list would be the can of worms. We all can agree that it
> would be another Cock et al paper, but getting the list of people that
> are the "et al" would be complex.
>
> Tiago
>
>
>
> ------------------------------
>
> Message: 5
> Date: Wed, 23 Dec 2015 02:38:22 +0000
> From: Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com>
> To: Tiago Antao <tra at popgen.net>
> Cc: biopython at mailman.open-bio.org
> Subject: Re: [Biopython] DOI for BioPython
> Message-ID:
>         <CAB=-
> b2aEPQRGOc+jpoGMDda-zeXYjnwPOcsP9fqo+WtutfzFYg at mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
>
> Since this is a python library I'd go with the mantra of there is only one
> way to do things, thus, one citation to keep things simple :)
>
> More seriously, I'd be much happier with a convention where you'd cite the
> major biopython paper, the specific module papers if there are any, and
> include the hash of the revision you used. This should be included in the
> wiki page where we define 'citation guidelines'. It is simple, helps
> everyone (although occasional contributors might feel robbed), and is fully
> reproducible.
>
> Including DOIs for every release is only partially reproducible, since it
> relies on a third party service to map between the DOI and the release,
> which adds an unnecessary, in my opinion, extra layer of complexity. Also,
> it doesn't help those that do not know biopython to learn about it since
> they will have to Google it's publication.
>
> It's a good discussion and a good point to raise, but there are still major
> rules in science we have to play by in order to keep people interested and
> generally move forward.
>
> As Tiago, I'm also against implementing pre-release DOIs. It'll create too
> much entropy that has no real benefits other than ideology.
>
> A qua, 23/12/2015, 01:03, Tiago Antao <tra at popgen.net> escreveu:
>
> > On Tue, 22 Dec 2015 16:38:37 +0000
> > Jo?o Rodrigues <j.p.g.l.m.rodrigues at gmail.com> wrote:
> >
> > > Also, Peter's paper is not *his* paper about the software. It's the
> > > entire team, or almost, on that paper and it describes the different
> > > modules and their purpose and functionality.
> >
> > One can discuss if it is good or bad, but basic pragmatism dictates
> > that the proper citing object is the 2009 paper, period. This for the
> > reasons stated here by others already. Myself I would not be able to
> > have contributed over the years if there was not a high-cited paper to
> > put forward as a justification.
> >
> > Having another (non-paper) citing object will damage the project and
> > has my full frontal objection. Things in science work as they work, and
> > while I might object to paper citation metrics, that is the
> > world we live in. I believe Biopython already lives too much on the
> > volunteering of a few individuals, removing the only thing that
> > provides "income" would be fatal. We actually need more (not less)
> > forms of "income".
> >
> > That being said, and at the risk of opening a can of worms, I find it
> > very unfair that many people that have recently contributed hard to the
> > project are not on the author list of the 2009 paper (you, for
> > example. Bow is another). That would be fixed with a new paper. Deciding
> > on the author list would be the can of worms. We all can agree that it
> > would be another Cock et al paper, but getting the list of people that
> > are the "et al" would be complex.
> >
> > Tiago
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL: <
> http://mailman.open-bio.org/pipermail/biopython/attachments/20151223/560ce6cc/attachment.html
> >
>
> ------------------------------
>
> Subject: Digest Footer
>
> _______________________________________________
> Biopython mailing list  -  Biopython at mailman.open-bio.org
> http://mailman.open-bio.org/mailman/listinfo/biopython
>
> ------------------------------
>
> End of Biopython Digest, Vol 156, Issue 19
> ******************************************
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.open-bio.org/pipermail/biopython/attachments/20151223/1dfd98ca/attachment-0001.html>

More information about the Biopython mailing list