[open-science] citing a code and data on a GitHub repository

Tue Dec 3 17:23:09 UTC 2013

I'm very sympathetic to Arfon's request as I often forget what I wrote in
my own code 20 years ago.  My best guess is that full text indexing (I use
Javadoc)  is as valuable as trying to create keywords and microformats. I
am reduced to Googling for my own code on Git/Bitbucket.

The mathematicians have well defined bibliographic categories (e.g. PDEs)
and I'd recommend using them as they are widespread use. Not sure about
compsci. Names of algorithms are always useful so, for example I have
recently implemented / reused Zhang Suen Thinning, Otsu Binarization, Canny
edge Detection, Hough Transform algorithms for image processing and they
should be very discoverable by search engines if exposed.

I'd assume that the docs for a significant body of code would indicate the
language (Javadoc is distinctive).

On Tue, Dec 3, 2013 at 5:10 PM, Arfon Smith <arfon at github.com> wrote:

> I'd be interested to hear if there's been any effort put into how to
> properly describe the function of a piece of code in a file like
> BibJSON. For example, knowing that a piece of code was written in
> Python, was for an astrophysics domain and performed coordinate
> transformations. It feels like that could be encapsulated in some kind
> of meta descriptor file which could then be indexed (and searched).
>
> Has anyone seen anything like this done?
>
> Cheers
> Arfon
>
> On 12 November 2013 14:15, Piotr Migdal <pmigdal at gmail.com> wrote:
> > BibJSON is more of format. But I didn't know that, Emanuil, thanks for
> > mentioning!
> >
> > Also I got (from Rainere Silva) a link to a nice discussion on the topic:
> >
> http://lists.software-carpentry.org/pipermail/discuss_lists.software-carpentry.org/2013-October/001048.html
> >
> > Sheila, thanks for pointing to another discussion.
> >
> > Regards,
> > Piotr
> > BTW: http://offtopicarium.wikidot.com/
> > - an unconference for young scientists, programers and activists from
> > various fields; all topics allowed
> >
> >
> > On 12 Nov 2013, at 17:15, sheila miguez <shekay at pobox.com> wrote:
> >
> > I saw some discussion happening about this on one of the
> software-carpentry
> > mailing lists, but since those aren't for the general world, more
> discussion
> > could happen here or on the sci tech mailing list I created for people to
> > discuss technical details and implementation ideas for tools they work
> on.
> >
> > Here's a conversation I posted and then a good reply from Aron Ahmadia.
> The
> > WSSPE workshop is coming up, and he suggested reading through the
> > contributions on citation.
> >
> > http://wssspe.researchcomputing.org.uk/contributions/
> >
> http://lists.hackingscience.org/pipermail/scitech-hackingscience.org/2013-October/000001.html
> >
> >
> >
> >
> > On Mon, Nov 11, 2013 at 12:27 PM, Emanuil Tolev <emanuil at cottagelabs.com
> >
> > wrote:
> >>
> >> A mix of BibJSON fields can probably *describe* such a reference
> >> satisfactorily in a machine-friendly format. http://www.bibjson.org/
> >>
> >> If I had to do it in my own work, I'd work up from that, trying to make
> >> sure it's consistent with whatever format I'm using for the rest of the
> >> bibliography (e.g. IEEE). Most of them have some measure of support for
> "a
> >> thing with a URL". I realise that's not ideal though, and BibJSON is
> only
> >> good as far "what pieces of information do I put in" goes, not, say,
> >> formatting them.
> >>
> >> Greetings,
> >> Emanuil
> >>
> >>
> >> On 11 November 2013 17:50, Piotr Migdal <pmigdal at gmail.com> wrote:
> >>>
> >>> Hi!
> >>>
> >>> Does any of you know what is the preferred way to cite code (placed on
> >>> GitHub)?
> >>>
> >>> I saw the question here:
> >>>
> >>>
> http://academia.stackexchange.com/questions/14010/how-do-you-cite-a-github-repository
> >>>
> >>> Also, a similar question (i.e. citing things not being journal
> >>> publications) for  OEIS:
> >>>
> >>>
> http://mathoverflow.net/questions/131245/how-to-cite-a-sequence-from-the-on-line-encyclopedia-of-integer-sequences-oeis
> >>>
> >>> Do you know what is "good practice"?
> >>> Is there some (perhaps informal) place trying to standardize the format
> >>> (e.g. in BibTeX) or in general worth looking for a reference?
> >>>
> >>> Regards,
> >>> Piotr
> >>> BTW: http://offtopicarium.wikidot.com/
> >>> - an unconference for young scientists, programers and activists from
> >>> various fields; all topics allowed
> >>> _______________________________________________
> >>> open-science mailing list
> >>> open-science at lists.okfn.org
> >>> http://lists.okfn.org/mailman/listinfo/open-science
> >>> Unsubscribe: http://lists.okfn.org/mailman/options/open-science
> >>
> >>
> >>
> >> _______________________________________________
> >> open-science mailing list
> >> open-science at lists.okfn.org
> >> http://lists.okfn.org/mailman/listinfo/open-science
> >> Unsubscribe: http://lists.okfn.org/mailman/options/open-science
> >>
> >
> >
> >
> > --
> > sheila
> > _______________________________________________
> > open-science mailing list
> > open-science at lists.okfn.org
> > http://lists.okfn.org/mailman/listinfo/open-science
> > Unsubscribe: http://lists.okfn.org/mailman/options/open-science
> >
> >
> >
> > _______________________________________________
> > open-science mailing list
> > open-science at lists.okfn.org
> > http://lists.okfn.org/mailman/listinfo/open-science
> > Unsubscribe: http://lists.okfn.org/mailman/options/open-science
> >
> _______________________________________________
> open-science mailing list
> open-science at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/open-science
> Unsubscribe: http://lists.okfn.org/mailman/options/open-science
>

-- 
Peter Murray-Rust
Reader in Molecular Informatics
Unilever Centre, Dep. Of Chemistry
University of Cambridge
CB2 1EW, UK
+44-1223-763069
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.okfn.org/pipermail/open-science/attachments/20131203/075074c0/attachment-0003.html>