Re: References to registries -> debian-med policy?
On 21.09.17 23:13, Dylan Aïssi wrote:
> Hi Steffen,
>
> 2017-09-16 18:37 GMT+02:00 Steffen Möller <steffen_moeller@gmx.de>:
>> I have added a sentence to the Debian Med policy and also pushed
>> the "Registry:" entry for the UpstreamMetadata wiki page that our
>> policy is pointing to.
>>
> Thank you for this.
>
> After a discussion with Fabien (from OMICtools) in August, we switched
> the name of the SCR_XXXXXX IDs from "RRID" to "SciCrunch" (for a
> randomly example [1]). But, now I see that you come back to "RRID". In
> the UpstreamMetadata wiki page, you also use the name "RRID" for these
> IDs. Just to be sure, what is now the recommended name for these IDs?
>
> Best,
> Dylan
>
> [1] https://anonscm.debian.org/cgit/debian-med/phyml.git/commit/?id=edabc350f1
That is a tricky one. I had changed to SciCrunch and one of the first
such changes
from RRID to SciCrunch was corrected back to SciCrunch - may just have
been some git
conflict that was resolved the other way. I did not feel like bike
shedding about
it much at the time, I just wanted to get things in, convinced that we
could easily
change to whatever any time, so I then continued with RRID.
A look at table 1 in
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4648211/
indeed indicates that SciCrunch is more exact than RRID, especially since
https://arxiv.org/ftp/arxiv/papers/1707/1707.03659.pdf
states that OMICtools' identifiers would also be valid RRIDs. I do not
know about bio.tools.
So, yes, for the savant Debian packagers I think SciCrunch is better
as a better identifier in the metadata file than RRID. But when we
put the RRIDs on our task pages I am not so sure. People want
to read "RRID" somewhere and at least as of today, they would likely
feel unhappy when we point them to something else than SciCrunch.
For the public presentations I could imagine to have just one field "RRID:"
that puts IDs from both resources to the degree that these are available.
Does that make any sense? If there is some consensus to this, then I would
edit the UpstreamMetadata page accordingly.
Best,
Steffen
Reply to: