Discussion List Archives

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Assigning DOIs to dictionaries

  • To: "Discussion list of the IUCr Committee for the Maintenance of the CIFStandard (COMCIFS)" <comcifs@iucr.org>
  • Subject: Re: Assigning DOIs to dictionaries
  • From: James Hester <jamesrhester@gmail.com>
  • Date: Wed, 10 Jun 2020 16:44:19 +1000
  • DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025;h=mime-version:references:in-reply-to:reply-to:from:date:message-id:subject:to; bh=JlAHsJwKVwScupdxR7zDJnrSnvKm6MahpMjvQB5jS3s=;b=XPxiaz4Vg6X5Hb8e/va+gMLlb+AEpQO1bXpXF2Vorwlfl7glEj9LARUdVWJbrGqajniu/O8pi9rRvEHdUngq8+8YAJ6NJV9HItecPhzqkBMBd2t3UHeGY8qK4Hq5uoDVVp0Yb1mEnwiQmBvtv14Jsn7p+xZRScPMASVwi7biLtPVSa+i/iWkpclT8M6FGRN2+/gsvUw5dOfgSJalHUQJu6gMHVQD48SXlvtRIy3F340yKwDvLGy9z1Fnahyc0BxB1HBue6TF7vHyqvMBu4jFiOv+YzZBucrv+iEgyPEFcg1EkekIz4fuiIFKIkeGU5BtKGnD5jpDQB3TsNUUsrniTA==
  • In-Reply-To: <CAM+dB2cP+kbftpDWotNVTXDV5042c6VBkp6bacC+1SOx72oipQ@mail.gmail.com>
  • References: <CAM+dB2cP+kbftpDWotNVTXDV5042c6VBkp6bacC+1SOx72oipQ@mail.gmail.com>
Dear COMCIFS,

In view of the lack of response to the DOI suggestion below, I have taken the initiative and published the most recent version of the DDLm core dictionaries on Zenodo, which can be viewed at https://zenodo.org/record/3887473.  This deposit was created automatically from a release on Github. It is certainly not as polished as the official IUCr repository, but does provide a stable, citable object. Feedback on this is of course welcome.

In lieu of a decent concept of an organisation on Zenodo (in the sense of multiple accounts having access) I have created a COMCIFS "community". What this means is that anybody may deposit items for COMCIFS into Zenodo, and provided the curator of the community gives the tick, that item will be added to a list of community items (as well as appearing in the contributor's Zenodo list regardless).

I have attempted to resolve the authorship issue for the Zenodo deposit by including those who submitted pull requests to the Github repository (this is automatic) as well as Nick and Syd as the original authors of the DDLm version, and put everybody in alphabetical order. I have noted the derivation of the DDLm dictionaries from DDL1 by listing Vol G as a related item.

Clearly I have not followed my original suggested workflow but it is something we can aspire to.

best wishes,
James.

On Mon, 27 Apr 2020 at 17:14, James Hester <jamesrhester@gmail.com> wrote:
Dear COMCIFS,

In the past we have discussed assigning DOIs to our CIF dictionaries via the mechanism used by the IUCr for their publications. So far this discussion has not yielded much fruit.  A DOI is quite useful as it allows us to refer to other dictionaries via DOIs and URLs that are pretty much guaranteed to not become outdated and are machine-actionable.

Instead of the IUCr publication mechanism it occurred to me that we might use an outside service like Zenodo to deposit canonical dictionary versions. How this would work is as follows:

(1) A new dictionary version would be approved using the current mechanisms
(2) The dictionary would be deposited at Zenodo using a COMCIFS or IUCr official account and a DOI reserved (ie document is not yet published)
(3) This DOI would be edited into the dictionary if necessary
(4) Further technical editing would be undertaken as needed
(5) The final dictionary is published on the IUCr website and the 'Publish' button pushed on Zenodo

Ideally the 'author' for Zenodo and citation purposes would be either 'COMCIFS' or 'IUCr'. The authentication password could reside with the COMCIFS secretary(s). I have chosen Zenodo because it is not affiliated or funded by a commercial entity and I have some experience publishing there, but alternative suggestions are welcome.

The minor downsides I see with this scheme are that (1) the DOI would contain the string 'zenodo', which is sort of against the spirit of the underlying object being able to move locations (2) a data DOI such as that delivered by Zenodo points to a 'landing page', not to the object itself. The resources are instead linked from the landing page; as far as I can tell these links are predictable (DOI URL + '/files/<filename>').  I'd prefer an even simpler way to point a program at a DOI and get back the file I wanted, but I digress.

Thoughts?

James.
--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148


--
T +61 (02) 9717 9907
F +61 (02) 9717 3145
M +61 (04) 0249 4148

Reply to: [list | sender only]