Some digital data are also not everlasting. They don't get used up as such, but data about people is sometimes only consented to be stored for a finite period of time after which they are to be deleted. My understanding is that with this digital data and physical samples that only last until they are gone, then the DataCite DOI still resolves to the landing page for these data, but on the landing page it is clear why the data are no longer available. The landing page metadata is to persist in perpetuity.
Some forms of DOI for real world objects do exist as the geological example shows, there are others including the UK national mapping agency, Ordnance Survey's Topographical Object Identifiers (TOIDs) that refer to physical objects, features of their geometry and attributes. There have been interesting discussions about when to mint a new or reuse a TOID, for example with a building which gets extended and then redeveloped.
I appreciate this is a side issue to what Miriam is asking about. As for people as a subject of study or their views, then clearly ORCID references work only when these are researchers, not the subject.
Andy
http://www.geog.leeds.ac.uk/people/a.turner/
-----Original Message-----
From: Research Data Management discussion list [mailto:[log in to unmask]] On Behalf Of M. Casula
Sent: 20 September 2013 09:25
To: [log in to unmask]
Subject: Re: DataCite DOIs and tissue samples
Dear all,
Thank you for the quick feedback, I now have an idea of how to proceed. I would like though to elaborate a little on the original post.
DOIs are digital object identifiers, so implicitly they are *intended* as identifiers of digital objects rather than real objects (for me there is a difference between a digital identifier of an object and a digital object identifier). This has some implications for their use because the metadata accompanying a digital object (a data set when talking about DataCite DOIs) is quite different from the metadata that would accompany a real object. This statement alone could probably be the start of a lengthy discussion, but it is tangential to my original post. Similarly, I would say that ORCID is intended as means of tracking (resolving) people, namely researchers, while DOIs are intended for citation. Again, another statement that is subject to individual opinion, but again which is tangential to my original post.
My original post relates to the fact that DOIs provide a means for the tissue samples with which I am working to be cited and potentially reused (which imo is the more important benefit). These samples (objects) have metadata that are broadly consistent with the DataCite metadata schema and so a DOI could be minted that resolves to a landing page that displays the metadata about the object (as is accepted best practice) and not the object itself. If a researcher is interested to use the object to which the metadata corresponds, there would be a mechanism whereby they could access the object. For a data set this would probably be a link to the actual digital object, for tissue samples it would be by way of contact with the curator. So, to this point minting a DataCite DOI for a tissue sample is perfectly feasible. The issue though is that DataCite DOIs are intended *always* to resolve to the object to which they correspond (because if they are cited they need to be available) and a tissue sample may not always be around because eventually it could be completely used up.
What I need to do have a plausible and acceptable means to work around the mentioned 'issue'. Perhaps IGSN could evolve into or provide the basis for a 'SampleCite' framework, and while I agree that DOIs can be used for real objects, my question is whether they should be ? While DOIs resolve in the first instance to metadata, the metadata are simply a 'wrapper' for the object-they are only information about the object, and it is the object itself that must be available, at least in a citation framework (rather than say a tracking framework). Perhaps the answer in the short term for the tissue samples lies in a combination of CERIF and DataCite, where the latter provides the permanent identifier and the former qualifies the origin and use of the samples i.e. specializes the DataCite metadata. I don't know enough about CERIF to know whether this is feasible but I will certainly look further into both CERIF and IGSN.
I am really very grateful for the posts.
Thank you,
Miriam
________________________________________
Van: Research Data Management discussion list [[log in to unmask]] namens Keith Jeffery [[log in to unmask]]
Verzonden: vrijdag 20 september 2013 9:48
Aan: [log in to unmask]
Onderwerp: Re: DataCite DOIs and tissue samples
Miriam, all -
Using any appropriate identifier is fine; the key thing is that the metadata associated with that identifier is sufficient to determine (a) that the object being referred to is not digital; (b) relevant description for discovery; (c) relevant restriction as to use (e.g. has to be viewed in the museum or can be sent for examination) and (d) relevant context (who collected, when, why,ownership, custodianship, related aatasets, publications, equipment and/or facility used for examination..... All of this can be recorded using CERIF (Common European Research Information Format) http://eurocris.org/Index.php?page=CERIFreleases&t=1 which can interoperate with (i.e. map/generate) common metadata formats used in research.
Best
Keith
Keith G Jeffery Consultants
Prof Keith G Jeffery
E: [log in to unmask]
T: +44 7768 446088
S: keithgjeffery
President ERCIM www.ercim.eu ([log in to unmask])
President emeritus euroCRIS www.eurocris.org Past Vice President VLDB www.vldb.org Fellow (CITP, CEng) BCS www.bcs.org
----------------------------------------------------------------------------------------------------------------------------------
The contents of this email are sent in confidence for the use of the intended recipient only. If you are not one of the intended recipients do not take action on it or show it to anyone else, but return this email to the sender and delete your copy of it.
----------------------------------------------------------------------------------------------------------------------------------
-----Original Message-----
From: Research Data Management discussion list [mailto:[log in to unmask]] On Behalf Of Jens Klump
Sent: 20 September 2013 08:24
To: [log in to unmask]
Subject: DataCite DOIs and tissue samples
Hi Miriam,
The questions you asked with respect to DOIs for tissue samples are the same questions we asked ourselves in the project "Publication and Citation of Primary Research Data" (STD-DOI) <http://www.std-doi.de> if we wanted to assign DOI as identifiers to geological samples. Due to the specific use cases of geological samples and due to the business model for DOIs at the time we decided against using DOI and set up our own handle-based system, now modelled after the example of DataCite. It is now in operation as the International Geo Sample Number (IGSN). More information can be found at <http://www.igsn.org> and <http://dokuwiki.gfz-potsdam.de/datawiki/doku.php?id=igsn:start>.
The IGSN persistent identifier infrastructures could be used for other types of samples or disciplines, too.
Since the inception of the IGSN things have changed and it is clearly feasible to use DOI for physical specimens.
Can DOI be used for physical objects?
Yes, because DOI is a digital identifier for objects, not only an identifier for digital objects.
Where should the DOI point in the case of physical samples?
The DOI should point to a digital representation of the physical object, i.e. a "landing page" with metadata on this object.
What about specimens being consumed or degrading with time?
The consumption of degradation with time of specimens is a common case in geochemistry. The specimen, if published, is still an object that is being referenced by data and literature and has thus become part of the record of science. It is therefore in line with common practice in scientific communication to keep a reference to the object of investigation, even if the object no longer exists.
A really significant feature of DOI and IGSN is their metadata element "relatedIdentifier" which allows to point from one identifier to another identifier, e.g. data to literature, and codify the nature of this relation, e.g. "isCitedBy". This element allows a much tighter integration of literature, data and samples, aiding both systematic and serendipitous discovery.
Citation of specimens is certainly an important factor in the acknowledgement for curation of the specimen.
The primary goal for the development of the IGSN was to create a system of unambiguous, worldwide resolvable names for geological samples. [1] Unsystematic naming of samples made close to impossible to compile synthesis studies of global geochemistry. I am sure the issue of sample identification is also relevant to tissue samples [2].
Kind regards,
Jens
[1] Lehnert, K. A., and J. Klump (2012), The Geoscience Internet of Things, in Geophysical Research Abstracts, vol. 14, pp. EGU2012-13370, Copernicus Society, Vienna, Austria. [online] Available from:
http://meetingorganizer.copernicus.org/EGU2012/EGU2012-13370.pdf
[2] Engel, M. (2012), Falsche und verunreinigte Zellen - Akademisches Risiko durch falsche Etikettierung, Forschung Aktuell. [online] Available from: http://www.dradio.de/dlf/sendungen/forschak/1708955/
(Accessed 21 March 2012)
> ------------------------------
>
> Date: Thu, 19 Sep 2013 20:15:29 +0000
> From: "M. Casula" <[log in to unmask]>
> Subject: DataCite DOIs and tissue samples
>
> Hello,
>
>
>
> Sometime ago an inquiry was posted on my behalf regarding the use of
> DOIs for tissue samples to enable their citation in derivative works.
>
>
>
> The added value of aknowledging the source should not be
> underestimated, as it gives due credit to the curators and donors, and
> promotes use of the samples for further research. This is infact what
> is motivating my interest to assign identifiers to tissue samples.
>
>
>
> While in principle it is feasible to use DOIs with tissue samples (by
> having them resolve to metadata about the samples themselves) the
> question arises as to whether a digital object identifier should be
> used for something real as opposed to something digital? Although this
> may seem somewhat academic, there is a practical issue in this
> particular case because tissue samples are gradually consumed and will
> eventually no longer exist. In which case, I wonder whether the use of
> DOIs to identify real things is advisable?
>
>
>
> Any advice and/or opinion would be welcome.
>
>
>
> Kind regards,
>
> Miriam
>
>
>
>
>
> Miriam Casula, PhD
> Neuropathology Department
> Academic Medical Centre
> University of Amsterdam
> Meibergdreef 9
> The Netherlands
> Tel: +31205665649
>
> ________________________________
>
> AMC Disclaimer : http://www.amc.nl/disclaimer
>
> ________________________________
>
> ------------------------------
>
> Date: Thu, 19 Sep 2013 23:46:18 +0100
> From: Andy Turner <[log in to unmask]>
> Subject: Re: DataCite DOIs and tissue samples
>
> Hi Miriam et al.,
>
> It is my understanding that a DOI itself is digital, but it can refer
to a physical sample or other physical data like it can refer to some digital data. Physical samples are clearly data and informational and using DOI's to refer to these is a good idea. Physical samples that are scanned to produce digital data about the sample may not last as long as the digital data. Whether it is worth assigning a DOI comes down to how long that sample is going to be around. If it's not likely to be around for long enough for it to be reused then it probably isn't worth it.
What that means in practice is probably at least weeks, but then I suppose that depends on how fast the field is.
>
> In terms of tissue samples, these clearly are data in my mind. Indeed,
by extension, whole organisms are too in stored contexts. The thing I struggle with is if there is scope to use a DOI to refer to an individual person or a collection of people that may have been part of a study, are not stored for further study, but can be found via addresses for further study. If there is a DOI for a living person, I'd really like to hear about it.
>
> Regards,
>
> Andy
>
> ________________________________
> From: Research Data Management discussion list
[[log in to unmask]] On Behalf Of M. Casula [[log in to unmask]]
> Sent: 19 September 2013 21:15
> To: [log in to unmask]
> Subject: DataCite DOIs and tissue samples
>
>
> Hello,
>
>
>
> Sometime ago an inquiry was posted on my behalf regarding the use of
DOIs for tissue samples to enable their citation in derivative works.
>
>
>
> The added value of aknowledging the source should not be
underestimated, as it gives due credit to the curators and donors, and promotes use of the samples for further research. This is infact what is motivating my interest to assign identifiers to tissue samples.
>
>
>
> While in principle it is feasible to use DOIs with tissue samples (by
having them resolve to metadata about the samples themselves) the question arises as to whether a digital object identifier should be used for something real as opposed to something digital? Although this may seem somewhat academic, there is a practical issue in this particular case because tissue samples are gradually consumed and will eventually no longer exist. In which case, I wonder whether the use of DOIs to identify real things is advisable?
>
>
>
> Any advice and/or opinion would be welcome.
>
>
>
> Kind regards,
>
> Miriam
>
>
>
>
>
> Miriam Casula, PhD
> Neuropathology Department
> Academic Medical Centre
> University of Amsterdam
> Meibergdreef 9
> The Netherlands
> Tel: +31205665649
>
> ________________________________
>
> AMC Disclaimer : http://www.amc.nl/disclaimer
>
> ________________________________
>
> ------------------------------
>
> End of RESEARCH-DATAMAN Digest - 17 Sep 2013 to 19 Sep 2013
(#2013-131)
>
***********************************************************************
--
Dr. Jens Klump
Centre for GeoInformation Technology
Phone: +49 331 288-1702
FAX: +49 331 288-1703
E-mail: [log in to unmask]
_______________________________________
Helmholtz Centre Potsdam
GFZ German Research Centre For Geosciences Public Law Foundation State of Brandenburg Telegrafenberg, D-14473 Potsdam
"Digital information lasts forever - or five years, whichever comes first."
(Jeff Rothenberg, RAND Corp., 1997)
________________________________
AMC Disclaimer : http://www.amc.nl/disclaimer
________________________________
|
