Duplication of deposit requirements
Some authors are affiliated to more than one institution, so their articles are relevant to more than one repository. From an Open Access (OA) point of view, it only matters that they deposit once, to one repository, but from a University administrative point of view, the article might need to be duplicated in different institutional repositories.
In theory, the author ought to be able to choose which repository to deposit to, and the other(s) can harvest from there. RSS feeds are very useful as a way to watch when our authors are depositing in another repository, whilst we're operating on a relatively small scale and for authors with unique surnames. But the problem with an RSS feed is that it alerts me at the time of record creation, and I then have to create a calendar appointment for myself at the end of an embargo period, so that I can get the full text file to add to our repository: this is perhaps a problem that only we at WRAP face, because we only want full text items. In any case, watching RSS feeds is possible for one or two authors, but not on a grand scale.
I wonder if, in the future, we might end up maintaining lists of our authors who have informed us that they deposit elsewhere, and then running yearly processes to harvest records and items from those repositories. We would have to have a way of uniquely identifying those authors in the other repositories. Every repository has a slightly different aim and character and type of content within it, so the data from each repository would need to be handled differently, before it could go in our repository. This sort of a large-scale process is just not feasible for us. (This is pretty much what our medical authors already expect us to do with their content that is available on OA in PubMed - yes, it would be better if the content went round the loop the other way, but in practice, we don't have the resources to format their work appropriately for PubMed anyway. I don't know of any IR that does this - yet.)
There are two other ways around such a process: 1) insist that authors deposit multiple times in multiple repositories; 2) all repositories enable authors to specify other repositories for their content to be sent to, upon deposit.
Clearly, I can't make option 2 happen! There is no reason for an author to deposit more than once at present, because we don't have a mandate. But if we did have one and the other institution had one, we would be requiring authors to deposit their work more than once.
Asking for multiple deposits is nothing to do with OA and our authors will be aware of it and resistant to it. The best reason I can give to authors is that having lots of copies of their work available in OA repositories will help the chances of their work being preserved in the long term. For those medical authors, the long term intention would be to reverse that loop, so the duplication of effort for CR and IR deposit would be a developmental phase only (in theory), but for those affiliated to more than one mandating institution, there is no other way round duplication, that I can see.
Jenny Delasalle
10 Feb 2009 11:32
| 
Comments (5)
| 
Report a problem
Loading…
code Gorilla
I’ve always been bemused by the struggles and contortions we seem to go through to avoid the Multiple Copy situation.
If we actually EMBRACED the idea that a particular deposit will be (a) incomplete and (b) distributed, then all of these issues disappear.
Yes, we need to build an infrastructure that percolates the missing data from where it is to where it’s missing, and resolve the thorny issue of data inconsistency (and two people claming to be authoritative… and disagreeing).... but surely if we’d spent as much effort into getting this sorted as we’ve putting into wanting full text and the “right deposit locus”, it would all be a non-issue
.... or is the problem not really a technical one?
Is the problem a political one? Where the institutions are demanding to hold the one and only copy?
10 Feb 2009, 12:33
Jenny Delasalle
I don’t think that the politics are about who has the copy “of record”. I think the publisher has that, undoubtedly.
I’m not trying to avoid multiple copies… I don’t mind that in the least, but I just don’t see how I can explain to authors that they must deposit more than once. I don’t have to yet, because we don’t have a mandate in force…
My point is precisely about how do we build that infrastructure that gets missing data to where it is missing? Mandates are discussed on grounds of OA and not on grounds of REF returns and admin needs. I think that this is at least a little misleading, unless and until we have a way to harvest records from other places and to ensure that copies are passed from IRs to CRs appropriately. I don’t doubt that technical solutions can be found. They just aren’t in operation yet, and WRAP particularly does not have the expertise to develop any.
In our particular repository, we’re not aiming for a complete record of everything published by every author, precisely because we do want only full text items. But we are working with those at the Uni who do want a record of everything published by everyone (to be stored in what is now being called a CRIS, rather than a repository).
There are other reasons why we have kept WRAP separate from such a system, not least because of the metadata quality. But that is a whole different issue!
10 Feb 2009, 12:59
Theo
@codeGorilla It’s not so much the fact there there will be duplicates existing in multiple repositories that is the problem here, but that authors will be asked to deposit twice. Its a human/ social problem we have on our hands.
You and I both know that there are technical solutions we could implement right now with a little work -> Repository Junction plus SWORD = deposit in multiple repositories.
As a famous author once wrote ’...the future is already here. It’s just not very evenly distributed.’
10 Feb 2009, 16:58
Gavin Baker
Jenny, if you haven’t seen them, I recommend my blog posts on this topic:
http://www.gavinbaker.com/2009/02/05/authors-i-dont-care-where-you-deposit-just-do-it/
http://www.gavinbaker.com/2009/02/09/sussing-out-the-details-on-locus-of-deposit/
11 Feb 2009, 18:23
Sam
thanks for those gavin, really interesting
02 Mar 2009, 17:46
Add a comment
You are not allowed to comment on this entry as it has restricted commenting permissions.