DefiningImageAccess/Resource/River
From ImageWeb
| DefiningImageAccess/Resource/River | |
|---|---|
| Focus:=Repository}} | |
| [[PublishedBy::{{{PublishedBy}}}]]}} | |
| [[UsedBy::{{{UsedBy}}}]]}} | |
| [[Describes::{{{Describes}}}]]}} | |
| Defining Image Access | DefiningImageAccess/RelatedWork |
River report
http://www.jisc.ac.uk/uploaded_documents/RIVER%20Final%20Report.pdf
The River report is about versioning in repositories, was submitted to JISC about March 2006, and was written by:
- Sally Rumsey & Frances Shipsey, then of London School of Economics and Political Science
- Michael Fraser & Howard Noble, Oxford University Computing Service
- Mark Bide, Hugh Look & Deborah Kahn, RightsCom
Their discussion of versioning touches on one matter that is of fundamental importance to our aims of identifying coreferences in different repositories, "functional granularity" (cf. INDECS project), which assrts that identifying two entities as being the same is only possible in light of functional requirements. (We have considered this issue in the context of data webs, and recognizing this assertion, have adopted a view that the community-specific nature of a data web implies some set of functional requirements for resolving identity questions, but that there are details here that remain to be articulated).
This paper indulges some discussion of what constitutes a "best" version of a document, which I do not see as being particularly helpful to consider. More even than identity, "best" must surely be dependent on functional requirements?
The report notes OpenDOAR as a useful resource, albeit with some limitations.
The report contains a survey of repository software systems that contains some information about the metadata handling provided (thopugh somewhat oriented toward versioning metadata rather than image metadata, but still of some relevance to our purposes - e.g. notes about metadata access protocols). It is also interesting to note that the report treats wiki software as a kind of repository software, but lacking metadata exposure (I note that Semantic Media Wiki may rtepresent a departure from the last point).
The survey includes a fair amount of useful information about PubMedCentral (with reference to Entrez for maching-to-machine metadata interactions).
The conclusions section rasies a number of issues and recommendations that are likely to be highly relevant to our work with repository metadata. The metadata-related recommendations of the report, if enacted, are likely to be of considerable assistance to the deployment of image data webs, assuming the metadata is available for images as well as complete publications. Finally, the reports and referemnces section reads like a "who's who" of areas that have been identified to us as having a bearing on our goals.
