I want to put forward what I believe is one compelling use case for a standard ingest API/Interface/Protocol (call it what you will). We use DSpace here at [Oregon State University], and as everyone knows, one of the biggest barriers to faculty submitting materials into an institutional repository (i.e. not DSpace-specific) is the amount of metadata they need to enter when submitting a resource. Now, say a faculty member has a large set of related (such as they are all part of a series) images that they are trying to submit into the repository - with DSpace (and other repository software, I’m sure), they are required to re-enter, for every image, every metadata field value, even if it is the same value for each image....Now, DSpace does indeed have one way of doing this - by setting up a default template for a collection. Unfortunately, this isn’t an acceptable solution, as we would need to create a new collection every time a new, unique series of resources was to be submitted. What would be better is if we could build a user interface that provided the user an option to "carry-over" their metadata from their previous submission.
In order to accomplish this "carry-over" feature, there are at least two approaches. The first approach would be to dive into the DSpace code, and enhance the software to do what we want....The second approach would be to create a separate web app that allowed us to easily create a custom UI / workflow particular to the problem we are attempting to address, and then have that app automatically submit the resources into our repository. The plus-side here is that we aren’t needlessly monkeying around with the DSpace code, we can use whichever programming language / framework that is comfortable to us (such as Ruby on Rails), and we don’t have to upgrade this system every time DSpace is upgraded - unless the method of submitting items into DSpace changes.
And this brings us to the point: If DSpace defined a standard, consistant ingest interface (or API or protocol...), then this would allow us to build these types of apps quite easilly. And, if there were a standard ingest that was implemented by all scholarly repositories, then if we were decided to step away from DSpace, or wanted to deposit items into more than one repository, we could - without needing to re-program our app. To me, that's a fairly convincing use case for a standard ingest interface for scholarly repositories.
Posted by
Peter Suber at 6/24/2006 02:47:00 PM.
The open access movement:
Putting peer-reviewed scientific and scholarly literature
on the internet. Making it available free of charge and
free of most copyright and licensing restrictions.
Removing the barriers to serious research.
