Thanks for this Amber and to the others who have been contributing to this
discussion.

To return to what I've realised was my actual point:

This email list is for discussion of repository issues across the range of
repository types or domains or whatever: research output repositories (be they
institutional or wider); learning materials repositories; scientific data
repositories; imagebanks; assessment itembanks; eportfolio repositories, etc.
(all of which may be "institutional" or not). Therefore, when we are discussing
things on this list it's important to keep in mind the range of communities
members come from and be clear that not everyone uses the same terms as everyone
else. Different communities have their own terminologies as well as their own
drivers and their own standards and their own separate history and their own
priorities (and, dare I say, prejudices?).

The term "institutional repository" has a very clear meaning in one domain.
This doesn't mean that others understand that meaning.  It is also a term that
may be too vague for those outside the "IR" domain- there isn't anything in it
that implies "research outputs" or "research papers" to an outsider. It looks
very general.

I've noticed that since sending the initial email, I've started (on-list and
off-list) saying "repositories of research outputs" to distinguish this type of
repository in my own communications. I'll continue to do that where I think it's
needed in my own communications as long as I find it useful (and add
"institutional" where that is useful).

Just so the IR people don't feel picked on- I was asked last week why on
earth "data repositories" was listed as a separate type on the DigiRep wiki when
surely ALL the repository types hold data. I explained that this is shorthand
for "repositories of scientific data from experiments" (at least I think that's
what it is- I'll probably cause another avalanche of emails just with that
statement). And don't get me started on how the term "learning object
repository" looks to the uninitiated (check out the name of my DigiRep project
below in my signature!).

This list could be said to be forming a broader community of practice
encompassing a number of established communities of practice. I don't know but I
strongly suspect that the "IR" contingent is in the majority. That can affect
dynamics in any group.

I do strongly believe we can all learn from each other.

Apart from that, in terms of defining "repository" I can't really add much to
what Phil and Amber have said- they said it well (although I do think the
discussion there is useful too, e.g. Richard Green's comments).

Best
Sarah



Quoting Amber Thomas <[log in to unmask]>:

> 
> Hi All,
> 
> Fascinating discussion, and very reassuring that we're identifying the same
> kinds of issues. I agree that "repository" is a pliable  notion, especially
> when we move beyond the e-prints model into  learning and teaching, research
> data, e-admin content.
> 
> Firstly, are we mixing up "the Institutional X" and "an institutional x"?
> I'm
> sure it makes a big difference to JORUM whether its seen as "the National
> Service" or "a national service". The capital letters have big connotations!
> And as Phillip Hunter said, some services end up institutionally managed
> because of funding as much as anything else.
> 
> My tuppenceworth is that there are indeed different ways of viewing "a
> repository" ...
> 
> Repository as software
> define it by its functionality for getting, putting, discovering and
> retrieving
> illustration: the whole VLE/MLE/LearningPlatform debate. We know you can
> provide VLE functionality without a particular piece of "VLE software".
> Maybe
> this is the same. The E-framework allows us to discuss functionality neutral
> from any particular piece of software, hence reference models etc.
> 
> Repository as filestore
> define it as a collection of digital content stored/managed in a particular
> way
> illustration: lots of FE colleges are using a repository mainly to access
> their NLN materials because a repository is good for managing a big set of
> content packages. That seems to be looked down on as undeveloped repository
> use, but why? There's a storage issue, and the repository solves it. Why
> should it do more than that?
> 
> Repository as a set of practices
> define it as a set of processes and policies around copyright clearance,
> metatagging, submission, reviewing etc
> illustration: archiving
> There's archiving activity, and there are archives. You can undertake
> archiving practices without focusing on any particular archive. One person
> might manage their C:drive more carefully than a whole institution might
> manage its repository!
> 
> 
> Personally, I think it's the practices that are the starting point. If a
> repository is a suitable solution then people will see the need for it. If
> they've already got a service mix (intranet, a VLE, a network drive) that
> supports those practices then in a sense they've already got repository
> functionality.
> 
> 
> 
> Do definitions matter? They do to us, but not to many others. If I could
> have
> a pint for everyone that's told me google is a repository ... :-)
> 
> 
> 
> Amber
> 
> 
> 
> Amber Thomas
> Project Leader
> JISC DEL: Promoting Use of Shared Content in the West Midlands
> Bredon 76
> University of Worcester
> Henwick Grove
> Worcester
> WR2 6AJ
> 
> email: [log in to unmask]
> mobile: 07913 842421
> http://www2.worc.ac.uk/wm-share/
> 
>