Re: your first para, I just wanted to note that it is quite dangerous to
use the word 'item' in this kind of discussion without being clear which
usage of 'item' is intended.  I'm also not quite clear what you mean by
"the files" :-)

'Item' is used by both the OAI-PMH (essentially for a conceptual
collection of metadata records - proper definition in my previous
message) and by FRBR (for a "single exemplar of a manifestation") - i.e.
it is used in two completely different ways.  In SWAP, we chose to use
'copy' rather than 'item' partly for this reason:

http://www.ukoln.ac.uk/repositories/digirep/index/Model

The question of "what is of interest in a repository?" is still open I
think.  In SWAP, we argued that 5 key entities are of interest -
Scholarly works, expressions, manifestations, copies and agents - of
which only copies and agents are concrete physical/digital entities, the
others are conceptual.  So we certainly said that more than just the
"files themselves" are of interest to the outside world.

Why?  Because, we argued, the easiest way for the outside world to
understand the relationships between any two copies (i.e. between any
two "files") is to understand the hierarchy of manifestations,
expressions and scholarly works "above" those "files" and the
relationships between those things.  This is not the only way to model
the scholarly communication world of course.  It (arguably) fits well
with the direction of travel in the library cataloguing world but, as I
mentioned before, it is conceptually quite challenging and that may be
its downfall.

In the SWAP work we toyed with leaving out one or more of the layers
(e.g. merging scholarly work and expression into one entity) but always
came back to requiring the 4 layers to correctly model the scholarly
communication space.

Andy
--
Head of Development, Eduserv Foundation
http://www.eduserv.org.uk/foundation/
http://efoundations.typepad.com/
[log in to unmask]
+44 (0)1225 474319 

> -----Original Message-----
> From: Repositories discussion list 
> [mailto:[log in to unmask]] On Behalf Of Les Carr
> Sent: 07 May 2008 08:25
> To: [log in to unmask]
> Subject: Re: Google, OAI and the IRs
> 
> On 6 May 2008, at 17:34, Andy Powell wrote:
> >> From the perspective of the "web", the "resources of interest" (or 
> >> just "resources", since that is a key piece of terminology in the 
> >> definition of the web) are the individual files that can be 
> >> downloaded via the web (HTTP) protocol.
> > Yes... though note that the Web architecture use of "resource"
> > encompasses anything that is of interest including digital objects, 
> > physical objects and conceptual entities - so it is much 
> more than the 
> > "files that can be downloaded".
> 
> Agreed, but I was responding to your point that the resources 
> of interest in a repository are the FILES and that OAI had 
> chosen to identify the wrong things. I was making the counter 
> point that the resources of interest in a repository are 
> abstract bibliographic items (items in the literature), not 
> the files themselves. At least that was the case for the 
> "repositories" represented at the OAI meeting.
> 
> > "Note that the identifier described here is not that of a 
> resource.  
> > The
> > nature of a resource identifier is outside the scope of the 
> OAI-PMH."
> > so in that sense the OAI-PMH ignores (i.e. doesn't model) the 
> > "resources of interest".
> Not of your interest, no :-) It is certainly possible to 
> argue that the "OAI-PMH items of interest" are not of 
> interest to anyone but OAI service providers!
> 
> >  Hence the need for initiatives like SWAP which say, 
> "here's a view of 
> > the world that we need to share metadata about" and "here's 
> the kind 
> > of metadata we want to share about the entities in that world-view".
> 
> Quite - it's the worldviews that need to be improved.
> 
> OAI defines one worldview - and a potentially very skewed one 
> at that.  
> It is a worldview that is about the bulk exchange of 
> bibliographic metadata. SWAP - which is based on FRBR - 
> provides another world view that is based on the world of 
> cataloguing publishers' products. The Web provides a 
> worldview which is all about serving "information resources".
> 
> None of these reflects the reality of a repository accurately 
> enough to describe a repository's holdings in terms of the 
> modern scholarly communications processes and the 
> scholar-as-information-provider- worldview.
> --
> Les
> 
> PS By the way, getting back to the original topic, I think 
> that this split between worldviews makes it very difficult 
> for a repository to serve an accurate Google SiteMap. I don't 
> think any of the platforms keep track of when a particular 
> web page changes, but we all know when an item's metadata has 
> changed. The two are not the same, especially when repository 
> software is upgraded, or Institutional Visual Identity or 
> Branding is changed. URIs may be persistent, but the HTML 
> files that they resolve to may undergo all sorts of changes 
> that a repository fails to keep track of.
>