You (and Jerry) have argued convincingly that METS does not have to
be shackled to the OAIS IP. However, I have not heard any convincing
arguments that it shouldn't be.
I guess I'm unclear on what about the METS classification of metadata
makes it more useful than borrowing the OAIS categories would be. For the
project I'm involved in, it would be less so, in part for reasons
cited in the "METS vs. OAIS" email. As far as I'm concerned, METS is
very valuable for (not an exhaustive list, just my favorites)
1)allowing granular linking to files
2)allowing for metadata extensions
3)linking divs to amd and dmd and files to amd
4)implementation in XML schema
The taxonomy of metadata, however, is not one of METS' strong points in my
The OAIS treatment of metadata is flexible, complete, and very well
defined. On the other hand, very nice tools and extension schemas have
been designed based on METS 1.0. So even if everyone agreed with me on
the superiority of the way OAIS organizes metadata, there would be a very
understandable issue of inertia.
On Fri, 23 Aug 2002, Brian Tingle wrote:
> At 04:28 PM 8/23/2002 -0400, Clay wrote:
> >My argument was and remains that METS does not map simply and
> >unambiguously to the OAIS data model. I think there is one true
> >OAIS data model, and that METS does not map to it as simply and
> >unambiguously as it could.
> Maybe I don't understand the point of a reference model, but I thought it
> was for, well, reference. You are not supposed to implement a reference
> model. You implement protocol specifications.
> When different people implement things differently to meet local needs,
> then the reference model gives them a common vocabulary, so that when they
> sit around at lunch at conferences or write to each other comparing what
> they have done, they can understand what each other are talking about.
> In _Understanding OSI 7-Layer Model_, Neil Briscoe notes "that most
> protocols in day-to-day use work on
> a slightly modified layer system. TCP/ IP,
> for example, uses a 6-rather than a 7-layer model. Nevertheless, in order
> to ease the exchange of ideas, even those who only ever use TCP/ IP will
> refer to the 7-layer model when discussing networking principles with peers
> from a different networking background."
> My feeling is, if network engineers don't worry that their systems don't
> implement their ISO reference model, we should not worry if our
> implantations of archives don't match up paragraph and sub paragraph to the
> reference model.
> -- Brian Tingle
> Development Programmer, California Digital Library