On Feb 21, 2008, at 3:40 PM, Melanie Schlosser wrote:

> I'm working up a METS profile to describe a complex, distributed
> electronic text project, with the ultimate goal being a preservation
> information package of the whole thing. The edition itself takes the
> form of a Cocoon-based website that delivers up page images, HTML
> dynamically generated from the TEI-encoded manuscript, and contextual
> materials (including text, images, audio, and video). The edition as a
> whole does not have a clean hierarchical structure - it is rather made
> up of a complex interaction of independent parts. I am planning on
> including a structMap (I think it would be type="physical") with a
> shallow div structure, based on grouping files by type and purpose
> (e.g. a div for master page images, one for TEI files, etc.). It seems
> redundant to me, since I will also be organizing the fileGrps in a
> similar way, but it's the only way I have come up with to represent
> all of the 'pieces' in one structMap.
>
> I would like to include a logical structMap, as well. One option seems
> to be to create a 'partial' structMap for one or more discrete
> sections of the project. For example, I could have a structMap
> representing the manuscript itself, organized by page, with links to
> page image files and the appropriate sections of the TEI (leaving out
> the contextual materials entirely).
>
> My question is, what has to be represented in the structMap? Does it
> need to describe the entire object that is the subject of the METS
> doc, or can it be more narrowly focused and include only some of the
> component files?
Minimally mets requires at least one structmap with one empty div  
child element. That is enough to validate against the schema. That  
being said, it doesnt need anything but that. It can be as narrow as  
no fptrs and as expansive as you can put in a finite file.
>
>
> Thanks,
> Melanie
>
> -- 
> Melanie Schlosser
> Metadata Librarian
> Scholarly Resources Integration
> The Ohio State University Libraries
> 614-688-5877
>

Francesco Lazzarino
Florida Digital Archive Programmer
[log in to unmask]