[SPG_Active_Members] Fwd: Standardization of schema for a "Preserved Digital Object"

[Richard]

In article <CAMn7poAb7S7Kx+HZd64zzr3OjNWUjpyB-MtQo+1CA=bkHvSB-A at mail.gmail.com>,
    Henry Gladney <hgladney at gmail.com> writes:

> Ref.: HMG, *Preserving Digital
> Information*<http://www.springer.com/3-540-37886-3>,
> Springer Verlag, 2007 ISBN 978-3-540-37886-0

At ~$100, this is a bit of a pricey reference.  I'd prefer that
standard references be available online at least in summary form.
Particularly since I can't pop over to the local bookstore and skim
this book to see if its worth $100.  Maybe I'm just cheap :-).

> Don Post has emphasized that, if the thinking underlying TDOs (Trustworthy
> Digital Objects) is to have any large effect, it must be represented in a
> widely accepted data interchange standard.

Are we talking about something like the Dublin Core profile?

> What is interesting about a TDO is its definition of structure--a definition
> that uses only 4 concepts: (1) that a thing can contain another thing, which
> might itself be a TDO; (2) that things, or places within them, have useful
> relationships to other things, or to places within those other things;[1]
> (3) that things can be uniquely identified by name or by location; [2]
> (4) that the recursion implicit in (1) is grounded in objects of well-known
> types--e.g., MS Word files, other kinds of files, objects in the real
> world, or any other kind of thing you can think of.[3]
> 
> Obviously, this needs to be fleshed out with references to other digital
> information standards, such as those for asymmetric key cryptography.

I'm in the process of building a catalog of my computer graphics
museum artifacts based on the Dublin Core profile with some minor
additions/extensions.  I had come to a similar conclusion regarding
what I needed to add: a hierarchical containment relationship was
needed to describe whole systems decomposed into smaller and smaller
assemblies, i.e.

    SGI Onyx Reality Monster
        Rack 1
            Module 1
                CPU Card 1
                CPU Card 2
                ...
                CPU Card 8
            Module 2
                CPU Card 1
                CPU Card 2
                ...
                CPU Card 8
			Power strip
			Skins
			Cables
        Rack 2
            Module 1
                Graphics Card 1
                Graphics Card 2
                ...
                Graphics Card 8
            Module 2
                ...
			Power strip
			Skins
			Cables
        Rack 3
            ...
        Rack 4
            ...

Because my museum is a hands-on working equipment museum, I need to
swap parts of different provenance to create working systems and I
need to track all this metadata of the parts in order to have a
tractable history on all the items.

This might descend all the way down to provenances on individual chips
and other electrical components when component level repair is
required.

For digital objects, the audit trail of an individual object isn't
required so much as the similar containment and provenance metadata.

I'm consulting with an archivist currently to create my initial
catalog schema for use with Collective Access
<http://www.collectiveaccess.org>.  I will publish the schema once the
initial version is available.  It might serve as a useful basis for
the proposed metadata of digital objects.
-- 
"The Direct3D Graphics Pipeline" -- DirectX 9 draft available for download
 <http://legalizeadulthood.wordpress.com/the-direct3d-graphics-pipeline/>

      Legalize Adulthood! <http://legalizeadulthood.wordpress.com>