Nor would <container type="1.0 cu. ft. box">1</container> validate,
since the type attribute here will not allow spaces. I suggest
thinking of type as a kind of normalized key which can then be used
to select a label. That is type="box" renders as "Box" -- or "Carton"
if you have a localized interface in French, say.

Even if the text of the label is predictable, the 'label' attribute
is probably the better choice:

<container type="box" label="Box">1</container>

-Brian

On May 19, at 11:55 AM, Michele R Combs wrote:

> Interesting approach. I'm not clear what you mean by "relationship
> encoding" -- could you post a sample of how one of these postcard
> boxes
> would look in EAD? You put this in the "type" attribute, perhaps?
> Now
> we use the type attribute along with the contents of the <container>
> element to display the name of the container to the end user, viz:
>
> <container type="Box">1</container>
> displays in the online finding aid as
> Box 1
>
> If I do
> <container type="1.0 cu. ft. box">1</container>
> I'm going to get
> 1.0 cu. ft. box 1
> which doesn't seem as easy to read/understand, plus it doesn't
> match the
> label on the container which simply reads "Box 1."
>
> Michele
>
> -----Original Message-----
> From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf
> Of Russell D. James
> Sent: Monday, May 19, 2008 12:02 PM
> To: [log in to unmask]
> Subject: Re: Weird containers
>
> Michelle,
>
> The first place I trained as a lowly graduate assistant had a way of
> doing this for non-standard and standard containers before the
> advent of
> EAD that I've found works well now that we have EAD. They have every
> box measured in its relationship to a cubic foot, their standard (and
> many other archives) in measuring. So the normal "Hollinger box" that
> measures 10"x12"x15" is close to a cubic foot and is the largest box
> they use so they call that a "1.0 cu. ft. box" and then there is the
> smaller size everyone makes that is roughly 10"x15"x4" and they call
> that a "0.33 cu. ft. box," and so forth.
> If every box is measured by its relation to a cubic foot and is
> standardized in your repository, then it works fine to know what
> size of
> boxes you are looking for when you go searching.
>
> It also helps to add the function of the box - for instance, if a
> box is
> created to hold postcards and they fit just perfectly, then call it a
> "0.xx cu. ft. postcard box" or if it is one for artifacts, then
> call it
> a "0.xx cu. ft. artifact box." EAD allows this type of relationship
> encoding (at least, I've never had the parser say "no").
>
> Russell
>
> -----Original Message-----
> From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf
> Of Michele R Combs
> Sent: Monday, May 19, 2008 8:16 AM
> To: [log in to unmask]
> Subject: Weird containers
>
> A question for the collective wisdom: how do you encode material in
> non-standard containers? We have a set list of "accepted" terms for
> primary <container> element (Box, Package, Oversize, Mapcase, etc) and
> secondary (Folder, Drawer, Reel, etc). Occasionally we have items
> that
> are in something weird -- a metal box, for example, or a camera
> case, or
> a canvas bag, or a tube, or...
>
> Is it better to only use the <container> element for a consistent list
> of standard container types, and put the oddities in the abstract or
> other descriptive element? Or should one use the <container> element
> always, even for strange things? I can see arguments for both
> approaches and wondered what others have done.
>
> Thanks
>
> Michele
>
> ++++++++++++++++++++++++
> Michele Combs
> Manuscripts Librarian
> Special Collections Research Center
> E.S. Bird Library
> Syracuse University
> 222 Waverly Ave.
> Syracuse, NY 13244
> ++++++++++++++++++++++++

--------------------------------------------------
Brian Sheppard
University of Wisconsin Digital Collections Center
[log in to unmask] (608) 262-3349