I have the same problem with my quilt collection inventory which you
will see at MCN.
> The information you are including in the label attribute, e.g.,
material-materialname was intended to go into either an element called
<facettype> or an attribute of the same name in the <physfacet> element.
In the rush of last minuet changes, it didn't make it into Beta but I
have suggested its addition to version 1. I personally prefer the
Would this be a satisfactory solution?
>From: Richard Rinehart[SMTP:[log in to unmask]]
>Sent: Thursday, October 02, 1997 4:47 PM
>To: Multiple recipients of list EAD
>Subject: physdesc/physfacet feedback
>I've been using the EAD to encode finding aids to art collections in our
>museum (which are online at
>http://www.bampfa.berkeley.edu/search/collectionguides.html) and I'd like
>to offer some feedback on changes to the DTD I think might be useful. I
>know the EAD comes out of the library/archive worlds, but it is already SO
>enormously useful and much-needed in the museum community that I think some
>minor considerations which could benefit all, could also make the EAD even
>more precise and useful for museums.
>Some feedback has already been given on the need for more detail in the
>PHYSDESC/PHYSFACET tags, which is where I also feel a little tweaking is
>needed for collections which need to provide more precise information on
>the physical aspects, since these are primary access points for us.
>Below I've included some tagging along the lines I've been using for
>describing typical museum/art objects. This tagging is currently legal, but
>is not perhaps what was envisioned with the PHYSDESC tag so far. In
>addition to the AAT for controlled vocabular terms, we use the CDWA
>(Catagories for the Descriptions of Works of Art) as the structural
>organization for our museum records. The CDWA is not about syntax at all,
>and is very flexible that way, allowing us to use it as a model for our
>collections database, and (hopefully) in our related systems such as
>finding aids to allow integration.
><UNITTITLE>Yellow Table on Yellow Background</UNITTITLE>
><PHYSDESC SOURCE="OTHERSOURCE" OTHERSOURCE="CDWA"
><DIMENSIONS>41 x 50 inches</DIMENSIONS>
><PHYSDESC SOURCE="OTHERSOURCE" OTHERSOURCE="CDWA"
><PHYSFACET>oil on canvas</PHYSFACET>
><GENREFORM SOURCE="AAT" NORMAL="paintings">
><PERSNAME SOURCE="ULAN" NORMAL="Hofmann, Hans" ROLE="artist">Hans
>Basically, I found I needed more detail than the EAD currently allowed -
>for instance we needed to indicate materials used in the creation of an
>object specifically, where the EAD would put that under the broad tag of
>PHYSDESC or PHYSFACET. We need to separate this element out from other
>elements in the physical description of the object, such as processes used
>in it's creation, or dimensions. So, to add granularity to the PHYSDESC I
>used the available attributes of source to indicate where I was getting my
>lables from, and label to indicate the actual term taken from the CDWA that
>describes that facet of it's physical description. This level of info seems
>like it would make more sense at the PHYSFACET level, but is not available
>there yet. I used hyphens to indicate the level in the heirarchy of the
>CDWA that the term occurs at since there are duplicate terms at different
>Basically something like this allows the person marking up to use the
>general EAD tags which are themselves useful, but also to add some
>subject/domain-specific structure and detail to them, so art objects might
>get extra CDWA-derived attributes added, medieval manuscripts could use
>another organization, all in the same framework. If a "source" and "label"
>attribute were added to the PHYSFACET element, it would not be overly
>complex, EAD users could ignore them or use them. It would maintain the
>interoperability of all EAD documents because it does not add an entire
>specialized element, and yet it would allow more precise retrieval of
>certain types of records, especially in environments of finding aids for
>mixed types of collections. (I'm thinking of this one in particular as we
>are contributing our finding aids the a larger "UC-EAD" project, which will
>have a smattering of art objects/finding aids among many archival finding
>aids). Although "label" used to be an attribute of the PHYSFACET element, I
>feel that "source" would also be important for precision, since say
>medievalists may use the same term as art catalogers ("name" or
>"material"), but use it in very different ways.
>So, although it might seem like unneeded complexity, I can think if
>immediate practical benefits already. The CDWA comes from the Getty along
>with the AAT and ULAN and could be added as an allowed source, or left open
>for the user. I'd be curious to hear if current markup can already do what
>I'm looking for another way, or if this would create confusion about what a
>"controlled vocabulary" is or any other comments as I'm working with this
>Richard Rinehart | Berkeley Art Museum/Pacific Film Archive
>Systems Manager & Education | University of California
>Technology Specialist | 2625 Durant, Berkeley, CA 94720-2250
>[log in to unmask] | http://www.bampfa.berkeley.edu/
>& Board of Directors, Museum Computer Network, http://www.mcn.edu/