Hello:
Regarding opening components within components.

I have been working on a series of documents which contain other documents and have found a problem expressing
their structure in EAD.

There is no problem with a document that encloses another separate piece, thus (simplified):
<c05>LETTER ABOUT ENCLOSURE
<c06>ENCLOSURE</c06></c05>
or even a series of enclosures, providing that they are separate and you have finished describing the <C05>
document before you start on the <c06>'s.

What does not seem possible is to render the structure of a document that cites other documents while returning
between these citations to give more description of the base document.
An example of this is the inspeximus, where an existing set of documents are recited and confirmed. One goes like
this (and consists of 8 documents):

<c02>Inspeximus by Richard II confirming:
<c03>Inspeximus by Edward III confirming:
<c04>[Cross ref to existing document]</c04>
<c04>[Cross ref to existing document]</c04>
<c04>Document (given in full as no earlier surviving example)</c04>
More description of the <c03> document  ...[</c03>]
More description of the <c02> document (includes more <c03>'s)and ending with a list of witnesses and a
seal ...[</c02>]

only it can't (the option at present is to move the description of the end of physical document to join the rest of the
description of it at the start).  When you close a subordinate component (such as </c04> above) the the only options
are to close the next component up, open another at the same level, or open a <thead>.
Given that the initial component tag is still open, would it not be possible to continue describing that component
without compromising the overall structure of the finding aid, and is it possible to express this in the DTD (I suppose
for a start it would mean allowing more than one <did> ...)?
This does not just involve obscure medieval documents, as it is also necessary to describe the contents of bound
volumes which frequently contain items within the body of other items.  At present the EAD DTD does tend to be
biassed towards cataloguing runs of individual separate documents, so perhaps this thread could start to redress
that.

Richard Higgins
Durham University Library