I am playing guinea pig with EAD revised alpha. While I did suggest revisions and have been following developments from afar, I was not part of the intense conversations, so please allow for a certain degree of naiveté in the following questions and observations.
Listed in order of encountering them as encoding a small finding aid (not an exhaustive effort to test every element):
1) We are accustomed to marking up our repository as both a corporate name and including a link to the repository home page in one nest of elements, i.e.: <archDesc><did><repository><corpName><ref ... >My Repository's Name</ref>
This is not a big deal, it's just something worth noting: <corpName> will not take a <ref> inside it (also <ref> will not take <corpName> inside). Practitioners and migrating archivists would have to decide among:
· No <corpname> markup, i.e.: <repository><ref...> My Repository's Name </ref></repository>
· Some text for the <ref>, i.e.: <repository><corpName> My Repository's Name</corpName><ref...>Click here...</ref>
· Or stylesheet solutions
2) I struggled (idiotically) over why the schema was insisting on two language declarations until I realized that <languageDeclaration> can be a child of control as well as a child of <languageDeclarations>. If there really is a need for such a wrapper, documentation will be helpful here.
3) I'm not sure where the copyright statement for the finding aid would go. Expecting it somewhere near publicationStmt. It could always go into a generic paragraph, but maybe a special location, since it is rights statement, would be more proper.
4) Since <controlNote> seems to be the only permitted note within <noteStmt> I wonder if noteStmt is necessary in <fileDesc>.
5) I am assuming that the @identifier was for a database number, ORCID ID, or such. Thus, for example, I did: <persName identifier="http://lccn.loc.gov/nr2004015084" role="Creator" rules="AACR2" source="NAF">Warren, G. K. (George Kendall), d. 1884</persName>. Am I right? What would I do if I have two or more such identifiers?
6) chronList is not available everywhere. Must go inside <biogHist>
7) Is <chronologicalSet> basically <eventGrp>? It's a little funny that the schema will allow me to do <event><event><placeEntry><placeEntry> but not <event><placeEntry><event><placeEntry>. Not sure what the intention is here.
8) What is <div> within <scopeContent> intended to do--merhaps merely what it says on the tin, that is, make sections? If so, should <div> be available in <biogHist>, too, to allow for sections like Early life, Family life, Career, Discoveries, Influence, etc? Or perhaps that would make <biogHist> out of sync with EAC-CPF.
9) Still not clear about how to express <controlAccess> for "subject-ness" of non-topical things, and "geographical-ness" of topical things. E.g. corporate name subject, personal name subject, geographic subdivisions. And unsure of best use of localType (if indeed the intent was for this purpose) also an EAC-CPF thing, I think. Which of these makes sense (if either!):
· <corpNameEntry encodingAnalog="61020"> <part localType="corporateSubject">Harvard University</part> <part localType="topicalSubdivision">Buildings</part> <part localType="formSubdivision">Photographs</part> </corpNameEntry>
· <subjectEntry encodingAnalog="61020"> <part localType="corporateName">Harvard University</part> <part localType="topicalSubdivsion">Buildings</part> <part localType="formSubdivision">Photographs</part> </subjectEntry>
And how about:
· <subjectEntry encodingAnalog="650 0"> <part>Black-and-white photography</part> <part localType="geographical">United States</part> <part localType="chronological">19th century</part> </subjectEntry>
I'm happy to send my test EAD file and seasonally-appropriate (Easter colors!) cascading style sheet for oXygen to anyone who requests it.
That's all for now!
Collections Services Archivist
Harvard University Archives
Cambridge, MA 02138
voice: (617) 384-7787
fax: (617) 495-8011