That’s a great question, as I bet there’s a lot of variation out there. As for “undated” dates, we’ve decided not to use the normal attribute in our encoding for those (but we do use the normalized attribute for every other date instance). That said, the RLG Best Practice Guidelines for EAD suggests that if you know of certain cutoffs for the dates involved (see page 6), you can encode those like so:
However, I actually don’t know of anyone that’s currently using the normalized dates for much of anything (aside from, “if the date occurs in the inclusive span, return result”). I would like to use them, for a number of different services, but right now they aren’t being used at all really. (Perhaps they are being used by such systems like the Archivist’s Toolkit?).
I am conducting an informal poll as to how organizations are using normalized dates.
Brown University and members of the Rhode Island Archival and Manuscript Collections Online consortium (RIAMCO) have switched from dtd to schema so our prior normalized format for undated materials is no longer valid:
<unitdate encodinganalog="245$f" type="inclusive" normal="">undated</unitdate>
Using the schema, the following two examples will validate:
1. <unitdate encodinganalog="245$f" type="inclusive">undated</unitdate>
2. <unitdate encodinganalog="245$f" type="inclusive" normal="0000">undated</unitdate>
- How are other organizations formatting the normalized for undated materials? Right now we are leaning towards option 1 for undated materials.
- Regarding normalized dates in general, what do organizations use as their rationale for using or not using normalized dates? Our current plan is to keep using the normalized format for all dated materials but we are most curious as to what might be the argument against using normalized dates.
Thanks in advance!
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
Jennifer J. Betts
John Hay Library, Box A
Providence, RI 02912
E-MAIL: [log in to unmask]
TEL: (401) 863-2148
CELL: (401) 480-1173
FAX: (401) 863-2093