I’m not sure if this will help at all (as I haven’t looked back through the responses) but a similar thread came up over a year ago (search for “normalized dates” on the EAD listserv, or go here: )


In response to that thread, I hastily jotted down some thoughts for a blog post, located here:


And -- as an update to that post -- at ECU we are no longer normalizing dates at the file level.  I wish that we still were, to be honest, but we aren’t really using that data at this time.  Still, for EAD-aggregate databases, I continue to think that it might be a great feature to enable researchers to create completely virtual collections, wherein they could assemble result-sets of every component level that had been encoded with, for example,  “July 1830”. 


Still, it’s hard to justify the time/planning involved if there’s no immediate utility.  But, if you do plan to encode to that level and you don’t have a usage in mind, you could make all of your EAD records easily available on the web so that perhaps someone else could make use of them.



Mark Custer




From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Nathan Tallman
Sent: Thursday, May 06, 2010 9:29 AM
To: [log in to unmask]
Subject: Date normalization


Good morning,


I'd like to take a quick and extremely informal survey.  How many institutions are normalizing dates at the file level?  My past institution only normalized dates at the top, series, and subseries levels.  The rational was that it's not worth the time to normalize at the file level because there aren't enough publishing options available that utilize the data.  (Plus, there's a script to normalize dates if it's needed.)  My current institution is looking at best practices for our encoding and I'd be interested to hear what others are doing .



Nathan Tallman