So now we know that it technically can be done, but should it be done? 


I personally agree with Joyce’s earlier post.  The UNC symposium was certainly a thought-provoking and interesting day, with many perspectives voiced about privacy and ethical issues surrounding the large scale digitization of entire manuscript collections.  


Most of the time,  decisions about use and restrictions are made at the collection level upon acquisition, as are our decisions about processing priority.  Most repositories discourage restrictions, or, at the very least, try to put an end date on the restriction.  Prior to EAD online finding aids, most of us probably would not have given a second thought as to whether restricted portions of a collection would be mentioned in the container list of a finding aid.  


During later processing – which is often when the EAD finding aid is created – the processing archivist may discover individual series or files that contain records that should be restricted for either privacy or ethical issues (social security numbers, medical records, etc.) .  However,  the fact that this material exists within the collection should not be hidden from researchers.  The EAD finding aid merely points to and describes the collection, it does not display the collection.   If personal names need to be hidden for some reason, why not simply change the folder title to something more generic ? And, use the appropriate EAD tag to offer information about use or access restrictions – which may change in the future.  


Having said that, I do think it’s important to be more careful when digitizing entire collections and presenting them online.  With all of the talk about large scale digitization of manuscript collections, I’ve noted that many take the position that large scale digitization is defined as digitizing everything in the collection.  As desirable as that may be, I feel that there are certainly items/files within the collection that should not be digitized.  These will most probably be discovered and identified by the processing archivist, who also creates the finding aid and, who, in my opinion, should be making the “final cut” (appraisal) for full digitization.  


The beauty of having the digital files/digitized collection presented online with, or as part of the EAD finding aid is that researchers know immediately what is available online and what is not.   And again, should that material become unrestricted, the digital files can be integrated at a later date, without changing the finding aid.    I think it’s also important for us to be realistic about workflow and resources and try to minimize the effort required for additional encoding, later changes, and individual file or item redaction in the finding aids.   


Barbara D. Aikens


Chief, Collections Processing

Archives of American Art, Smithsonian Institution

Ph: 202-633-7941

email:  [log in to unmask]


Mailing Address

Archives of American Art, Smithsonian Institution

PO Box 37012

Victor Bldg., Suite 2200, MRC 937

Washington, DC  20013-7012




From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Custer, Mark
Sent: Tuesday, April 21, 2009 12:22 PM
To: [log in to unmask]
Subject: Re: Ethical issues raised by EAD encoding


Michele, that makes perfect sense (and I completely overlooked that since I’ve yet to use that attribute)! 


So, since I’ve never used @audience, it would be easy for me to use it exclusively for redactions (i.e. “internal”) if so needed.  However, if someone else is using “internal audience” for something slightly different, I suppose that that wouldn’t be the case.   But that still leaves questions in regards to sharing EAD records.  Either way, though, I don’t think that it would permit you to set an expiration date (unless I’m overlooking yet another general attribute or simpler solution)?


As for the bigger picture, a processing archivist currently has control over a finding aid that he/she authors.  They are able to make the decision about the level of identifiable granularity that they want to provide (though these decisions are certainly not impartial, or always fully considered), but there’s a question lingering about whether such a decision – whether documented or not –  will be upheld after the collection starts to go online and/or if finding aids begin to incorporate editable features (for example, by permitting researchers to add their notes). 


In the case of fully digitized collections, though, this is simply not a question anymore, as there will be many items in the collection that haven’t been fully looked at or comprehended before being made accessible to many-more-than-before.


This is fascinating from a research/access point of view, but it will inevitably produce new privacy issues (legal, ethical, and both) that will need to be addressed.  There is little denying, I’d contend, that a search engine has the power to re-bestow currency, if only temporarily, to outdated or even false information.  And, in my opinion, that’s certainly something that should be considered during this transition.





From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Michele R Combs
Sent: Tuesday, April 21, 2009 10:50 AM
To: [log in to unmask]
Subject: Re: Ethical issues raised by EAD encoding


Or you could just use the “audience” attribute which is available for all EAD elements.  Surround the name in question with a PERSNAME element, set the @AUDIENCE to INTERNAL, and make your publishing process create whatever visible indication you want – a blank, a black bar, the word [name redacted], whatever.




(be green - don't print this email!)


Michele Combs
Manuscripts Librarian
Special Collections Research Center

Syracuse University Libraries
222 Waverly Ave.
Syracuse, NY  13244


[log in to unmask]




From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Custer, Mark
Sent: Tuesday, April 21, 2009 10:35 AM
To: [log in to unmask]
Subject: Re: Ethical issues raised by EAD encoding


This, of course, led me to wonder if EAD should have a redacted tag (or attribute