The organization suggested below is a leftover from the days of typed finding aids when the human eye did the searching on the page.  The human eye - brain can look at this organization and realize what  it really says: Abilgard, Mark , General ----  and so on.  It is  natural to combine that information.  At our archives we used to have many finding aids in that fashion.  As we have computerized  we have abandoned all of that because it splits up too much information which should be kept together.  In a simple type of computer search which displays folder titles the arrangement below would display  " General, 1989-1998" but wouldn't inform you that it was the file of Abilgard, Mark.   To do that would require additional processing of each hit to see if it had a parent which contained the rest of the title. Note how Abright, Bill does contain it and for no other reason than it has one file.  I would call that an inconsistency in the structure. It is like mixing sub-series and individual files in the same series.  So,  each search result for a file will need to have  a routine to look for a level which may or may not be there.   The benefit of the EAD, and searching in general, is to allow researchers to search across our finding aids.  Often the only reference they need is to see file folder titles.  And the search display result doesn't need to be much more than a simple table display.    So, I would say that each folder needs to have its complete title, and information, because it may well have to stand on its own in a display.  Encoding which has Abilgard, Mark - General, 1989-1998 preserves the entire information together and allows it to be displayed with a minimum of effort. 

Just my thoughts. 


Michele R Combs wrote:
[log in to unmask]" type="cite">

FWIW, we use this construction that Barbara cites quite often.  Just seems to make a lot of sense, both intellectually and XML-ly.




From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Aikens, Barbara
Sent: Thursday, April 15, 2010 12:23 PM
To: [log in to unmask]
Subject: Re: Promoting Container Levels


I wouldn’t necessarily call the example provided as a “false level”.   There are times when one needs what I refer to as a “folder grouping” and I believe it is an intellectual component level that also works well for display.   


For example, in plain text:


c01:       Artists Files


                cO2        Abilgard, Mark  

            cO3     General, 1989-1998

            cO3     Printed Material, 1989-1998, undated

cO2     Abright, Bill, General, 1987-1999, undated  

cO2     Adams, Hank Murta

cO3     Announcements and Press Releases, DW Gallery, 1985-1996  

cO3     Collector Records, 1993-1997, undated

cO3     Consignment Records, 1991-1996  

cO3     Correspondence, 1981-1997 (2 folders)  

cO3     Loan Forms, 1996  

cO3     Printed Material, 1985-1997  

cO3     Shipping Records, 1989-1997, undated

cO3     Photographs of Adams, undated  

cO3     Other Photographs, circa 1981-1995 (3 folders)  

cO2     Aebersold, Jane Ford, General, 1982 1

cO2     Alexandrov, Simona, General, circa 1994-1996, undated 1

cO2     Anderson, Daniel, General, 1987-1996




Barbara D. Aikens


Chief, Collections Processing

Archives of American Art, Smithsonian Institution

aikensb @



202-633-7994 (fax)


From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Ethan Gruber
Sent: Thursday, April 15, 2010 8:21 AM
To: [log in to unmask]
Subject: Re: Promoting Container Levels


The way I see it, the important organizational information is contained in the @level attribute, so I avoid using numbered components at all.  Just using <c> and a @level avoids situations like this, I think.

Ethan Gruber

On Wed, Apr 14, 2010 at 11:58 PM, Mark Carlson <[log in to unmask]> wrote:

This has been my experience as well.  I call it "Encoding For Display" or EFD for short (EAD's wicked cousin).  I've seen all kinds of weird things pass my eyelids just because someone wanted a particular display that the stylesheet hadn't (yet) been programmed to accommodate.  Little did they know that I was their yellow brick road to Oz and all they had to do was to click their heals together three times and say "There's no EFD, There's no EFD".

On Wed, 14 Apr 2010, Joyce Chapman wrote:

This could be different from what is under discussion here, but I've
seen "false levels" at multiple institutions while cleaning up batches
of EADs. As far as I can tell, the main reason component padding has
been done historically is for display purposes. Some processors in the
past may not have understoond CSS or didn't have access to the finding
aid CSS files. While processors may not know what CSS is, they can
clearly see that they control indentation through component levels.
Add a component level and you immediately see increased indentation in
your rendered display. I think it's easy for processors not to
understanding why this is a bad practice if they are new to
technologies like EAD/CSS/HTML or if no one stressed the separation of
encoding and display in their training.


On Wed, Apr 14, 2010 at 6:29 PM, Fox, Michael <[log in to unmask]> wrote:

I just have to comment on an aspect of EAD that may or may not be implied in the original post, one that has nothing to do with the technical solutions suggested.

To be very clear, the statement that

My understanding correct EAD encoding:

        * c1 - Series
        * c2 - Subseries
        * c3 - File


        * c1 - Series
        * c2 - File

is not literally correct.   There is no assumption in EAD that any level of component relates to any level of intellectual arrangement.  A <c01> might be a sub-collection, series, subseries, file or item.  The same principle applies to every component level.  The LEVEL attribute is available to make such designations if one wishes.

I'm not sure I understand what a "false level" might be.   Sometimes I have seen individuals attempt to encode containers as component levels- a real no no.   But of course we do create levels of hierarchy within our arrangements that organize more than describe subordinate units of "real stuff."  As I understand, German archives actually have a term that describes such levels that is rendered by the value "class" for the attribute LEVEL.  It is short for classification or in German "Tektonik".  It is not classification in the sense of Dewey but rather refers to the structural elements of a hierarchy.

Michael Fox

-----Original Message-----
From: Encoded Archival Description List [mailto:[log in to unmask]] On Behalf Of Mark Carlson
Sent: Wednesday, April 14, 2010 1:10 PM
To: [log in to unmask]
Subject: Re: Promoting Container Levels

That's true.  One needs to remember that component levels are nested and
that valid EAD documents require <did> within component levels, so you
can't just remove the false component level and expect it to work.
Assuming that you are using the EAD DTD, this script appears to work
(although I haven't tested it extensively).  Contact me offlist if you
want to pursue trying it.  Mark

<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
<xsl:output method="xml" doctype-public="+//ISBN 1-931666-00-8//DTD
ead.dtd (Encoded Archival Description (EAD) Version 2002)//EN"
doctype-system="ead.dtd" indent="yes"/>
<xsl:template match="* | processing-instruction() | comment()">
<xsl:copy-of select="@*"/>

<!-- Change this to match the <did> of the false level -->
<xsl:template match="c03/did"/>
<!-- Change this to match the component level of the false level -->
<xsl:template match="c03">
<!-- The following should match the component level of the line above -->
<!-- This should match the next component level down from the one being
matched above -->
<xsl:copy-of select="descendant::c04/*"/>

On 4/14/2010 7:37 AM, Nathan Tallman wrote:

As far as I know, there isn't an easy way to do this.  But it never
hurts to ask, right?

Past encoding practices at my institution inserted a false c0 layer in
EAD finding aids.  For example,

    My understanding correct EAD encoding:

        * c1 - Series
        * c2 - Subseries
        * c3 - File


        * c1 - Series
        * c2 - File

    Past practices at my institution:

        * c1 - Series
        * c2 - Subseries
        * c3 - False layer to enclose files
        * c4 - File


        * c1 - Series
        * c2 - False layer
        * c3 - File

Other than hand-coding, is there a way to promote the c4s to c3s and
eliminate the false level?  I haven't used Archivist Toolkit or many
other collection management software packages that might have this


Nathan Tallman
Associate Archivist
American Jewish Archives


Joyce Chapman
NCSU Libraries
Metadata and Cataloging/
Digital Library Initiatives
[log in to unmask]



L. Dale Patterson			[log in to unmask]
Archivist-Records Administrator		973-408-3195
United Methodist Church Archives	fax: 973-408-3909
Madison, NJ 07940