From AJDGOMEZ at telefonica.net  Mon Nov  1 04:31:43 2004
From: AJDGOMEZ at telefonica.net (AJDGOMEZ@telefonica.net)
Date: Mon Nov  1 04:30:51 2004
Subject: [Eac-l] Re: EAC DTD error
Message-ID: <b7d10bb96a.bb96ab7d10@teleline.es>

Barbara, Daniel,
I think I understand both arguments, and, in my opinion, some day EAC 
will have to face this refinement (names/entities). However, from my 
point of view, at the moment the DTD is ok, as a beta version, and 
further refinements before testing it would generate confusion.
Regards
Alejandro Delgado-G?mez
City Council of Cartagena
3000 Informatics

----- Mensaje Original -----
De: Daniel Pitti <dpitti@virginia.edu>
Fecha: Viernes, Octubre 29, 2004 5:24 pm
Asunto: [Eac-l] Re: EAC DTD error

> Barbara,
> 
> I am just about to rush off to a meeting which will occupy me for 
> most of 
> the afternoon. The distinction you make about names vs entity 
> brings back 
> my library authority work. This well may be a point of divergence 
> between 
> library authority control and archival authority control, that is, 
> whether 
> it is about the name or the entity (or perhaps entities) bearing 
> the name 
> (or same names). I'll wait to hear more from the archivists before 
> venturing much further out on the limb.
> 
> Daniel
> 
> At 11:02 AM 10/29/2004, you wrote:
> >I cannot quite follow what the issue is here as my version of the DTD
> >and Tag Library information both show "corpname", etc.   Can you 
> tell me
> >where in the DTD and the Tag Library you are trying to make this
> >correction?  I can't find it.
> >
> >I'm also not clear about what you are asking.  As you know, in
> >authority records or packages of authoritative information about an
> >identity (EAC recognizes identities) the focus is on the names to be
> >used to represent that entity (the identity) - and those in turn 
> can be
> >associated with one or more actual persons (or corporate bodies, or
> >families).  An entity of "person", "corporate body", or "family" can
> >certainly exist, but in the EAC context, the EAC instance is not
> >actually the person, but rather the name(s) associated with that 
> person>within a particular identity - possibly linked to the name 
> of a real
> >individual which could be seen as the surrogate for the "person," 
> but it
> >is not actually the person (and sometimes not even a real person -
> as for
> >fictitious characters, etc.).   So if you are giving names to the
> >entities - won't you need both the persname, corpname, famname as 
> well>as person, corporate body, family?
> >
> >I feel I'm not following the goal of the argument...can you help
> >explain what's issue here?  (I wish we could talk and not email...are
> >you by a phone today?) - bt
> >
> >Dr. Barbara B. Tillett, Ph.D.
> >Chief, Cataloging Policy and Support Office
> >Library of Congress
> >101 Independence Ave., S.E.
> >Washington, D.C. 20540-4134
> >U.S.A.
> >
> >tel.: +1 (202) 707-4714
> >fax: +1 (202) 707-6629
> >email: btil@loc.gov
> >
> > >>> Daniel Pitti <dpitti@virginia.edu> 10/29/2004 10:09:41 AM >>>
> >
> > >Per-Gunnar Ottosson has pointed out to me that the @type values on
> ><eac>
> > >in the beta dtd are different from the values listed in the Tag
> >Library.
> > >After a bit of discussion, we concluded that the Tag Library is
> >correct
> > >and the DTD (and thus the two schemas) are in error.
> > >
> > >We would like to change the values in the DTD from:
> > >
> > >corpname | persname | famname
> > >
> > >to
> > >
> > >corporatebody | person | family
> > >
> > >The logic: an EAC instance describes the corporate body, 
> person, or
> > >family, and not the name, as such.
> > >
> > >Does anyone have any counter arguments to this correction?
> > >
> > >Daniel
> 
> _______________________________________________
> Eac-l mailing list
> Eac-l@mailman.yale.edu
> http://mailman.yale.edu/mailman/listinfo/eac-l
> 

From stephen.yearl at yale.edu  Tue Nov  9 10:36:13 2004
From: stephen.yearl at yale.edu (Stephen Yearl)
Date: Tue Nov  9 11:40:53 2004
Subject: [Eac-l] proposed changes to %a.ea
Message-ID: <5.2.1.1.2.20041109100502.02289750@sjy2.mail.yale.edu>

All:

Currently the EAC DTD has the following defined parameter entity:

<!ENTITY % a.ea
   'ea                   NMTOKEN                                 #IMPLIED'>

Which I would like you to consider rewritten as:

<!ENTITY % a.ea
'ea                   CDATA               #IMPLIED
easys             NMTOKEN       #IMPLIED'>

There are two changes here:

1. ea is now CDATA instead of NMTOKEN.

This would allows the specification of an xpath-- or other-- expressions 
for those descriptive resources that do not have unambiguous identifiers. 
It would also permit the common | delimiter between a MARC field and its 
subfield.

2. a (new) attribute easys (abbreviated from encoding analog system) may 
now appear in all locations that @ea can appear.

Currently the only location where one may specify an encoding analog system 
is using @encodinganalogsystem on <eacheader>. I venture that in a complex 
EAC instance resources and information from many different systems and 
vocabularies can be employed, each one mapping to a different 
encodinganalog. The new attribute locally qualifies and allows for cleaner 
interpretation of @ea's value. In <resourcerels>, for instance, one will 
switch in and out of various encoding analog systems when describing 
resources: MARC data in <bibunit>, EAD data in <archunit>, and so forth.

Here is a fragmentary example to better illustrate the proposed change:

<!--BIBLIOGRAPHIC-->
<resourcerel reltype="origination" easys="MARC21" ea="035|a" type="monograph"
                 system="yul.orbis" syskey="(CStRLIN)CTYG0658040-B">
                 <bibunit>
                     <title ea="245"> Television and antisocial behavior: 
field experiments [by]
                         Stanley Milgram [and] R. Lance Shotland.</title>
                     <imprint>
                         <publisher ea="260|b"> Academic Press.</publisher>
                         <place ea="260|a">New York, NY</place>
                         <date ea="260|c">1973</date>
                     </imprint>
                 </bibunit>
             </resourcerel>
             <!--ARCHIVAL-->
             <resourcerel reltype="origination" easys="EAD2002" 
ea="ead/archdesc/did" type="manuscript collection"
                 system="mssa.locDB" syskey="num=1406&amp;typ=ms">
                 <archunit>
                     <origination>
                         <persname>Stanley Milgram</persname>
                     </origination>
                     <unittitle> Stanley Milgram papers,<unitdate
                         type="inclusive">1927-1993</unitdate> (inclusive). 
</unittitle>
                     <physdesc>
                         <extent>90.25 linear ft. (156 boxes)</extent>
                     </physdesc>
                     <repository>Yale University Library, Manuscripts and 
Archives</repository>
                     <abstract> The papers consist of correspondence, 
research and data files,
                         writings, and course material, which document 
Stanley Milgram's work as an
                         innovative researcher and teacher in the field of 
social psychology. The
                         papers highlight Milgram's work on obedience to 
authority, television
                     violence, urban psychology, and communication patterns 
within society.</abstract>
                 </archunit>
             </resourcerel>

To parallel easys, I'd also suggest that <eacheader'>s 
@encodinganalogsystem also be shortened to this abbreviated form.

regards,

St.