On 2008-06-04 14:36 , "Fox, Michael" <[log in to unmask]> wrote:

In this world of data interchange, wouldn't it be better (safer/more useful) to use the tools within existing DTD/schemas (repeating elements with distinguishing attributes and recursion) to accomplish this without resorting to local variations that will not play well in the larger universe of data?

Yes, indeed. This is what my second example was set illustrate, but the schema needs to be modified to accept xml:lang. This does not seem like an unreasonable modification to the schema to me.

Adding "<xs:attribute ref="xml:lang"/>"  to the definition of the a.common attribute group in the XSD, and importing the XML namespace "<xs:import namespace="" schemaLocation=""/> is all that would be required.

A hack by any other name is still a hack, even if it is dignified by a local namespace.

Fun, though!


Stephen Yearl
Systems Archivist
Yale University Library::Manuscripts and Archives

Michael Fox

-----Original Message-----
From: Encoded Archival Description List [mailto:[log in to unmask] <mailto:[log in to unmask]><mailto:[log in to unmask]> ] On Behalf Of Yearl, Stephen
Sent: Wednesday, June 04, 2008 12:05 PM
To: [log in to unmask]
Subject: Re: Bilingual ead finding aids

It sounds to me like the questioners are looking for a more generic solution. If you are working with the EAD Schema you could declare your own element not in the EAD namespace and use that. Oh, I don't know, something like:

            <my:seg xml:lang="fre"  renderorder="1">My French unittitle</my:seg>
            <my:seg xml:lang="en" renderorder="2">My English unittitle</my:seg>
            <unitdate normal="2008">2008</unitdate>
            <container type="Box">1</container>
            <container type="Folder">1</container>
            <my:seg xml:lang="fr" renderorder="1">My French scopecontent</my:seg>
            <my:seg xml:lang="en" renderorder="2">My scopecontent unittitle</my:seg>

<seg> being an arbitrary segmentation of text and xml:lang being used for a great deal of specificity regarding the language identified (its value should be consistent with <><> ).

It would be pretty straightforward to render this 'master' document into either a French or an English EAD compliant finding aid using XSLT. An additional benefit of using this approach would be that one could selectively apply fonts for the rendering of more 'exotic' languages, if present. Or you could leave the document as is, remembering that there are mechanisms/tools and standards for selectively validating namespaces present in a given XML instance. In this case one would validate only the EAD and ignore everything associated with the "my:" namespace. (This fragment may or may not happen to be valid EAD since from the 'EAD perspective' the unittitle and the p are empty elements, the "my:" is ignored).

There may be other variations and refinements on the above; an additional one might be globally allowing the xml:lang attribute. In that case the scopetent might be:

            <scopecontent xml:lang="fr">
               <p>My French unittitle</p>
            <scopecontent xml:lang="en">
               <p>My english scopecontent</p>

Just a couple of thoughts,


Stephen Yearl
Systems Archivist
Yale University Library::Manuscripts and Archives

On 2008-06-04 12:20 , "Mark Carlson" <[log in to unmask]> wrote:

This could be accomplished using <descgrp> and nesting in EAD.  This would allow you to encode a single finding aid that incorporates multiple language translations.  You could set the TYPE attribute of <descgrp> to specify the language within that group:

=>indentifiers, etc. <eadid><eadheader>

<did> High-level description </did>
<descgrp type="fre">
French language data for <bioghist>, <scopecontent> and other descriptive elements </descgrp> <descgrp type="eng"> English language data for <bioghist>, <scopecontent> and other descriptive elements </descgrp> etc.


Fabienne Queyroux wrote:
> Hello to everyone
> Please allow me to transmit to the list a request from a young
> archivist, Tifenn Hamonic.
> F. Queyroux
> Bibliothèque de l'Institut de France
> "Good morning,
> As a student of the Ecole nationale des chartes (Paris), I'm actually
> carrying out  the Catalogue of the archives of archaeology kept at the
> Institut de France (Paris). In this purpose, the description of
> archives has been encoded in xml using the dtd ead. Then, these files
> were loaded into the software Pleade 3.
> The project has a difficulty that I would like to submit to you. We
> want to make the files available both in English and French.  Each
> description has to refer to the same description in the other
> language. I wonder which tag(s) should be used to make links between
> the French and English versions.
> More generally, I would be very interested to learn about any fully
> bilingual finding aids and the ways the encoding has been done and the
> search interface programmed.
> Can you please enlighten me on this
> subject?
> Thank you in advance for your help.
> Tifenn Hamonic
> Master II, Histoire et nouvelles technologies Ecole nationale des
> chartes [log in to unmask]
> <mailto:[log in to unmask] <mailto:[log in to unmask]><mailto:[log in to unmask]> ><mailto:[log in to unmask]><mailto:[log in to unmask]> "