Hi all,
I have several small, maybe ordinary, questions about encoding
information on a person in a <listPerson> (or on an organization in a
<listOrg>).
1. encoding biographical information as a discourse
Several projects in which I am involved include the making of authority
files (persons lists), where each record on a person contains, first and
always, one or more names (mandatory of course), optionnally birth
and/death dates, one or more "occupations", etc. (i.e. information that
can be structured using the existing TEI namesdates module elements),
*and* contains a text giving information about some of the person's
activities and history, in order to contextualize the edited text and
explain why the person is mentioned in it, which role he/she played,
what he/she did, etc. In other words, a text about some features of the
person's biography, several phrases in one or more paragraphs, like in a
biographical dictionary.
The <event>, <trait> or <state> elements do not seem appropriate, since
the discourse thus should have to be rewritten and split into sentences
or terms (one sentence per event, etc.).
The <p> element cannot be used within <person> when other more specific
elements are already used, even if the only other element used is
<persName>.
So I am thinking of :
- either using only <p>, and <persName> within it, but it seems that
there would be a great loss here, as nothing would thus explicitly tell
that the name contained in <persName> is the described person's name
(and other problems would happen, such as "how encode several parallel
persNames"). So I do not like this option
- or creating a new container element, such as
<biographicalInformation>, and using <p> within it.
This option, which has in a way been chosen in other specific standards,
would have my preference.
I wonder if someone here has had and solved the same problem ? The same
situation could of course happen for any organization (an <org> in
<listOrg>).
2. encoding bibliographical references as source of information in a
<person> or <org> element
Of course, one can use <bibl> (in <person>) or <bibl>, <biblFull> and
<biblStruct> in <org> (so that there is a difference between the content
models of <person> and <org>).
My point is that the result will not explicitly tell that the book or
resource described in <bibl> is the source for the information given in
the <person> element, or gives more information about the described
person. So, what do you think of this ? could it be a reason to
introduce some <bibliography> or <sourceDescription> element or
something like that ?
3. linking a <person> or <org> element to an external authority record
that describes the same entity (the same person or the same organization)
In order to do that, one can think of using attributes:
- @corresp. But here the correspondence is very close, since the
<person> element and the external authority file describe the same thing
; @corresp does not seem to express this correctly. Another problem is
that @corresp is often used, in our projects, to link the element to one
or more <persName> or <orgName> within the transcription of the indexed
text. We could remove these links of course, since the symmetrical links
also exist...
- @sameAs. But it seems that this would not be appropriate, since the
Guidelines say that this attribute may be used only when the two
elements linked have identical *content*. (see
<file:///C:/Normes-et-technos/TEI/P5/2-1-0/doc/tei-p5-doc/en/html/SA.html#SAIE>).
Of course I could create a new attribute, or use @sameAs in a more
extensive way than the Guidelines say.
Has anyone thought about this ?
The question may concern more TEI elements (a resource in <bibl>, a
<place> etc.). IMHO, could be considered important at a moment where we
try to Link our Data to other data ;-)
Thanks a lot for any answer
best,
--
Florence Clavaud
École nationale des Chartes
Paris (France)
|