Sounds like a cool project. But I'm afraid we're missing some crucial
information, here. If the sermon you are encoding is part of a larger
collection, are you encoding the larger collection? Or are you only
encoding specific sermons from it? If you are encoding the entire
collection, is it being encoded as
a) separate, essentially unrelated TEI documents,
b) one TEI document for the collection, each sermon in a separate
c) one <teiCorpus> for the collection, each sermon in a separate
or something else? And if you are only encoding specific sermons
from a collection, how are you handling a case of more than one
sermon from a particular collection?
In any case, it is probably not advisable to mis-use TEI attributes
so completely. The @n attribute, for example, is about a label or
sequence number that refers to the element being encoded, not the
number of units inside some larger structure elsewhere. Much better,
I think, to follow your thought of using <biblScope>. Here is a
portion of a bibliographic citation in <sourceDesc> which (I claim)
asserts that the encoded portion is sermon 7 of 9:
<biblScope unit="sermon" from="1" to="9"/>
<citedRange unit="sermon" from="7" to="7"/>
BTW, I concede the point that it is difficult to derive this kind of
encoding from the _Guidelines_, as there is very little discussion of
<biblScope> and <citedRange>, and not a single example of @from or
@to other than when the unit is pages.
The value of the @ana attribute is defined as one or more pointers
(typically to <interp>s, <note>s, <fs>s, or <category>s) separated by
whitespace. Thus the value "thematic, lectio_continua, unknown" is a
set of 3 relative URLs and means "there are three files named
'thematic,', 'lectio_continua,', and 'unknown' in the same folder
that this file is in; if you read those files, you will find either
a) 3 separate divergent interpretations of this <ab> element, or
b) 3 mutually consistent interpretations of the content of this <ab>
in different contexts."
which I am certain is not what you intended.
Even if you were using "thematic, lectio_continua, unknown" as
shorthand for "either 'thematic' or 'lectio_continua' or 'unknown'",
you still would have the problems that the implication is a file in
the same directory, and that the interpretation applies to the
content of the <ab> itself.
The good news is that the TEI provides a standard and relatively easy
way to classify a given encoded document (or even portion thereof)
with respect to a classification system. But before I babble on about
that mechanism (<classDecl> and <textClass>), it would be useful to
get an answer to my opening question, above, as the TEI system is not
really designed for classifying a document other than the one being
> We are building the corpora of protestant sermons (published in
> Geneva from 1550 to 1750) and we would like to integrate few
> additional information regarding the type of the publications the
> sermon is taken from. Our sermons are either part of a larger
> collection (book) or are published separately as a unique sermon.
> We would like to add that information (collection or unique )
> within the TEI header. If the sermon is a part of larger collection
> we would like to specify: - Number of sermons within the collection
> - Type of the collection (thematic, lectio_continua, unknown…)
> We considered including this information within the <sourceDesc>
> using an element such as : <ab type="composition" n="1,2,3…"
> ana="thematic, lectio_continua, unknown" Where the value @n
> corresponds to the number of sermons : n=”1” for a unique sermon,
> with higher values corresponding to the number of sermons in the
> collection. We are wondering if there exists better adapted manner
> to do so?
> Would it be possible to add this information directly into
> <biblStruct>? For example: <biblStruct type="composition"
> n="1,2,3…" ana="thématique, lectio_continua, unknown"> And
> specifying in the case of the collection the pages as follows:
> <biblScope unit="page" from="12" to="34"/>