On 15/08/10 16:21, Natalia Ermolaev wrote:
> I am working on project digitizing Russian journals from the
> 1920-30s. My first step is to encode the tables of contents (rather
> than generating a TOC based on the articles – I will encode the
> articles themselves at a later point, contingent on continued funding
> for the project).
> journals. I’ve been experimenting with listBibl, but am not sure it
> is the best element for the kind of TOC I’m aiming for. Here’s an
> example of what I’ve done so far:
It all depends on your needs and approaches as to what a ToC is... But
listBibl seems a reasonable way to do it to me. If you find you need
something slightly more flexible, you could just use 'list' with 'item's
which then contain 'bibl' elements.
> However, I’m not sure where to put the information about year and
> issue number. Any tips on how to use listBibl, or perhaps a more
> suitable element for this project would be much appreciated!
I think I'd have a <listBibl> for each issue, with a <bibl> for each
article. It would then be fairly straightforward to generate a
teiHeader for each article when you do get around to transcribing them.
I think I understand your problem... since you're encoding all of the
entries you feel it is redundant to include a <date> and perhaps <idno>
for each of the <bibl> entries. Since I'd be tempted to reduce the
burden on processing these, I would be tempted to just to repeat them
inside each <bibl> element. However, if you are processing this with
XSLT it would be easy enough to put them just in the first <bibl> and
then in processing go and get them from this location. I know some
might argue it is against the spirit of XML and the re-use of data...
but I would just put them in each <bibl> because then no special
processing is necessary to make them complete. (And if someone else
picks up this file at some point, they will understand it.)