CSD (privately)

[Koontz John E]

On Tue, 7 Oct 2003, R. Rankin wrote:
> Wally Hooper has a conversion program that works within IDD.  The
> particular ASCII/ANSI values can be plugged in.

I think this approach would work for the CSD.  Because it used a single
set of characters plus some supplementary digraphs it is easier to handle
than files which use different sets in different fields and subfields.

> I looked at this problem last June, and the "other dialects" or "other
> languages" field could be used.  It's set up so you can specify any
> number of other languages.  it would make very heavy use of that one
> field, but it would work.  And there are other fields for grammatical
> class, gloss, etc.

Well, I still wonder, because one of the functions of the fieldname in
cases like this is to key what formatting is needed.  If the other
language field is used only for consistantly formatted data fields and the
language in question is coded as a new initial subfield, then this might
work.  It might be useful if they could add a field or set of fields for
comparative citations.  The information we need in that context has
certain subfields:

1) Language name.
2) Phonemic form, optionally tagged to indicate that it is an internal
reconstruction itself.
3) Source form, as transcribed in the original source.
4) Source gloss.
5) Source code, usually some mnemonic plus a page or item number.

There has to be some way to attach comments or remarks to this at this
level, which might be a 6th subfield or a paired field.

This can be handled as multiple fields, but in that case it's desirable
that editing copy consolidate the fields on one line or successive lines.

It can be useful in the phonemic form and gloss to be able to bracket the
part compared, but whatever bracketing there is should be invisible to
searches, e.g., nu<bdhaN> should be locatable as nubdhaN.

The other sorts of fields we used were Proto-X, and Pre-X, where X is some
language, a rather limited set of these, and Other Reconstructions, for
citations of reconstructions by others.  Those could be handled as a
variant of the general citation field.  We had a general comment field.
In comments in multiperson work it's useful to be able to sign and date
your comment.

I've often thought that we needed something like CVS - a tool for keeping
track of changes.