[tei-council] Place followup

Daniel O'Donnell daniel.odonnell at uleth.ca
Mon Mar 26 09:37:33 EST 2007


I'd been too tired to add some additional comments on the place name
test when I submitted my example last night. But here they are now.

I found a couple of minor inconsistencies in the ODD: notBefore and
notAfter failing to be available in places where the examples say they
should be, examples that should text data in elements that the ODD
requires child elements for. I also wonder if we shouldn't have someway
of indicating point-in-time for all elements that have
from/to/notBefore/notAfter elements: like birthdays (where we use
@date), acts of incorporation or census data are valid only for specific
points in time and it felt silly using either from or from/to with the
same date to indicate that Lethbridge changed its name on 1885-10-15 or
had 1850 women in common law relationships in the 2001 census.

The main issues I had, however, were the inability to create structures
within groups of the main elements (i.e. placeName or placeTrait) and,
in the case of the river, to express parts of the place (this last is
not too hard, I suspect though).

So, to give an example, in the case of Lethbridge, various online
sources supply quite a number of Native American names for the city in
various spellings. I went and checked with an elder from the local
Peigan community, and not all of these are equally valid or acceptable,
even though all are out there in sources somewhere. There were at least
three spellings for the most common semi-official name of Lethbridge
among the Blackfoot--sikohkitoks, sik-oki-toks, and sik-ooh-kitoks--of
which the first is the correct current spelling and the others probably
formulated by non-speakers. The other terms all refer to the specific
area in which Lethbridge is, but not all are used to refer to the city.
The one that means "place where we slaughtered the crees" refers to a
major battle that took place here in the very early 1870s just as the
first European settlers started establishing themselves on the townsite.
I don't believe it is current and if it was, I doubt it is polite.

It would be nice to be able first of all to qualify names (something I
guess you can do with type), and secondly somehow group and distinguish
among alternatives, commenting on mistakes if applicable. Also, almost
every source I saw glossed the Native names the first time they came up.
I did this by using foreign within placeName, but I can see that causing
problems if you only gloss "unusual" foreign names: so I could see in
the case of something like Rome using placeName xml:lang="en" for Rome,
and placeName xml:lang="it" for Roma but wanting to gloss whatever the
Blackfoot word for the city is.

Secondly, in the case of Lethbridge I went to Census Canada, where there
were a number of newly released placeTraits. But there is no way of
keeping the internal (tabular) structure Census Canada or even group
them as a subsection of the placeTrait. Leaving aside the ability to
introduce tables--though these are commonly used by geographers--it
would be nice to be able to do something like this at least:

<placeTrait type="population">
  <placeTrait type="singleMales20-24">20</placeTrait>
  <placeTrait type="singleFemales20-24">19</placeTrait>
</placeTrait>

Finally, in the case of the Oldman river, I wanted to be able both to
group locations (some places on the river are both lat-long,
confluences, and towns and it would nice to indicate their relationship
to each other--a similar problem to the above) and discuss the river in
terms of the different geographical areas it went through--mountainous,
foothills, and plains. Here I was getting my information from a study of
the river and they tended to discuss the river in this way.

One option for this would have been to divide the river into multiple
places, one each for each region:

<place>
  <placeName>Oldman river</placeName>
  <place>
    <placeName>Mountainous region</placeName>
    <location/>
  </place>
  <place>
    <placeName>Foothills region</placeName>
    <location/>
  </place>
  <place>
    <placeName>Plains region</placeName>
    <location/>
  </place>
</place>

The trouble with this, though, is that the same source discussed the
climate in two regions--mountains and plains (which affected foothills
and plains). I thought I'd then try to tie placeTraits to location
ranges, but there is no corresp or targets att available.


-- 
Daniel Paul O'Donnell, PhD
Director, Digital Medievalist Project http://www.digitalmedievalist.org/
Associate Professor and Chair, Department of English
University of Lethbridge
Lethbridge AB T1K 3M4
Canada
Vox: +1 403 329-2378
Fax: +1 403 382-7191




More information about the tei-council mailing list