Le 24/09/2012 20:16, Jameson Quinn a écrit :
(Resend; last time bounced due to photo attachment)
So, I see that this thread is heating up again, and a progress report
is in order.
[...]
I've looked around a bit at ancient uses, insofar as that's easily
available on the web.
* In petroglyphs from the classic era, it is far more common to find
the numbers incorporated into glyphs (such as date glyphs), with
vertical bars. Also the cloverleaf zero is common in classic
numbers, and often mixes in the same inscription with the shell zero.
* In codices (postclassic), usage is more similar to modern.
Vertical-barred numbers are still present in dates, but the naked
horizontal-barred numbers, square-spaced numbers are common. I'm
not enough of an expert to know how the text flow is working, but
if I'm right that successive digits are alternating colors (red,
black, red, black) then there are examples of both horizontal
flow, vertical flow, and 2-by-2 vertical flow. Also, just as in
modern usage, I haven't seen any cloverleaf zeros.
* I've seen nothing to suggest to me that the "face" forms of the
digits are ever mixed with the dots-and-bars forms. I believe that
for ancient, and certainly for modern, purposes we can ignore the
"face" forms.
About the ancient use : those numerals have also been used in other
mesoamerican scripts beyond the Maya script. as said here :
http://www.ancientscripts.com/ma_ws.html . So I guess the numerals could
be called "mesoamerican" rather than Mayan:
* In the Zapotec script, described here
http://www.ancientscripts.com/zapotec.html . You can clearly see
some numerals (4 and 8) in the stelae 12 and 13 from Monte Albán :
http://commons.wikimedia.org/wiki/File:Monte_Alban_Stela_12_%26_13.jpg
* The Epi-Olmec / Isthmian script, described here
http://en.wikipedia.org/wiki/Epi-Olmec_script and here
http://www.ancientscripts.com/epiolmec.html . You have plenty of
numerals in the few examples I found :
o stella 6 from Cerro de Las Mesas, with Long Count date of
9.1.12.14.10 http://en.wikipedia.org/wiki/Cerro_de_las_Mesas
o La Mojarra Stela 1 described here
http://en.wikipedia.org/wiki/La_Mojarra_Stela_1 . The Long
copunt dates are easily seen here :
http://www.ancientscripts.com/images/lamojarra.gif
o a mask seen here :
http://www.mesoweb.com/reports/isthmian_mask_back.html (you see
a 9 close to the upper right corner)
o The tuxtla statuette
http://collections.mnh.si.edu/search/anth/?irn=8054511 ,
http://en.wikipedia.org/wiki/Tuxtla_Statuette , whwerw we see a
long count date 8.6.3.4.17 and a "vertical 8"
o The stella C. at tres Zapotes, with a long count date.
http://en.wikipedia.org/wiki/Tres_Zapotes#Stela_C
o A LaTeX package and a ad-hoc 8-bit fonts exists for Epi-Olmec.
It is available here
http://obelix.ee.duth.gr/~apostolo/Epi-Olmec/ and described here.
* The Ñuiñe script, briefly mentioned in
http://www.ancientscripts.com/ma_ws.html, where the examples given
clearly show some numerals (10 and 11).
As far as I understand, there is little agreement over the decipherment
of any of these scripts, and there is probably a long time to wait
before any of it is encoded in Unicode, but given the already encoded
and planned undeciphered scripts (Phaistos disc and Linear A at least),
I wouldn't be surprised to see one of these script being encoded before
the well understood but complex Maya script.
Your initial idea was already to make them multi-script (Maya + Latin),
so adding a few script to your proposal shouldn't be a problem. It might
even ease the situation, since we seem to be in a case similar to the
Aegean Numbers (U+10100–U+1013F ), used in linear A and B. I'm not a
specialist (at all !), but I would be surprised to find the complex
behaviour of the numerals in the Maya script (which seem to scare
Michael Everson) reproduced identically in all the above scripts. In
particular, it seem absent from the Epi-Olmec/Isthmian script.
* I'm enclosing a picture of the bills
<http://picpaste.com/billetes.jpg> I happen to have around right
now. There's one of every bill in current use (1,5,10,20,100,200)
except the 50. As you can see (assuming this comes through on the
list; otherwise, just google "billetes quetzal"), all 3 of the
bills with a 0 use noticeably different variations of the "shell
zero" glyph. The 20 and 100 have two stripes in their zeros, the
200 has four of them in its. (The 10 and 50, of course, have no
zero in base 20)
Good pictures of the bills (including the Q50 one ) can be found her
http://en.wikipedia.org/wiki/Guatemalan_quetzal#Banknotes .
By the way, I recently saw a post from an associate professor of
matematics looking at ancient number systems in (Xe)TeX. He says "I’d
love to be able to do something similar with the Mayan numbers. I tried
for a while, but couldn’t get them to work." The reason of his failure
is the lack of unicode encoding.
http://divisbyzero.com/2012/08/30/ancient-number-systems-in-xetex/
Do people think I should include any of this investigation of ancient
usage in my proposal?
As you've probably guessed by now, I think you should... but I have no
experience at all in the encoding process!
How many examples of modern usage, aside from currency, are worth
including in a proposal, assuming that they will take me an hour or
two each to find distinct uses, or a day or two to go visit the
publishers and get unlimited examples?
I think it's more the diversity of usages than the number which counts.
Out of pure curiosity, I have the following question :
Do modern Mesoamericans still speak languages using a vigesimal system ?
If yes, how do they write numbers like 127 (especially in informal use
?). 127 ? 6,7 ? In letters ? With Mesoamerican numerals ? As discussed
here (see
e.g.http://www.unicode.org/mail-arch/unicode-ml/y2012-m04/0102.html and
answers ), some Inuits have felt the need to introduce a base 20 numeral
system, but I've no idea of their success or failure. If you find some
non-decorative modern everyday use of Mesoamerican numerals, even
handwritten, I think it will be difficult to object the encoding.
