more messages (2) FROM CHINOOK-L

[Mia@RedPony]

I hate to say this, but this is not trivial.

For example, I have been working on fonts for Athabascan for over a year
now. Apache uses rising tone nasals. These are not supported directly as
glyphs in Unicode. This means a) the support has to be developed to build
the glyphs dynamically; b) the code has to be embedded in the html
documents; c) the code will not work for some browsers and oh-by-the-way
they ALL have to be tested.

Alternatively, a font has to be developed that supports the glyphs directly.
You may not, by law, take existing fonts, modify them, and redistribute them
because they are the property of the big font companies and are covered
under intellectual property law. So you must either get the font companies
to modify their fonts to the specifications of the Native language, or you
must develop your own, from scratch. (Definitely somewhat non-trivial; the
existing fonts have been developed over 400 years.) Then, whichever of these
two directions you took, you must find a way of making sure that the new
font is available on the viewer's workstation. You can embed them in a .pdf,
but then the text cannot be used for other purposes; this is a bad thing in
the Indian communities where there is little enough material without it
being hung up in capitalistic "protection features".

Finally, most of the browers are written in C++; C++ uses base classes; the
base classes are dependent, ultimately, on two primitive standards, ISO-639,
which defines the language codes, and ISO-3166, which defines the country
codes. Using any of the standardized routines for Internationalization
(I18N) or Localization (I10N) requires having referencable values in the
base class that define these two standards. There is a pretty good overview
of the process here, http://www.linuxjournal.com/article.php?sid=6176, for
Linux, which is Open Source.

To modify a browser that has the necessary base class support for a
particular language is not overly difficult. The programmer power required
depends on whether one wishes to implement what is called the "interface" in
the Native language, or simply modify it to support Native languages.

If one wishes to modify the interface, one should go all the way and make it
culturally appropriate. Remembering that what is culturally appropriate for
one tribe is culturally offensive for another, the interface needs to be
configurable with graphics, colors, and text support. Then, the words and
messages, help text, menu displays, all need to be "converted" into the
target Native languages. HOW one does this is a major language extension
questions.

My friend who wrote the paper in Navajo for his grandmother said he had to
take words for concepts that seemed related in Navajo. When I did my Hopi
presentation on DNA, I crafted words, much as photomicrograph, typewriter,
computer and alternator were crafted. The Elders were in general pissed with
both of us. Sigh.

So how you get this massive amount of text "translated" is a major issue.
What do you do when there are no words in the language for the icons. For
example, on my tool bar for Outlook, which I am using to write this, are
"Undo", "Paste", "Check", "Spelling", "Attach", "Priority", "Sign",
"Encrypt" and "Offline". Of these, "check" and "attach", possibly "sign",
can be expected to exist in most Native languages. But I'll bet only some of
the really modern ones like maybe Cherokee have words like "spelling" and
"priority". And I'll bet that none have "encrypt" and "offline". To "borrow"
the words from English kind of defeats the purpose here, and encourages the
development of a complex Nanglish. . . . probably not pretty.

You also want to use long descriptive words, like the Hopi word for
computers, which essentially describes what it does and consequently is not
a good label for an icon.

And it goes on.  . . to do this, one needs a Group (large group,
preferrably), of committed people, with a space, and enough money to support
the task. Once you have to have a job and be in school full time, you don't
have much left over for volunteering.

Our task would be made much easier if we could get someone to change at
least the ISO language standard to include Native American languages, and
also to convince the font companies to support us with distributable fonts
that included the necessary special characters, instead of expecting that
our Native people and miscellaneous scholars will spend from $99-$129 for
One Single Font. When we install a new operating system, or package like
Photoshop, we get 1000's of fonts for English that also support all the
Romance languages; some even support Germanic and Cyrillic, and there are
the occassional full-ISO 10646 fonts, like Lucida sans Unicode, and the
Microsoft 80-bazillion MB one that you can download.

I'm actually planning on building a "font engine" this summer that will be
distributed for free to Native communities and for a nominal cost to
academic researchers, and will take a font and create the necessary
characters for the target language using Unicode methodologies. The reason I
am distributing it this way is because if you buy a font from the font
companies, or get one in a distribution, and You Modify It Yourself, then
you can use it as much as you want. You can't sell it or send it to someone
outside your organization, however. With the Red Pony Font Engine, you will
be able to make your own fonts, send the engine and the specs to your
friends, and they will be able to build Their Own fonts that will look
exactly the same as yours. This way, people will have fonts for posters,
calendars, cards of all kinds, graphics, whatever they can dream up.

. . . .this sure ran on. . . .
--> Just a Few Thoughts!

Mia


----- Original Message -----
From: "coyotez" <coyotez at OREGON.UOREGON.EDU>
To: <ILAT at listserv.arizona.edu>
Sent: Saturday, May 10, 2003 12:40 PM
Subject: more messages (2) FROM CHINOOK-L


> Date:         Thu, 19 Nov 1998 16:16:57 +0900
> Reply-To:     The Chinook Studies List <CHINOOK at LINGUIST.LDC.UPENN.EDU>
> Sender:       The Chinook Studies List <CHINOOK at LINGUIST.LDC.UPENN.EDU>
> From:         Mike Cleven <ironmtn at BIGFOOT.COM>
> Subject:      Re: Chinook browser development (was  Fwd: Hawaiian language
web
>               browser released (fwd)
> Content-Type: text/plain; charset="iso-8859-1"
>
> Now that Netscape source code is "open", anyone can develop add-ons to it,
> including other language versions.  To get this done, it would have to
mean
> that one of us is familiar with writing code - or we can convince someone
to
> develop it for us.  Any takers?
>
> I kept notes on this once upon a time - y'know......File-Print,
> Edit-Preferences, etc.
>
> First suggestions "File" = "Ikta", "Edit" = "Mamook", "Save" - "Iskum",
> "Undo" - "Kilapi", etc.  Some of the compounds needed are going to be
> interesting.
>
> Just for fun we could all develop this as a communal whiteboard page;
> arguing out the possible terms.......
>
> Also, there's Opera (www.operabrowser.com I think - maybe
> www.operasoftware.com).
>
> And as a "by the way", I noticed www.slanguage.com listed somewhere -
maybe
> the Herald Tribune.  Maybe Jeff's page is already listed there.....
>
> Mike Cleven (hiyu siah kopa huloima illahee)
>
> Netscape's Universal Localization Program: Netscape ULP homepage:
> <http://www.mozilla.org/docs/l10n>
>
> Just thought I'd let y'all know I've been having a look at the Netscape
ULP
> - purdy interestin', if I do say so myself.  I'm willing to slog through
> the "how to" of the damn thing if the rest of you can help me find
suitable
> translations/renderings of the drawbar commands and other interfaces.  Not
> that a Jargon browser is going to have broad appeal; it'll just be
"cool" -
> apparently we also can't call it Netscape, according to the rules of the
> project, so we also need a suggestion for a name.  "Wawabox" or
"Wawahouse"
> occurred to me (as replacements for "Communicator").
>
> But as I was reading, it occurred to me that the Netscape ULP could have
> profound use within the First Nations/Native American linguistic
> communities.  I don't know if anyone out there is working on this for
> Cherokee or Navajo or any other major American native language, but I
> suspect someone is already working on Inuktitut/Inuvaluit and Cree
> variations on the theme.
>
> So it's occurred to me to suggest it to you Salishan list people to try
and
> evolve a "standard" Salishan vocabulary for browser use; I know that there
> are wide disparities in languages within the language family, but maybe
> this is an opportunity to derive "New Salish" technological terms and
> ideoms......cooperation between linguists and actual tribal community
> members would seem to be a must.  The alternative would be to develop
> separate Secwepemc, Nlaka'pamux, Lushootseed, etc. versions.....
>
> I'm going to turn a couple of Kwakwala and Tsimshian people I know on to
> the idea; at least in their cases it's a little more
straightforward.......
>
> David Lewis
> Confederated Tribes of Grand Ronde
> Department Of Anthropology
> University of Oregon
>
>