[Edu-sig] More about Internationalization (i18n) / Unicode

[kirby urner]

I've been fostering debate around work about i18n ("internationalization")
a tried and true topic in Geek World bar none, and otherwise known,
paradoxically, as "localization" as in "make all these Ubuntu menus appear
in the Basque language."

I remember being in Lithuania and booting up Blogger, to find my control
panel in Lithuanian.  Cool!

We're brainstorming a promo / commercial for OST (oreillyschool.com)
harvesting footage at OSCON, where I'm suggesting a mostly non-English
sound track with students bragging about the jobs they got or the skills
they learned from our courses + mentors, in various non-English languages.

"Mostly non-English" is likely way too radical to make the final cut as a
story-board for this promo (maybe another one), but I do hope we'll have at
least some soundtrack in fluent Russian, with sub-titles added later.

Why?

Because the two alumni / active students I've recruited for the promo, are
fluent Russian speakers, so duh, right?

But zoom out and here's the bigger debate:

should we think of English as the language of instruction or more just an
important language for documentation?  It's an important decision.

In the latter case, it would be important to have our Python track
materials in fluent Portuguese, including of course the source code, ditto
in China we would use mostly Chinese.

However, the docstring APIs in the Standard Library would stay as they
are.  The Standard Library is to remain in English.

So I'm all for keeping it straight Python 3.4.1 or whatever, i.e. I don't
want to have truck with these  "replace the keywords" versions, because we
want to standardize on Python.org distros, known to be best.

Your viability in the job market depends on not being too specialized.

In the former case (English as the language of instruction), we go for
"dumbed down" or "for dummies" English ala Business English (BE) as the one
and only language of instruction and make that a per-requisite for our
computer courses.

Mentors such as myself need only know BE and we'll be able to do our jobs.

That's true now in fact, but not because of any decision or ideology,
simply because of geography and history.  The school grew up in the corn
fields of Illinois.

Only now are we experiencing the ripple effects of globalization / Unicode
/ cosmopolitan best practices.

We could very well hire a native Chinese-speaking counter-part for my
Python teaching job tomorrow, but then we'd have to launch a new version of
the text as well, by either the current author (Steve Holden) in
translation, or by someone equally well versed in Python.

Usually translations lose something don't you think?

Maybe a friend of Steve's who writes in her or his own voice would be the
ideal candidate.

Anyway, it's not like we have to settle this debate right away.

I think many colleges and universities are looking at the same issue.  I
invite others to jump in with their musings.  What's you long range
forecast of how things will go?

Case in point:  Portland State works with a sister campus abroad (Fudan
University) where many of the courses are in Chinese, naturally, and yet
videos of US English-speaking professors -- e.g. my friend Allen Taylor,
author of 'SQL for Dummies' and an electronics engineer -- are included in
for-credit English language courses as well, so it's a mix.

Allen is a very skilled teacher (I've been to many of his talks) and
lectures on cruise ships these days.  Having him on video in China is not a
waste of anyone's time, if they have the skills to follow, and many Chinese
do.[1]

My point:  a heterogeneous environment is what I expect, with many students
pursuing their areas of concentration using two or three human languages,
actually the standard for an educated person going way back.  I don't
expect a mono-lingual future and don't see why anyone would.

Obviously, Unicode was / is a huge undertaking and we're only beginning to
reap the rewards of this investment.

One of those rewards is the ability to break free of the limitations of
ASCII's 127 and later Latin-1's 256+.

We're ready for the millions of characters that comprise our heritage.

I, for one, am looking forward to the diversity that brings to our Python
community.  Carl Trachte and I have been focusing on i18n within the Python
community for some time now.[2]


Kirby

[1]  http://worldgame.blogspot.com/2005/01/shanghai-city.html
[2]  http://pyright.blogspot.com/search?q=unicode
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/edu-sig/attachments/20140629/2ec3c739/attachment.html>