Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!news.ultranet.com!news.sprintlink.net!cs.utexas.edu!swrinde!sdd.hp.com!hplabs!hplextra!news.dtc.hp.com!hpscit.sc.hp.com!hpax!brunner
From: brunner@cup.hp.com (Eric Brunner Contra)
Subject: Re: Sources on language issues on Internet
Sender: news@hpax (News Admin)
Message-ID: <D69xFA.8p3@cup.hp.com>
Date: Thu, 30 Mar 1995 22:06:45 GMT
References: <hectorh.3.000F5196@umich.edu>
Nntp-Posting-Host: hpwsky.cup.hp.com
Organization: Hewlett-Packard
Lines: 63

hectorh@umich.edu (Hector Hernandez) writes:
...
: I am looking the impact (cultural, social, etc.) of English becoming "the 
: language of the net."  I wonder what kind of effects this will bring as the 
: Internet becomes more ubiquitous. I am particularly interested in the short 
: and medium term effects in developing countries and if there are any policy 
: issues that can be considered.

Hmm. I do i18n work, formerly at Sunsoft, now at HP. Some comments. We (the
older header-people/mailer-weenies/interface-junkies) have pretty much gotten
the email world we live in (this excludes ubiquitous odd-ball mailers out in
hyper-space: Mac mailers, DOS mailers, integrated office suites containing
mailers ... the kind of stuff that isn't part of our rfc821/rfc822 world) 8-bit
clean. Last time anyone checked only a hand-full of Bnews sites were still
even in the suspect, let alone the known-to-step-on-the-8th-bit-dirty list,
out of the ~100,000 uucp/tcp transported smtp relay sites constituting some
of our idea of "the net".

A significant amount of traffic exists, and support for the local language
within the context of newsreaders, now html browsers for scripting systems
that require more than 8 bits -- e.g. asian languages. Whether you run into
a UNICODE-solves-everything advocate, or an I've-got-my-SJIS/EUC/ROMAN8/HP15
advocate (for any value of everything to the left and right of EUC), or some
one like myself working on ISO10646 (utf8), vendors have fielded partially
internationalized, partially localized systems for years.

Where there is an implicit english is in our rfcs describing, or not describing
our common network tools -- the talk rfc doesn't require talk providors to
support code sets, hence languages, other than ASCII, hence English.

But ignoring the trendiness of the net, and the cost issues around native
language support (I'll know that I'm sucessful if my architecture makes it
cost effective for marketing to sell my refrigerators to the esquimoux --
that's a joke, but I really am working on Inuktitut and Cree syllabaries
as "proof" an architecture works economically, not just in Japan/China, but
in miniscule niche markets, of which there are hundreds), why not look at
other, necessarily earlier events where a language becomes part of an exchange
medium? Chinook Jargon comes to mind.

As to national policy makers, you should know that these frequently mandate
standards -- e.g., I'm doing 4 byte EUC because it is a Taiwan (ROC) hard
requirement. At Sun the provincal government of Qubec had requirements for
a keyboard that came to my personal awareness. The list goes on.

Pick up a copy of the XPG4 i18n docs and get an understanding of what the
XPG vendors are providing i18n and l10n programmers. Check out the active
newsgroups in asian encodings, buy the cjk packages for your Mac or PC or
your workstation, browse the culture newsgroups for language use. Buy a
copy of Ken Lunde's book, or any of the i18n/l10n books -- there are at my
count, a half-dozen good ones.

Finally, check if you've already got a conclusion before your research
really picks up. It looks as if you might, and if you are a US english
mother-tongue person, keep in mind that it just may _seem_ they way you
see it. The less subjective proof would be to as for someone in the IETF,
but not me, I'm busy, to start keeping stats on inet servers which have
inferable code-sets, hence writing systems, associable to them. We keep
per-protocol data, we could keep per-encoding data as well.

-- 
Kitakitamatsinopowaw (I'll see you again)

		-- Eric Brunner
