Keith Moore moore(_at_)cs(_dot_)utk(_dot_)edu said:
People can and will use their own languages on the Internet - in email,
on the web, and in domain names, and without regard to their location
in either the physical world, the currently topology of the network,
or the TLD of the host they are using at the moment. Furthermore, a
great many people use multiple languages (not necessarily including
English) is, so that a given person, host, or subnetwork will often
need to exist in multiple (potentially competing) locales at once.
Fortunately the IDN group is making very good progress, and I'm
confident that consensus around a concrete proposal will soon emerge.
Dan K dank(_at_)hq(_dot_)lindsayelec(_dot_)com says:
Well, People cope with the flaws reasonably well. The codeset loaded into
this email client and OS has a hefty smash of diacritial support. Most
languages with a western origin can be represented with some moderate
difficulties.
A Scientific American article on machine representation showed how uneven
the support is, showing some languages really take a beating from word
processing in general. The negative example was Farsi, which they
illustrated looks tragically bad when machine rendered without specific
technology support.
The 16 bit attempts for some ideographic languages seems substaintially usable.
One reason the IDN thing is so daunting is the work arounds are not that
bad. For instance, you can embed a backgroundless GIF into a web page and
have any ideogram link to a URL. That's nearly ideal in many ways. Storing
it as a bookmark, "favorite" whatever, the underlying machine language is
barely encountered. If the local Software browser stored the graphic neatly
and presented it well, the author would have total freedom to compose an
image of any sorts and have it persist indefinitely. Disorderly but functional.
That's why I think the work should continue and broaden, and somehow, I
don't know how, get more non-technocrats to try this stuff out. Not rush
into global piecemeal application. As of course discussed at length
previously, those are reasons to get the protocols perfected in the absence
of knowing how to apply them. Subtle work.
I'd liek to do more of substance other than theorize. I think I will study
the concepts behine unicode this weekend and try to develop a better
understanding of that work.
Regards to all,
Dan Kolis
Dan Kolis - Lindsay Electronics Ltd dank(_at_)hq(_dot_)lindsayelec(_dot_)com
50 Mary Street West, Lindsay Ontario Canada K9V 2S7
(705) 324-2196 Phone (705) 324-5474 Fax
(888) 326-5654 Pager Anywhere (888) DANKOLIS {Same #)
An ISO 9001 Company; SCTE Member ISM-127194
/Document end