Apps-discuss teach internationalization needs to their members.
http://www.saint-andre.com/ietf/i18n-intro.pdf
Then PRECIS (stringprep) has also its set of slides:
http://www.saint-andre.com/ietf/ietf81-precis-framework.pdf
These two problems are intricated because they share the same
problem: the sustainable use of Unicode.
At 23:19 20/07/2011, Florian Zeitz wrote:
For implementers caring about code/executable size that's probably
about the same.
From a specification/standards point of view I see the merit/your point.
The problem is that Unicode is still an inadequate asolution to
support computer network use by real people. And that majuscules are
still not supported what makes the whole considered things
"senseless" (I mean semantically lame) for Latin languages, in
particular French. I did not see the main point quoted in these
slides, which is orthotypography.
As long as one considers that the Internet is 7 bits ASCII I am
afraid we also do not perceive the problem correctly, what means that
we will most probably be unable to explore the right solution,
whatever it may be. I know getting rid in part of Unicode and
changing our reading of the Internet technology is something hardly
conceivable, hence inacceptable. However, this is what we just did
with IDNA2008.
Because everyone wanted to be free from Unicode versioning and users
could not accept Internet internal mapping. We now need
multilinguisation, i.e. the coexistence of every language on an equal
footing and a physhing-proof universal character system. The best
stringprep solution is no stringprep. This is RFC 3439 principle of
simplicity. I think Unicode is not RFC 3439 conformant, hence all our
problems. Sure we need an IETF temporary patch, but we also need a
sustainable long term solution. If we do not consider the later, IMHO
we will not build a good enough patch.
jfc
_______________________________________________
precis mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/precis
