Re: [idn] opting out of SC/TC equivalence

["James Seng/Personal" <James@Seng.cc>]

Hi Xiangdong,

Now, responsing directly to your note:

First, I hope we can agree that the wg is well aware of the TC/SC
problem (after such a long discussion thread) and we are all actively
seeking a solution to it. We may disagree with the approach but we are
certainly not disagreeing with the principle.

> you said it. Maybe we could not handle language problems in IDN, but
we could
> not forget that all script have language characteristics of
themselves, especially for
> Chinese. Internationalization means all users could use their familiar
ways to use
> domain name, not means all users must follow one ways.

Lets use logic to discuss this: A. domain names B. scripts C. languages

Your reasoning:
domain names handle script. A->B
script are used by languages. B->C
Therefore, domain names handle language. A->C

basic logic tells us that the last statement is not always true. And in
this case, domain names is not able to handle language. This is a known
limitation of domain names.

> Exactly, we are doing internationalization, but how do we explain for
> users what is internationalization, and what is multilingualism, I
> think ordinary users care about how to use , not what is.

Agree. Very few people recongized the differences.

While there is an desire to solve 100% of the users problem today, all
developments started with one small step at a time. Without solving the
first 10%, we wont know what the next 10% problem. This wg is not the
first nor will be the last one to solve the whole I18N problem. It is
one of the puzzle piece for the IETF and there are many other puzzle to
solve in future. So lets not put too much task upon ourselves.

> > Thus, I would modify your statement below to: "from each SCRIPT
> > users/engineers and ask them how they want their scripts to be use".
> > (ie. all han ideograph (CJK) users should talk about han folding
> > together if we going down the path of tsconv).
> >
>
> Prof. Tseng care more about users requirements, I think, a technology
developed without
> enough users requirements, it makes no sense.

And so am I. This is why I can understand Prof Tseng's well-intention.

But the "users set" in IETF context are not limited to any particular
locale so the opinion to the solution may be slightly different due to
different perspective.

Anyhow, lets not argue further on these details anymore but lets focus
on doing the job. The tsconv draft requires more work so lets put more
effort into that to make sure it is completed.

*cheers*

-James Seng