[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [idn] VIDN
- To: idn@ops.ietf.org
- Subject: RE: [idn] VIDN
- From: Jonathan Rosenne <rosenne@qsm.co.il>
- Date: Tue, 12 Dec 2000 22:44:01 +0200
- Delivery-date: Tue, 12 Dec 2000 12:53:00 -0800
- Envelope-to: idn-data@psg.com
Some languages, such as Hebrew and Arabic, commonly do not use vowels, making
automated transliteration impractical.
Jony
> -----Original Message-----
> From: owner-idn@ops.ietf.org [mailto:owner-idn@ops.ietf.org]On Behalf
> Of James Seng/Personal
> Sent: Tuesday, December 12, 2000 10:21 PM
> To: idn@ops.ietf.org
> Subject: [idn] VIDN
>
>
> I promised to send my comments to the mailing list on VIDN.
>
> I believe I am not only speaking for myself when I make these comments so if
> other people in the WG (other than the authors of VIDN disagree), please feel
> free to correct me.
>
> a) VIDN assumed that all non-English domain names are transliterated into
> English for current domain names which is seriously questionable. Sure,
> it works for Korean, but what about other languages?
>
> b) The NICs are responsible for operation of the ccTLD. Assumption that
> the NICs have the skill and capability to do language transliteration
> standardisation is once again beyond comprehension. (Some of the ccTLD
> may have such skill but they are not common.)
>
> c) Certain languages are used across different countries who may decide
> different transliteration rules. Worst, VIDN assumed that it is possible
> to have unified transliteration rules for each language.
>
> d) Certain languages which share the script which is used across different
> countries.
>
> e) Certain names may have contain different set of script which makes such
> transliteration system probably difficult.
>
> Going finer to the detail:
>
> f) The round-trip conversation is preserved by associating the names with
> an number ID. The I-D did not mention how this ID is assign, transmit
> and use.
>
> g) If VIDN names requires ID to resolve ambiguity, then look in another
> perspective, this ID could be the Unicode Codepoint of the IDN. And
> if ACE/UTF-8 is also a representation of Unicode Codepoint, it can
> be concluded that this ID is no different from ACE except it is assigned
> as a number. (This is the point raised out by Maynard Kang)
>
> On the other hand, there are certain good point on VIDN (Okay, I am playing
> devil adovcates again :-)
>
> 1. VIDN allows existing "some" registered domain names to be used
> without another registration.
>
> (I am saying "some" here because the author has claims 90% of korean
> names is working under this system but i dont think it solves the
> other languages)
>
> 2. Transliteration is definately not trival and hats off to the
> author who manage to do this for 90% of korean. It would be interesting
> to see how it would scale.
>
> 3. VIDN is in some aspect an extension of IDNA except of course a
> translitered english names is used instead of an ACE.
>
> ps: I dont see why been aggressive and threaten me after the meeting
> would change my opinion. It is probably better to argue with me
> based on reasoning.
>
> -James Seng
>
>