Re: [idn] (bias) summary of reordering discussion

["Mark Davis" <mark@macchiato.com>]

Count me as against the reordering; I think there has been some interesting
work done on it, but I don't believe it is necessary for IDN.

Mark

—————

Δός μοι ποῦ στῶ, καὶ κινῶ τὴν γῆν — Ἀρχιμήδης
[http://www.macchiato.com]

----- Original Message -----
From: "James Seng/Personal" <jseng@pobox.org.sg>
To: <idn@ops.ietf.org>
Sent: Tuesday, October 23, 2001 09:32
Subject: [idn] (bias) summary of reordering discussion


> On the questions posted:
>
> 1. Efficient - I accepted that reordering produce a shorter ACE string,
> sometimes as much as 20%. This means instead of a "zq--zxcvbnmasdfg"
> label, I get a "zq--zxcvbnmasd". I do not buy the arguments it is
> helpful to naked eye, to memorised or save RAM however.
>
> 2. Compression efficient in future since statistic - Lee's counter that
> compression "always SHORTER labels than usual". Mathematically, it can
> be proven this is wrong (very basic pigeon hole principle).
>
> 3. Referencing from established I18N organisation - ISO14651 is deem
> inappropriate and I agree with it. No alternative was proposed.
>
> 4. Stability of reordering - Lee's countered with the arguments that
> reordering tables would never changed. I am not sure if that is possible
> but I agree with the assessment that it is possible to design reordering
> to be stable. However, I like to see explictive statement in future
> draft.
>
> 5. Future additional of code points / changes to reordering - Lee's
> proposal is a two prefix solution, using prefix as a versioning tag. I
> do not like to solve a problem by creating others, especially one which
> makes it even more complex. Lee have yet to address the process of how
> future additional of code points or changes to reordering could be done.
> (IDN WG is not going to exist forever...I dream of finishing our work
> one day).
>
> 6. Reordering is never ending task - Lee's countered that so is
> Nameprep. My thoughts is two wrong dont make one right. (OTOH, Nameprep
> which is based on UTC work have explicit principles on how it can be
> done. And Nameprep is not subjective to frequency analysis changes which
> reordering is)
>
> Of the people participate in the reordering discussion
>
> Martin Duerst - explicit objection
> Mark Davis - not very supportive, not explicit
> Adam Costello - (no conclusion from comments)
> Doug Ewell - explicit objection
> Paul Hoffman - explicit objection
> Eric Brunner-William - somewhat supportive, not explicit
> Kenny Huang - (no conclusion from comments)
> James Seng - explicit objection
> Karlsson Kent - explicit objection
> Erik Nordmark - not very supportive
> David Hopwood - (no conclusion from comments)
>
> This is my rough read of the discussion on reordering so far (please
> correct me, and apologise in advance, if I am wrong).
>
> (Wearing my co-chair's hat) It is not a vote or even a strawpoll here
> obviously but I am trying to get a feel of the group consensus. But if
> these discussion is any indication of the group consensus, it does not
> indicate very little support for reordering. If there is other comments,
> please bring them forward soon. Thanks.
>
> -James Seng
>
>
>