New version, draft-faltstrom-idnabis-tables-02.txt, available

Martin Duerst duerst at it.aoyama.ac.jp
Tue Jun 19 04:45:50 CEST 2007


At 00:05 07/06/19, Patrik F$BgM(Btstr$B‹N(B wrote:
>
>On 18 jun 2007, at 09.45, Martin Duerst wrote:
>
>> For the purposes of this draft/memo/specification, a script is
>> considered unproblematic iff:
>> 1) ...
>> 2) ...
>> 3) ...
>> ...
>
>I understand more text should go into the tables-document, but it is  
>for me very very important to also get the core of the document  
>ready. We on this list should, I think, understand many of the  
>underlying arguments by now, so I hope you can handle the case not  
>all words and references are there yet. Making major updates AFTER  
>adding all "bling" to the document is quite tiresome.

I didn't ask about the document. I'd be fine with a list on the
mailing list. But it's just plain impossible to say "script
X is unproblematic" when there is no clear definition of unproblematic.

>First of all, the normative spec in the document will be the  
>algorithm. We have, as described in the idnabis document, seen all  
>problems with the existing IDNA that because of the codepoint spec  
>lock IDNA with Unicode 3.2. Only by moving to an algorithm instead of  
>explicit codepoints we will make it possible to be "more future  
>proof".

I think neither me nor Mark, nor actually Ken, have been arguing
that the actual table should be the normative piece. But what Mark
and me, and I think also Ken, have argued for is that the "step-by-step"
algorithm be replaced by purely descriptive language based on
set operations/logic.

>The table in the appendix of the tables document is non- normative as it is a calculation on the tables that existed at the  
>time of publication of the draft (and might include bugs regardless  
>of how much we try to find the errors).

Fine with me.

>As Harald says, many people have on this list and privately told me  
>the list is "too short", but I really need someone that know the  
>script, and know Unicode (and the history, including bugs,  
>discussions on the unicore list etc), to say what to add.
>
>I do not mind adding more scripts!
>
>I want to add more scripts!
>
>But I am not the person that can say what to add.

So you want somebody to say that there will be no more bugs in
script X, and then you can add script X to the list? And you want
to base this on past number of bugs? Which way? If a lot of bugs
have been found, that may be an indication that there are not
too many more left. Or, if a lot of bugs have been found, that
may be an indication of more benig hidden.

Regards,    Martin.

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp       mailto:duerst at it.aoyama.ac.jp     



More information about the Idna-update mailing list