<br><br><div><span class="gmail_quote">On 6/13/07, <b class="gmail_sendername">Harald Tveit Alvestrand</b> <<a href="mailto:harald@alvestrand.no">harald@alvestrand.no</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
The intent of "MAYBE YES" and "MAYBE NO" was:<br><br>- ALWAYS: We guarantee that these codepoints will be permitted in IDNs (at<br>this level of the standard).<br>- NEVER: We guarantee that these codepoints will never be permitted in IDNs
</blockquote><div>...<br><br>This is precisely the kind of information that belongs in Patrik's <span id="st" name="st" class="st">draft</span>.
Without a model of the intended usage, it is impossible to assess the
structure of the document. At least with this kind of information we
can begin to sensibly discuss the pros and cons of the changes over
what we had last December.
<br><br>However, it needs even more background information and
justification. Without knowing what you and Patrik meant by "Stable",
it is also impossible to assess how scripts should be assigned to that
category. After all, Thai is just as stable as Latin, if not more so,
depending on what is meant, yet you exclude Thai but keep Latin. I was originally
guessing that what you mean by "stable" is "has no characters that are
problematic for IDN", but in that case, Latin itself is not stable
because of the potential confusability between "1" and "l". Or, if
confusability is not the issue, then you really need to have some
examples of what exactly are the problems you are trying to prevent.
<br><br>That is, you need to provide more information as to what you
intend by "stable", with specific examples of scripts that you consider
stable and why, and scripts that you consider unstable and why. Failing
that, I don't see why every non-archaic script would not be stable,
thus obviating the need for your <span id="st" name="st" class="st">new</span> 4 distinctions instead of the 3 that we had up until a few days ago.
<br><br>Now, I found a hint of what <span style="font-style: italic;">may </span>be at issue in Patrik's further response:<br>
<br>
> Secondly, the ALWAYS and NEVER property values are only allowed on<br>
unproblematic scripts if we have a rough consensus that the<br>
codepoints will not move from ALWAYS to NEVER or vice versa given the<br>
algorithm we have to calculate the property value itself.<br>
<br>
If <span style="font-style: italic;">*that* </span>notion of stability is all that is being talked about, then it is very easy, and we have done it with a number of Unicode properties. Define the following:<br><br><div style="margin-left: 40px;">
Grandfathered_Always to be all characters that were Always under any previous Unicode version back to some base level (say 5.0)<br>Grandfathered_Never to be all characters that were Never under any previous Unicode version back to some base level (say
5.0)<br></div><br>Then modify the end of my message of June 13 to be:<br><br><div style="margin-left: 40px;">Then derive the following sets:<br></div><ul style="margin-left: 40px;"><li>Always = Grandfathered | (Favored &
<span>Functional) | Grandfathered_Always<br></span></li><li><span>Maybe_Yes = !Favored & Functional & !(Always | Grandfathered_Never)</span><span></span><span></span>
</li><li><span>Maybe_Not = (Archaic | (!Favored & !Functional)) </span><span>& !(Always | Grandfathered_Never)</span><span></span></li><li>Never = everything else<br></li></ul><br>Harald said a bit later:<span class="q">
<br><br>> So far, I've seen a lot of hand-wringing about the list of scripts being<br>too short, the list of scripts being Europe-centric, the arguments for<br>the list of scripts being too weak, the list of scripts including
<br>worrisome characters (IPA), but I have NOT seen ANY flat statement<br>"script XXX is unproblematic and should be included".<br><br></span>My response would be:<br><br>Each
script other than the archaic ones is no more problematic overall than
the Latin, Greek, and Cyrillic you have already included. Thus if you
include Latin, Greek, and Cyrillic, you should include them:
<br><br>Arab Arabic<br>Armn Armenian<br>Bali Balinese<br>
Beng Bengali<br>Bopo Bopomofo<br>
Buhd Buhid<br>Cans Canadian_Aboriginal<br>Cher Cherokee<br>Cyrl Cyrillic<br>Deva Devanagari<br>Ethi Ethiopic<br>Geor Georgian<br>Grek Greek
<br>Gujr Gujarati<br>Guru Gurmukhi<br>Hang Hangul<br>Hani Han<br>Hebr Hebrew<br>Hira Hiragana<br>Kana Katakana<br>Khmr Khmer<br>Knda Kannada<br>Laoo Lao<br>Latn Latin<br>Limb Limbu<br>
Mlym Malayalam
<br>Mong Mongolian<br>Mymr Myanmar<br>Nkoo Nko<br>Orya Oriya<br>Sinh Sinhala<br>Tale Tai_Le<br>Talu New_Tai_Lue<br>Taml Tamil<br>Telu Telugu<br>Tfng Tifinagh<br>Thaa Thaana<br>Thai Thai
<br>Tibt Tibetan<br>Yiii Yi<br><br></div></div>