From: Michael Everson (email@example.com)
Date: Thu Nov 06 2008 - 16:15:09 CST
On 6 Nov 2008, at 21:57, Asmus Freytag wrote:
> On 11/6/2008 1:33 PM, Michael Everson wrote:
>> No. Teuthonista usually stacks. For instance, diaeresis atop
>> diaeresis. Same as UPA.
> Good an example of flat and wide diacritics stacking. Just like I
I'm not looking at material in front of me. I'm tired from travelling.
>> There are also side-by-side accents, like Circumflex and Acute (if
>> I recall correctly; Karl can correct this). And in UPA, you find
>> circumflex and acute stacking as normal.
> Vietnamese has the circumflex and accent not stacking but half-ways
> stacking. Looks like the case of where you have one flat and wide
> and one tall and narrow, you get a range of behaviors.
I think you oversimplify, and in any case we have UPA and Teuthonista
both to contend with.
>> So it is quite possible that citations of both forms in the same
>> document could be required.
> Not clear that this has to be done in plain text.
Sorry, no. There are MILLIONS of citations in Teuthonista in projects
which have been going on for 150 years. The Germans will have a plain-
text requirement for this no less than the Uralicists did with UPA.
That really won't be negotiable, I am sure of it.
> If different languages / notations have different *preferences* it's
> not clear that these preferences *must* be presentable in plain
> text. They would look a bit odd, perhaps, but not be ambiguous.
This data must be respected, and there will be explicit discussion
texts about this published over the next while, in advance of encoding
> Rendering engines should be prepared to allow all modes of stacking,
> and CSS or similar protocols might need commands to set the
> preferred stacking mode, but as difficult as this would be to get
> support for, it still does not make a hard and fast case for this to
> be solved on the encoding level.
No one has made a hard and fast argument yet, Asmus. Kindly do not
Michael Everson * http://www.evertype.com
This archive was generated by hypermail 2.1.5 : Thu Nov 06 2008 - 16:17:12 CST