Re: missing characters: combining marks above runs of more than 2 base letters

From: Asmus Freytag <>
Date: Sat, 19 Nov 2011 10:58:15 -0800

What is missing here is the recognition that having a *standard*
way of expressing such features ("text scoring") is desirable. Suggesting
that this be left of one-off implementations or "some markup language"
  is not helpful to people interested in using such features.

What is, of course. equally lacking is the shared understanding
that Unicode is not the proper *standard* to hang this on. Adding
scoping undermines the plain text nature of Unicode
and turns it into its own markup language.

Assuming text scoring is indeed rather widespread, and
used in a variety of contexts, then the proper place to seek
for a common and standardized solution would be a text-styling
standard such as CSS.

I believe it would be helpful if, in addition to whatever requests
that individual proponents are making directly to the CSS working
group, if the UTC were to communicate clearly to the CSS working
group about features that had been requested but were ruled to
lie outside the scope of character encoding.

 From Unicode's perspective, it would be an advantage to be able
to point users to an existing solution on the markup level, instead of
merely positing that one "ought to exist".

Therefore, I would consider it in the UTCs own best interest if
if could utilize whatever liaison relationship exists to alert groups
like CSS about such missing pieces in the overall text architecture.

This would be a much better model of cooperation than what has
existed heretofore where both sets of committees work entirely
on their own and UTR#20 attempts, after the fact, to unravel
the resulting incompatibilities.


On 11/18/2011 5:58 PM, Ken Whistler wrote:
> On 11/18/2011 5:36 PM, Philippe Verdy wrote:
>> I have absolutely no clear way to represent sequences like in this
>> example that use such elongated diacritic applied to runs of more than
>> two characters.
> Nor should you expect to be able to represent such things in plain
> text. Such
> conventions are not appropriate to plain text. As usual, your regular
> alternatives
> are:
> 1. represent the content with structured markup, or
> 2. represent the presentation with pdf (or similar)
> Expecting out-of-the box rendering engines to link the two
> automatically is
> unreasonable, IMO.
>> Of course I can still use two half marks in the
>> plain-text (only on the first and last letter), but what does happen
>> if I cannot (and in fact don't) mark their joining above intermediate
>> letters ?
> You'll get *both* bad representation of the content and bad presentation
> of the appearance. Such a deal!
> --Ken
Received on Sat Nov 19 2011 - 13:04:59 CST

This archive was generated by hypermail 2.2.0 : Sat Nov 19 2011 - 13:05:02 CST