Re: Mark-Driven Script Categorisation

From: Philippe Verdy <>
Date: Fri, 18 May 2012 04:32:24 +0200

2012/5/17 Richard Wordingham <>:
> On Thu, 17 May 2012 20:41:19 +0200
> Philippe Verdy <> wrote:
>> Is it really the Latin letter x in question there, if it's use is to
>> be a visible placeholder to hold diacritic vowel marks ? The Latin
>> letter has the problem of is dual case (not found in the Lao script,
>> and a too large variation across many font styles, when the
>> multiplication sign × would probably fit better for its use as a
>> placeholder.
> You're the only other person I've met who thinks that it is
> U+00D7 MULTIPLICATION SIGN, but the evidence is against us.  Emacs,
> Ununtu and Windows all reckon that the Lao keyboard has 'x'

And Google's input tool (available as an extension for Chrome,
displaying virtual keyboards or implementing transliteration input
methods) does not even map any character on this key in the shift
state (in my opinion it should map the multiplication sign U+00D7 if
the Latin letter is not the prefered character).

Of course, Lao fonts will need to support the combination of × with
vowel marks correctly, just like they support it with the dotted
circle (which is apparently not the preferred symbol for showing the
Lao vowels marks in isolation as it is easily confusable with ວ at
small font sizes).

Another question would be : should the Lao chart in TUS still use the
dotted circle symbol as the placeholder, may be it could be as well a
dotted cross, leaving the plain cross only for actual use in texts,
and the dotted cross for defective uses of vowel marks ?

Are there other scripts that would prefer another placeholder symbol
as a bearer for their diacritic marks ? It looks like Arabic generally
prefers using a tatween for example (no problem : it is already
encoded, but why not a dotted tatween as well for the default display
of defective uses ?).

Other candidate bearer symbols : the small dotted square (centered on
the x-height or on the mathematical center line), the dotted upper
horizontal joiner in Devanagari (which could also include a plain
joiner working like the Arabic tatween), or a baseline horizontal
joiner (for cursive styles of Latin or Cyrillic, different from the
ASCII underline), or the half-rectangular symbol for space (in non
cursive styles of Latin or Cyrillic).
Received on Thu May 17 2012 - 21:37:11 CDT

This archive was generated by hypermail 2.2.0 : Thu May 17 2012 - 21:37:12 CDT