From: Peter Constable (petercon@microsoft.com)
Date: Tue Nov 20 2007 - 07:24:19 CST
> From: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org] On
> Behalf Of John Hudson
> Given that the vertical Mongolian script evolved from a right-to-left
> semitic script
> rotated 90 degrees counterclockwise, it seems to me that the obvious
> way to re-incorporate
> the script in a right-to-left setting would be to rotate it back,
> clockwise.
If you were embedding Latin-script text in to a RTL script, would you write the Latin-script text RTL because the writing it derived from was that way?
Doing what you suggest would present some implementation challenges: it means that an application needs to rotate each run of embedded Mongolian text 180°.
It also means that a trilingual reader that happens to known (say) Chinese and Arabic or English and Arabic as well as Mongolian will see Mongolian presented in opposite ways in polyglot text depending on which language it's combined with. I don't think that would be any more helpful for them than would be to present English RTL when embedded in Arabic.
IMO, the established conventions for Mongolian embedded in horizontal come from the most common scenarios to date, embedded into Chinese or English, and the conventions are LTR or, in short runs, vertically, and I don't think embedding in Arabic or Hebrew should change that.
Peter
This archive was generated by hypermail 2.1.5 : Tue Nov 20 2007 - 07:27:57 CST