John Hudson john at tiro.ca
Sat Aug 20 18:44:45 CEST 2011

John Jenkins wrote:

> One of the reasons why Apple includes the 'prop' table in its version of the TrueType spec is to handle exactly this problem, because it lets font designer specify certain glyphs as RTL (among other things).  It would probably be more productive to get this or similar functionality into the OpenType spec than it would be to convince Unicode to set aside part of the PUA for RTL scripts.  

Since PUA codepoints are, by their nature, most often font-specific -- 
i.e. you can't correctly display the PUA text unless you have the 
appropriate font --, it makes sense for their properties to be recorded 
at the font level, rather than relying on external resources.

However, I don't think the Apple 'prop' table per se would be an 
appropriate addition to the Open Font Format (ISO MPEG standard version 
of OpenType) because as spec'd the 'prop' table duplicates some existing 
OT data (combining mark glyph classification in the 'GDEF' table) and 
presumes a different mechanism for optical margin alignment than that 
provided for in OTL GPOS features.

OpenType already has a table, 'GDEF', that is equivalent to 'prop' in 
function if not in scope, i.e. that provides glyph-level properties and 
categorisation. An extension to the 'GDEF' table could provide a 
mechanism for indicating directionality and perhaps other properties for 
glyphs mapped to PUA codepoints.



Tiro Typeworks        www.tiro.com
Gulf Islands, BC      tiro at tiro.com

The criminologist's definition of 'public order
crimes' comes perilously close to the historian's
description of 'working-class leisure-time activity.'
  - Sidney Harring, _Policing a Class Society_

More information about the mpeg-otspec mailing list