[OpenType] OpenType "Recommendations": First Four Glyphs in Fonts

Ken Lunde lunde at adobe.com
Tue Jul 14 15:59:20 CEST 2015


Karsten,

By "first glyph" do you mean GID+0 or GID+1. If the latter, I would agree, as it is common practice for GID+1 to be a "space" glyph with a non-zero advance. I still think that there is value for GID+0 to be marking.

BTW, Acrobat once required that GID+0 be non-marking for its substitution fonts, but that requirement was lifted several versions ago. Acrobat is now Version 12, and if memory serves, this particular requirement was lifted in Version 7.

Regards...

-- Ken

> On Jul 14, 2015, at 5:52 AM, Karsten Luecke <karsten.luecke at kltf.de> wrote:
> 
> Message from OpenType list:
> 
> 
> Hello Mr Lunde,
> 
> given this and your OT list exchange with Adam, there could be a single 
> recommendation:
> Make space glyph the first glyph of the font.
> (It would "automatically" serves as .notdef – with or without a name –, 
> albeit a shapeless one.)
> 
> Best wishes,
> Karsten
> 
> 
> On 13.07.15 23:14, Ken Lunde wrote:
>> Vladimir and others,
>> 
>> Has anyone considered making changes to the "First Four Glyphs in Fonts" subsection of the "Recommendations" section of the OpenType Specification to agree with common practice or modern OpenType font consumers? See:
>> 
>>   https://www.microsoft.com/typography/otspec/recom.htm
>> 
>> We (meaning Adobe) looked into this about three years ago, and while such recommendations were meant to apply to OpenType fonts with TrueType outlines ('glyf' table), our conclusion was that we found that no modern environments depend on the presence nor characteristics of GIDs 1 through 3. GID+0 is special regardless of the underlying outline technology, but a single-item table doesn't make much sense. Also, the name of GID+0 is ".notdef" only for TrueType and name-keyed CFF fonts; CID-keyed CFF fonts have no glyph names.
>> 
>> Besides, I also suspect that that "Additional recommendations" are also suspect (pun intended). At the very least, we should consider adding a statement to the effect that these recommendations are based on older environments, and that modern environments no longer depend on them.
>> 
>> Regards...
>> 
>> -- Ken




More information about the mpeg-otspec mailing list