[XeTeX] support for variation & multiple master fonts

Jonathan Kew jonathan_kew at sil.org
Wed Dec 21 18:48:27 CET 2005


On 21 Dec 2005, at 4:44 pm, William Adams wrote:

> Adobe has pretty clearly stated on the OpenType list that they've no
> interest in doing Multiple Master fonts --- but they do advocate  
> that a
> type designer use the associated concepts when making fonts and use
> specific instances to get the optical sizes and other variations.
> Bummer about Apple deprecating them. It'd be nice if they'd at least
> freeze support and promise to maintain functionality.
>
> I've been playing with variations a bit and have some design projects
> which would benefit from them, so would like to see at least some sort
> of legacy support --- perhaps for MM this could be rolled into a
> separate back-end such as is envisioned for supporting non-Apple
> platforms?
>
> Like Peter, I think at least keeping a final version would be good.

If you re-read the original message, you'll see that if I go ahead  
and make the currently-planned changes, there will still be variation  
font support on Tiger. It just won't work on earlier OS releases any  
more. Does that matter?

I don't have any intention of removing it completely, so provided  
you're running 10.4.x, it'll still be there in the "latest and  
greatest" version.

Of course, if Apple pulls support out of the OS at some point in the  
future, there may be little option. So I can't guarantee that this  
stuff will still work on post-Tiger OS releases.

Just to clarify one other thing, in case some people are unsure of  
the distinction: this whole issue refers *only* to AAT-style  
"variations" (the width and weight axes in Skia) and the Type 1  
equivalent (Multiple Master axes). This is unrelated to AAT font  
features and OpenType layout features -- the stuff that supports  
complex scripts, alternate glyph selection, etc. None of that is  
expected to change (except perhaps by the addition of new features),  
as these technologies are very much alive.

JK



More information about the XeTeX mailing list