Dan:

If the fonts are reloaded when the image comes up (or the 'application' needs them) then you always get a fresh copy of the best font that fits your specifications. If one needs the same font on two platforms, then one will just have to buy it for both.

I sure don't mind having to reload a font on a new platform, or even load it every time I start up. It should go fast, and render only those sizes I've asked for...

Dave



At 12:58 -0800 2/14/02, Dan Ingalls wrote:
>"Mark Mullin" <mark at vibrant3d.com> asked...
>
>>So, QUESTION -
>>	If Squeak started using logic to get the platforms local typeface system to
>>render individual letters, and this low level 'give me an H in Merovingian
>>at 32 point italicized' was a nice snappy routine,  is it reasonable to
>>assume that Squeak could survive the introduction of this overhead and get
>>access to the underlying font system ?  Basically, you could look at this as
>>effectively throwing away the internal glyphs and just doing everything on
>>demand.  Of course, caching is often a good thing.
>
>Andreas and I have talked about doing just this for probably three years now.  You need a fairly simple cross-platform API for enumerating available fonts, styles and sizes, and then extracting the glyphs and other possibly useful metrics for any given one.  If caching is done in the form of (possibly augmented) StrikeFonts, then the overhead once a font has been extracted is zero -- it's exactly what we do now.  I think it's the way to go, and it just never quite made it to the top of either of our to-do lists.
>
>I believe you are right that this falls right into the conventional model of rights to use the OS fonts.  The only place where things start to get a bit fuzzy (legally) is if you develop some content on one platform, and then use it on another platform with the cached fonts.  What I hear increasingly these days is that even this would be OK as long as the source outlines are not used, but I can't say this with authority.
>
>	- Dan


-- 
_______________________________
David N. Smith
IBM T J Watson Research Center
Hawthorne, NY
dnsmith at watson.ibm.com