From: John H. Jenkins (email@example.com)
Date: Thu Mar 06 2003 - 18:06:34 EST
On Thursday, March 6, 2003, at 01:42 PM, John Hudson wrote:
> The problem you have is that Apple, despite being involved with
> Unicode from the earliest days, have only recently shipped an OS with
> native Unicode text processing available;
This isn't quite true. Unicode support has been available on Macs
since Mac OS 8.5, and, via GX, even earlier than that. For various
reasons (many of them admittedly legitimate) application developers
haven't been taking advantage of that support, but it has been there.
> this text processing is only available to 'Cocoa' apps, i.e. apps
> writtenly natively for OS X, rather than 'Carbonised' i.e. updated
> from previous versions;
This is also slightly off the mark. Carbonized applications can do
Unicode through MLTE and ATSUI, and there are some who do.
> there are still very few native Cocoa apps, and even the MS Office
> suite under Apple has very poor Unicode support compared to the same
> apps under Windows;
This is a Microsoft issue, not an Apple one.
> Apple have been saying for years that they would support OpenType
> Layout features in some way, but have yet to do anything;
We've been deliberately vague on this issue, because OTL support
(directly or indirectly) on the Mac is not high on the to-do list of
the people who actually call the shots. In fairness, both MS and Adobe
have had similar problems, where OTL support has been provided
piecemeal because of how it fits in with their long-term strategies.
In the meantime, the other issue for adding OTL support to the Mac is
largely a resource one. We're a distant second in the OS desktop race.
We don't have the resources that Adobe and MS have (and, if you ask
people over there, they will also express frustration at how much
they're expected to do with their limited resources). Under the
circumstances, since we would have to rewrite large chunks of our
layout engine in order to use OTL directly, it's no wonder that we
haven't done it. There hasn't been a significant call for it among the
bulk of our customers, and what engineers we have are set to other
It's the same reason why Unicode support has been slow to come for some
higher-end applications. Rewriting the guts of something that's
already in use is costly and difficult, harder (in many cases) than
writing it for the first time. It's no coincidence that InDesign had
CoolType support from the beginning, whereas Illustrator had to wait
for it to show up.
I won't deny that Apple's had more than its share of ball-fumbling, and
Unicode is a big part of that; but this is largely a side effect of our
stumbling through most of the 90's without having a "next generation
OS" strategy that we would actually follow through on.
> Apple continue to rely on their own AAT font format, despite the fact
> that almost no font developers are producing fonts for that format (GX
> by any other name smelling as sweet). So while I sympathise with your
> concern that the fonts might appear to be 'Windows-only', I think the
> proper target for your frustration is Apple, who have been
> systematically fumbling the ball for several years now.
The AAT font format has been around for the better part of a decade,
and free tools (more on which below) to develop AAT fonts has been
around for pretty much that whole time. I think it would be fairer to
say that it didn't make economic sense for font developers to build AAT
features into their fonts. The number of languages which *require*
complex typography is a minority of all languages (at least, in terms
of economic clout), and with few apps (until recently) supporting it,
there was little incentive.
>> Also, are you saying that the requisite font layout features are only
>> doable via OpenType?
> In the initial version of the font, yes. It is an OpenType font, using
> OpenType glyph substitution and positioning technology to correctly
> render Biblical Hebrew from Unicode encoded text strings. That said,
> if funding is available, it would be possible to make a version of the
> typeface in the AAT format, using that technology to produce the same
> shaping (I've some experience with AAT, but I'm not sure just how
> difficult it might be to achieve some of the cleverer contextual stuff
> I have in the SBL Hebrew OT font; AAT is a very difficult format to
> develop for, and Apple's tools leave a *lot* to be desired).
Our newer tools are substantially improved, but you have hit one of the
problematic nails on the head here. The core of the AAT layout
technology is the 'mort' table, which earned its name by being deadly
difficult to make. The newer 'morx' table we use now isn't much
better. The 'mort' table was designed in the early 1990's, when memory
was much more at a premium than it is now and when processors were
slower. The 'mort' is compact, fast, and powerful, but at a price.
They're not terribly easy to make. In particular, most of the
typographers I know are more artistic than geeky and tend to shudder at
the thought of designing a state table.
But not everything in a 'morx' requires a state table. Ligature
support is *really* easy to do, and has been for years and years. The
fact of the matter is that the bulk of the font designers out there
don't even *know* that there's a way to add ligature support to fonts
on the Mac. We've tried to get the word out, but obviously we haven't
Still, when and where people have come to use to ask for help, we've
done what we could to provide it. Frankly, few people have come.
> The best long-term solution is for Apple to follow through on their
> promise to support OpenType Layout features, so that we have a
> genuinely cross platform font solution.
As I say, we've been careful not to make public promises in any detail
on this issue. I'm not aware of any time when we've said more than
that we're hoping to provide OT to AAT layout table conversion possible
using our tools. We really can't commit ourselves on this.
Given the fact that many application developers are basically echoing
the same sentiment (why waste money developing for the Mac when I can
get 90% of the same customer base without spending the money), I'm not
sure it's entirely a matter of it being our fault, however. Certainly
I'm not sure that the best long-term solution to having competing OSes
is for everybody to simply switch over to Windows, either.
The best *short-term* solution is for someone to tell them that if
they're interested, they can contact us directly and we'll see what we
can work out. We could probably work out AAT support for their
specific font without too much trouble.
John H. Jenkins
This archive was generated by hypermail 2.1.5 : Thu Mar 06 2003 - 18:51:50 EST