On Wed, 2003-04-02 at 11:38, Edward H Trager wrote:
> OK, I have a few questions for Jungshik Shin, Edward Cherlin, and/or
> anyone else on this list who can answer these (although they may not be
> easy to answer):
>
> (1) Aside from the fact that Pango is written in C and SILGraphite
> (OpenGraphite) is written in C++ (and this in itself could lead to a
> flurry of opinions...), in your opinion, which library, Pango or Graphite,
> appears to have a better architectural foundation, or are they pretty
> similar?
They are apples and oranges.
Pango is a layout library, meant to be used by applications;
with sophisticated handling of attributed text, bidirectional
text, support for editing, a markup language for easy
programmatic creation of attributed text, etc.
SILGraphite is an engine for laying out text using one
particular font technlogy, with some simple higher
level capabilities.
Pango is designed to work on top of multiple different font
technologies and text layout engines; using Pango on top
of SILGraphite in fact makes a lot of sense architecturally
if you want to use Graphite fonts. Though nobody has
done the work to make that possible.
Pango, of course, has the better architectural foundation ;-)
[ In case it isn't obvious, that's a joke ... I don't think
you can compare them ]
> (2) If Pango is already fairly "mature" (quoting Jungshik), then why are
> Daniel Glassey, Alan Ward, et all working on SILGraphite instead? Is
> SILGraphite going to provide, or already providing, something that Pango
> does not? Or is this just a case of one group initially working on
> Graphite for Windows, and only later realizing the limitations of a
> single-platform approach, while the other group of Owen Taylor et al. were
> simultaneously working on Pango for Linux?
>
> (3) If you had to choose between one or the other, which would you use?
>
> (4) Is Pango going to become, if it is not already, the *de facto*
> standard for complex script rendering in the OpenSource world?
Hard to say. It doesn't seem to be happening right now.
My goal for Pango development has really been "the really good
layout library engine that meets the needs of GTK+ and GNOME".
That doesn't mean that it is any way tied to GTK+ and GNOME,
just that:
- I haven't been interested in removing the GLib dependency;
much work for only political gain. (*)
- There isn't much documentation about using Pango without
GTK+
- I haven't really had the time to promote the wider use of Pango.
> (5) What the heck is QT using ?
It's own code.
> I and a colleague are in the beginning stages of designing a library which
> will we are considering building on top of Pango. But, if Pango is not
> the right choice, or if SILGraphite would be a better choice, then of
> course we want to know that. Since we don't want to be tied to a single
> GUI toolkit, we assume that whatever QT provides is not really an option.
> But, of course if someone were to tell me that QT's solution is the best
> of the lot, then certainly I would take the time to investigate it.
Qt's solution is tightly tied to Qt, as far as I know.
I think a major problem with using SILGraphite as a complete solution,
is, that while it is cool technology, font creation will be largely
driven by Windows and thus to OpenType fonts. So, you really need
something that can support OpenType as well.
Regards,
Owen
(*) For desktop uses. Embedded uses are different and Pango's use
of GLib may be a problem there.
--
Linux-UTF8: i18n of Linux on all levels
Archive: http://mail.nl.linux.org/linux-utf8/
