[luatex] Why obsolete the fontforge name space?

[Yue Wang]

Hi

On Tue, Mar 24, 2009 at 3:28 PM, Arthur Reutenauer
<arthur.reutenauer at normalesup.org> wrote:
>> PS, on the dev-luatex mailing list
>> (http://www.ntg.nl/mailman/private/dev-luatex/2009-January/002217.html),
>> you doubted whether freetype2 will be sufficient for advanced fonts,
>> and in fact ft2 have most of the features you want:
>
>  From the list you made below, it seems that it doesn't because
> OpenType support belongs to HarfBuzz, as has already been said.
> FontForge already does all of that.
>

I know that project, and the idea is mentioned in the first mail of this thread.
Yes, FontForge solve all the problems FT2+HarfBuzz do, but:
- HarfBuzz+FT2 is more lightweigh/memory/computational efficient than
Font Forge.
- FT2 has been tested by thousands and thousands of software programs
(it is the basis of Unix GUI program + handhold device GUI + many
Windows and Mac Apps).
- HarfBuzz, though not public released, is the key component of Pango
[a part of GTK+] and Scribe [a part of QT application framework], so
it also has been tested by many software programs (All GTK+ and QT
program use HarfBuzz for text layout).
So, FT2+HB are the quasi-standard of the software world, This approach
is more robust and has much fewer bugs.
If there are sufficient funding, write such binding is beneficial.

Of course, I admit FontForge is useful in LuaTeX, and can solve many
problems FreeType cannot solve (like font generation).

>> So I think it is possible to write a similar binding.
>> Maybe we can apply for a Google SoC project for this?
>
>  TUG's candidature to GSoC has been rejected
> (http://tug.org/pipermail/summer-of-code/2009/000164.html), so, not this
> year.

bad to know that:(

>
>        Arthur
>

Yue Wang