[CREATE] Colorfont Workshop results

[Hin-Tak Leung]

(quite a few off-topic things...)

I like XeTeX too. That said, there are more than a few ugly hacks :-). Their dvi driver (a fork of an older version of dvipdfmx) has an option override to embed fonts with licenses which forbid embedding. The current upstream dvipdfmx - managed by a Korean team - still doesn't have that "feature". :-).

Werner Lemberg seems to prefer LuaTeX more (in some private communications) - in terms of support for non-Latin scripts. LuaTeX also have more extensive scripting support for use outside of LaTex-like environment, I think.

I don't know much about emacs 24's Arabic/Hebrew right-to-left support - afterall, it has only been released a couple of weeks ago :-), and I don't read Arabic/Hebrew... Would be interested to hear from native users what they think... 

--- On Fri, 29/6/12, Dave Crossland <dave at lab6.com> wrote:

> On 29 June 2012 17:26, Hin-Tak Leung
> <hintak at ghostscript.com>
> wrote:
> >
> > fontspec is nice, but rather tied into XeTeX?
> 
> The important thing is that XeTeX is libre software. Nasty
> hack, but
> you can import PDFs made with XeTeX into Scribus frames ;-)
> 
> XeTeX is helped a lot today by www.tug.org/texworks and I
> used XeTeX
> rather than Scribus for my development of Cantarell :-)
> 
> SIL has developed some nice (sadly, as yet unpublished)
> XeTeX type
> design development document-tools too. The idea is that the
> document-tool inspects the font and generates a document
> based on
> properties of the font itself - ie, generate
> immersive-reading
> typography with words that include every letter combination
> possible
> given the characters existing in the font, so that the
> spacing and
> kerning can be thoroughly checked.
> 
> If Scribus supported OT features, one could create such
> software for
> Scribus with python scripting to test all OT feature
> combinatorial
> possibilities :-)
> 
> For Scribus to support OT features, it need only take
> advantage of the
> QT text shaping, which is based on
> http://www.freedesktop.org/wiki/Software/HarfBuzz...
> 
> > One other thing Ricardo might want to check out is
> > Pango... anyway, there is a lot text-layout expertise
> > in the TeX/LaTeX community, the web-browser
> (pango...)...
> 
> ...so its Harfbuzz that is relevant here, not Pango, since
> Pango's
> OpenType support is delegated to Harfbuzz.
> 
> http://behdad.org/text/ explains some of the history
> here, but it is
> now rather out of date...
> 
> Behdad, do you plan to update that document? :-)
> 
> > emacs 24 seems to have gained the ability to do
> > Right-to-left directions i.e. displaying arabic/hebrew
> > the way it is intended.
> 
> That uses m17n, which is a non-OpenType complex script font
> format.
> Used only by emacs. :-)
> 
> > I also seem to remember some W3C specs/RFCs about fonts
> somewhere...
> 
> W3C has the CSS3 Fonts module, and the WOFF format.
> 
> The WOFF format is just compression, and doesn't effect
> OpenType features.
> 
> The CSS3 Font module specifies ways to access OpenType
> features and
> browsers are - as Ricardo said - slowly implementing this.
> Microsoft
> is leading here - MSIE10 will have full OpenType support -
> and Firefox
> trailing them.
> 
> -- 
> Cheers
> Dave
>