[HarfBuzz] Any comments on ISO OFFS (ISO 14496-22 & ISO 9541-4)

[Lars Knoll]

On Thursday 29 June 2006 02:34, mpsuzuki at hiroshima-u.ac.jp wrote:
> Hi,
>
> On Wed, 28 Jun 2006 11:08:31 +0200
>
> Lars Knoll <lars at trolltech.com> wrote:
> >> Do HarfBuzz developers have any comment on the standardization?
> >
> >There is one thing I would like to point out: A small part of the standard
> > has some patent restrictions on it (see
> > http://www.freetype.org/patents.html). It would be nice to finally get a
> > clarification from Apple on the issue, as I don't think as standard such
> > as OpenType (especially if it's going to be part of ISO) should have
> > patent restrictions on it. Maybe a standards body as the MPEG-4 committee
> > can achieve something here.
>
> Thank you for comment. Yes, the patent is the most
> important issue around TrueType technology. I expect
> ISO committee could achieve some conclusion, but I'm
> slightly pessimistic about MPEG-4 people's activity.
> Because, there is MPEG-LA but their licensing system for
> MPEG-4 has been unclear for a long time. In 2002, they
> had ever answered to EuroLinux as "the patents that will
> constitute the MPEG-4 Visual Patent Portfolio License
> support the charging of royalties on the use of MPEG-4
> Visual streams in Europe. ... Details of the actual license
> agreement are still being worked out", and no follow-up
> stories.
> http://petition.eurolinux.org/pr/pr18.html?LANG=en
>
> I've already reported the hinting patent issue to
> Dr. Komachi, a chief member of ISO 9541.
>
> >> So, I'm afraid, the approving OpenType font format v1.4
> >> itself does not support to implement portable OpenType
> >> renderer. More detailed specification is required, even
> >> if it's just another approval of Microsoft implementation.
> >> (there can be a nightmare: a MPEG-4 video player including
> >> Uniscribe can show closed caption correctly, but others
> >> cannot. Oops)
> >
> >Uniscribe is not the only implementation that implements the script
> > specific rules. Qt and Pango both implement the same rules for their text
> > layouting engines, as far as I know ICU also has some support for them.
>
> Thank you for giving overview. I guess, the reason why
> Qt, Pango (and ICU) follow the rules for Uniscribe is
> to utilize/divert the OpenType font products designed
> for Uniscribe (and "replace-ability", if there's).
> If there's something I miss out, please let me know.

Yes, we follow the same rules. The documents on Microsofts web site are the 
de-facto standard for support of these scripts through opentype.

Lars