From: Philippe Verdy (verdy_p@wanadoo.fr)
Date: Mon Mar 23 2009 - 07:36:24 CST
> [mailto:unicode-bounce@unicode.org] De la part de Mahesh T. Pai
> Roozbeh Pournader said on Sun, Mar 22, 2009 at 08:14:11PM -0700,:
> > > True; even Pango is likely to be outdated.
> >
> > Outdated how? As far as I can tell, Pango tends to support
> the latest > version of Unicode in each release, and major
> releases happen every six
>
> ugh. ugh. Make that:-
>
> "If you are using 2.0.X of OO.o, your version of Pango too is likely
> to be outdated".
>
> Because 2.0.X dates back to 2006.
>
> > Also, Unicode 5.1 support has been in Pango since March
> 2008, a > short while before the official release of the
> standard update > itself in early April 2008.
>
> I follow the relevant developer's list for pango. ;-)
But how is Pango's version issue related to the support of most scripts out
of the BMP ? These non-BMP characters are (for most of them) for simple
scripts that don't need advanced layout features (the supplementary
ideographic characters are still simple, even if there are many of them, and
most characters in the SMP also have simple layout with basic one-to-one
character-to-glyph mapping, allowing support in more basic fonts).
Most of the changes in Pango have been to support advanced features for
correct layout of the complex scripts in the BMP, or for looking up for
fallback fonts (I don't think that there are a lot of relevant fallbacks for
character blocks in the SMP or SIP).
The only "complex scripts" in the SMP I can see is for the western musical
notation, for which text layout engines are not enough (you need a specific
upper alyer protocol and a specific layout engine), and similarly, this will
be the case for Egyptian or Mayan hieroglyphs, whe they'll get encoded.
With the current state of Unicode, there's no problem in Ooo for supporting
non-BMP characters, since at least version 1.1, that was already Unicode
compliant (i don't know if this was completely the case for prior versions
of OOo or Sun StarOffice). If there were issues, it was not within Ooo
itself but in its dependant system requirements, and I think this is still
the case with Ooo v.3.
So effectively what I said initially was right and the question was badly
formulated as it was based on a misrepresentation of the problem. Well I
won't retry older version of OpenOffice because they have lots of other
problems or missing features, but the problems were certainly not what was
exposed when Peter affirmed that OpenOffice 2.x did not "handle" Unicode at
all when asking for a "confirmation" here about the same status in version
3.
Given that OpenOffice is completely free of charge and easy to install, I
don't understand why he did not simply took a try by himself to experiment
with the huge progresses that havge been made in the recent releases,
notably in terms of GUI interface (and memory management, editing features,
better support for tables and page layouts and styles...), but not
significantly in terms of Unicode support which is very good since long and
mostly equivalent and as much correct as older ones (with just minor changes
in the embedded ICU library for supporting more locales, but not much for
supporting more scripts or character blocks.
In fact this is the integration of ICU that has allowed excellent level of
support of Unicode features in OpenOffice, as well as better support for
transliteration, case mappings, collation and sort, date/number formating,
... The other source of excellent internationalisation is also the
integration of data from the CLDR project and other internationalisation
projects or efforts). The current version of OpenOfice cna now really
compete with MSOffice for most usages (including when working with tables,
summaries, indexes, and section heades/footers, or when embedding ther
documents from one program to another (all things that were not impossible
to do but really complicated in prior versions of OpenOffice).
I really appreciate to have now OpenOffice 3 installed on several of my
systems, withouthaving to buy supplementary licences of Ms Office which is
installed now on just one system (and I don't want to buy more versions,
given that I'm the only user of these systems and don't need multiple
licences at the same time). This gives more freedom of use with really
equivalent features most of the time, including for more advanced edtiing or
custumizations of documents.
But OpenOffice or even MSOffice is not the only software I use and need
everyday, and other editors are also needed, from IDEs like Eclipse or
Visual Studio, to more basic Notepad or Notepad++ and GNU editors like Vim,
or command line tools like GNU sed, or script engines for more complex text
editing with Perl or Python, or various reporting tools to automate the
creation of some documents. There's definitely not any single suite that can
match all the needs for text editing, in either rich text or plain text.
Most of these tools were chosen for various reasons, but most of them have a
ccompliant support of Unicode (with the exeception of Notepad++ whose
interest remains its speed of loading, but whose caveats are many bugs or
limitations for true internationalization)
This archive was generated by hypermail 2.1.5 : Mon Mar 23 2009 - 07:40:18 CST