Fw: [I18n] Sun's press release ?

["Edward Lee" <edward.@kimo.com.tw>]

-----原始郵件-----
寄件者: Owen Taylor <otaylor@redhat.com>
收件者: i18n@XFree86.Org <i18n@XFree86.Org>
日期: 2000年8月31日 AM 08:51
主旨: Re: [I18n] Sun's press release ?


>
>Pablo Saratxaga <pablo@mandrakesoft.com> writes:
>
>> Kaixo!
>>
>> On Wed, Aug 30, 2000 at 02:13:27PM -0700, Steve Swales wrote:
>>
>> > Actually, it's far simpler to use than it might seem from the description
of the
>> > modules.  GNOME, when built on Solaris works quite well with our
input/output
>> > method technology, unmodified.  Our goal is to provide Sun's tried & true
>> > technology to the open source community, so that systems like GNOME and KDE
>> > which already use the underlying X I18N framework can reach their full
potential
>> > for internationalization on open-source platforms, as well as on Solaris.
>>
>> Mmh, Gnome (wel Gtk+ in fact) plans to include string rendering by itself
>> (trough pango http://www.pango.org/), and I hear rumors that KDE (or Qt ?
>> I'm not sure here) have similar plans.
>>
>> Do you think that some integration and code reuse could easily be reached ?
>
>Repeating, with less authority, some of the thigns that Steve and
>Hideki have said, there are multiple things being provided here, that
>go beyond text rendering. While code reuse may be possible, it is far
>from the only thing that GNOME/KDE get out of this. Some of the main
>benefits I see are:
>
> - It provides UTF-8 locale and complex-text locale support for legacy
>   apps (GTK+-1.2, Motif, etc.)
>
>   Now, this support won't be as good in all ways as what can be
>   provided by a system like Pango. But the difference between seeing
>   your UTF-8 filenames as junk in, say, Acroread, and seeing them
>   correctly is huge.
>
> - IIIMF is an important step forward for input methods.
>
>   XIM is a rather poor match for a modern system like GTK+-2.0/Pango
>   (problems include locale dependent encodings, dependencies on X
>   rendering, poor support for switching input methods on the fly,
>   etc)
>
>   IIIMF is designed so that, for GTK+, we can completely bypass
>   the Xlib rendering/input method code and bind to it directly.
>
> - Some code reuse is most likely possible of code from the ouput
>   methods for Pango.
>
>   How/if this would work is still something that needs to be worked
>   out. It might be possible to create an adapter module for Pango
>   that allowed output modules to work. A more straight forward port
>   of any interesting modules to the Pango framework might also be
>   possible.
>
>[ Note, I have not yet seen any of the code, and only have a general
>  idea of the components of this code release. ]
>
>From my conversations with the Sun people, I think they recognize that
>this code that they are providing doesn't necessarily solve all the
>problems we need to address for X, and are interested in working with
>the open source community to move forward, even if that means that
>some of their past work would end up being eventually abandoned.
>
>But it certainly a very significant contribution that will do a lot
>for providing solid internationalization on open source platforms.
>
>Regards,
>                                        Owen
>_______________________________________________
>I18n mailing list
>I18n@XFree86.Org
>http://XFree86.Org/mailman/listinfo/i18n

--
Warm Regards,
Edward Lee
--------------------------------
大道廢，有仁義。智慧出，有大偽。
--------------------------------



To Unsubscribe: send mail to majordomo@linux.org.tw
with "unsubscribe xcin" in the body of the message