OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

cgmo-webcgm message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: Re[2]: [cgmo-webcgm] 24 hours to comment


Stuart made an interesting comment, but forgot to copy the TC list.  Since 
it actually taps into some concerns/thoughts I have had, I'll forward it 
and my thoughts...

At 10:42 AM 4/3/2006 -0700, Galt, Stuart A wrote:
>Looks ok to me.  Although I think that adding weird characters is an
>expansion of the spec and while allowed in the profiles it is somewhat
>dangerous because you are not guaranteed vendor support. (IMHO)

I'm not sure I understand completely the comment, but  "adding weird 
characters" presumably refers to allowing unicode in graphical text and 
non-graphical text.

"an expansion of the spec" is what puzzles me.  This was in 1.0 -- utf8 and 
utf16 were allowed in graphical text and non-graphical text, with *no* 
restrictions.  2.0 really doesn't change anything from 1.0, except:  fixes 
the CSL tail designators (PPF); and, clarifies in 3.1.1.4 about non-ASCII 
in URIs (and that clarification apparently either preserves or establishes 
IRI compatibility).

There are a couple of real issues lurking here.  It has always been the 
case that you could make a valid WebCGM 1.0 file with Farsi in graphical 
text and Vietnamese in apsids and fragments.  But don't expect widespread 
interoperability.  Interoperability probably won't happen outside of 
European sets, and maybe the CJK pages.  For WebCGM (or *any* 
internationalized standard), there is a real issue -- how do you define a 
limited repertoire within the tens-of-thousands of recorded unicode characters?

I think it is intractable in general, to write it (unicode-page 
restrictions) into the WebCGM standard itself, unless that standard is 
*only* for the use of ATA and S1000D interchange.  Otherwise, it has to be 
done in specific Cascading Profiles, or local (e.g., East Asian) technical 
communities that define their limitations (de facto or de jure).

On the other hand, maybe someone does have a restricted set in mind, that 
would suffice in WebCGM itself and would pass W3C muster?  (E.g., maybe 
western european, plus eastern european, plus CJK.)

Thoughts?  Proposals?

-Lofton.

>
>
> > -----Original Message-----
> > From: Lofton Henderson [mailto:lofton@rockynet.com]
> > Sent: Sunday, April 02, 2006 2:12 PM
> > To: Benoit Bezaire; cgmo-webcgm@lists.oasis-open.org
> > Subject: Re[2]: [cgmo-webcgm] 24 hours to comment
> >
> > Benoit (& everyone) --
> >
> > At 08:25 AM 3/30/2006 -0500, Benoit Bezaire wrote:
> > >[...]
> > >I did notice it. I do find it better, but still a bit harsh.
> > Again, I'm
> > >not in a position to propose something significant better,
> > so that will
> > >do (i guess).
> >
> > How about this one:
> > http://lists.oasis-open.org/archives/cgmo-webcgm/200604/msg00002.html
> >
> > Is it less harsh, while still clarifying the confusion?
> >
> > -Lofton.
> >
> >
> >
> >




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]