The Unicode Consortium Discussion Forum (CLOSED)

The Unicode Consortium Discussion Forum (CLOSED)

The forum has been closed, but prior postings are accessible for reading.
 Forum Home  Unicode Home Page Code Charts Technical Reports FAQ Pages 
 
It is currently Fri Dec 19, 2014 12:34 pm

All times are UTC - 6 hours [ DST ]





Post new topic Reply to topic  [ 8 posts ] 
Author Message
 Post subject: PUA characters
PostPosted: Wed Mar 07, 2012 5:50 pm 
Offline

Joined: Mon Feb 01, 2010 6:18 pm
Posts: 79
I just want to bring up the PUA again. Now that we have the distinction of DVO vs EAVO, should there be a conversation about the common use of PUA characters outside of the East Asian context, and whether they should be DVO=S or T instead of U?


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 3:11 am 
Offline
Unicode Guru

Joined: Fri Dec 04, 2009 9:25 pm
Posts: 76
To a large extent, the "default" value for DVO is U, and DVO=S characters are the exception. More or less, DVO=S is used when the semantics of the character is incompatible with U.

It seems to me that since we don't have any semantics for PUA characters, they should have the "default", i.e. U. We can't really use T, since we cannot say that they have a different shape.


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 3:54 am 
Offline

Joined: Mon Feb 01, 2010 6:18 pm
Posts: 79
My concern is that if we have Private Use characters outside of CJK name characters - MUFI, CSUR, etc. - that the implementation mechanisms that use DVO=U are going to bypass the semantics and meanings of the PUA agreement for a given font. If we set an expectation of DVO=T, then the presence of a <vert> substitution allows for a T or S-as-T execution, and its absence would be treated, by default, as DVO=U. Thus, the font designer alone will determine the vertical semantics of PUA characters, rather than having software impose an interpretation that all PUA characters should remain unchanged in vertical contexts.


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 8:57 am 
Offline
Unicode Guru

Joined: Fri Dec 04, 2009 9:25 pm
Posts: 76
I don't want to solve the OpenType problems here, but

vanisaac wrote:
If we set an expectation of DVO=T, then the presence of a <vert> substitution allows for a T or S-as-T execution, and its absence would be treated, by default, as DVO=U.


Regardless of whether one uses DVO or EAVO, I think there is consensus that 'vert' needs to be applied to U occurrences (as well at T, of course). There is currently no other way to signal to a font that a glyph is upright, and a number of fonts depends on that signal.


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 1:48 pm 
Offline

Joined: Mon Feb 01, 2010 6:18 pm
Posts: 79
emuller wrote:
Regardless of whether one uses DVO or EAVO, I think there is consensus that 'vert' needs to be applied to U occurrences (as well at T, of course). There is currently no other way to signal to a font that a glyph is upright, and a number of fonts depends on that signal.


Is that consensus expressed in UTR #50? I mean, we can say that EAVO/DVO "can be used in the absence of other information, but can be overridden by markup ... and application preferences", but I think that these informative Technical Reports will probably be followed more religiously, and our OpenType/AAT/Graphite features will be ignored because an application will only check for a vertical substitution for EAVO/DVO=T characters. How many applications are going to even offer the option of vertical form markup or preferences, given a Vertical Orientation property that does the job for them? Should the conformance section explicitly state that it "should be used only in the absence of other information. Document markup, preferences in a layout application, and other defined vertical forms should be followed in preference to EAVO/DVO property values at all times."?


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 3:34 pm 
Offline
Unicode Guru

Joined: Fri Dec 04, 2009 9:25 pm
Posts: 76
vanisaac wrote:
Is that consensus expressed in UTR #50?


No because it is a consensus on a particular class of implementations, namely with OpenType fonts. That belongs to OpenType.

vanisaac wrote:
I mean, we can say that EAVO/DVO "can be used in the absence of other information, but can be overridden by markup ... and application preferences",


It's already in section 3.

vanisaac wrote:
Should the conformance section explicitly state that it "should be used only in the absence of other information. Document markup, preferences in a layout application, and other defined vertical forms should be followed in preference to EAVO/DVO property values at all times."?


EAVO/DVO is not a last resort method, which is were one would usually see this kind of wording.


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Fri Mar 09, 2012 7:17 pm 
Offline

Joined: Mon Feb 01, 2010 6:18 pm
Posts: 79
So Eric, I guess my question is, given a discrepancy between a font feature substituting a vertical glyph form and a Vertical_Orientation=U, should the application display the transformed vertical glyph form, or the upright form?

If a vertical font feature should override DVO/EAVO, what is the functional difference between the VO=U and VO=T property values?

Likewise, how should a discrepancy between VO and markup or program settings be handled?

Is there an expectation that these conflicts should be handled in a specific way, or is it going to be random?

Most importantly, if there is an expectation of how such a conflict is handled, does UTR #50 reflect the expected behavior?

If the expectation is that VO properties are ignored in the presence of font features, markup, and program settings, does this not, in fact, make the DVO/EAVO properties a de-facto last resort vertical presentation property?

Lastly, if the expectation is of a consistent presentation across applications, and I can get some acknowledgment that UTR #50 currently does not unambiguously prioritize in the case of discrepancies, how should UTR #50 be updated to reflect the expected prioritization of font features, markup, application settings, and the Vertical Orientation properties?


Top
 Profile  
 
 Post subject: Re: PUA characters
PostPosted: Mon Jun 11, 2012 7:19 pm 
Offline

Joined: Mon Feb 01, 2010 6:18 pm
Posts: 79
I'd just like to make one last request that PUA characters be classified as HO=Tu, SVO=Tu, and MVO=Tu in the revision 5 paradigm, so that implementations will, as their default behaviour, check for font-defined vertical/horizontal features before assuming an identical glyph in both horizontal and vertical layouts. This will enable seamless use of Private Use CJKV ideographs, without any change to existing fonts, but also provide for an implementation of a PUA standard where characters should be treated otherwise.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 8 posts ] 

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: Yahoo [Bot] and 2 guests


Quick-mod tools:
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Template made by DEVPPL.com