Minutes of the January 21, 2003 meeting of the PWG Character Repertoires Committee

Elliott Bradshaw, 2/14/03

Attendees

General Discussion

Need an agreed on way to query the printer capabilities.  Use of “implict rules” is less valuable.

May not be necessary to specifically account for Microsoft variations.  Maybe this can be handled with a vendor extension mechanism.

Consensus that adding attributes to the semantic model is a good goal.

People seemed to appreciate the background material on, e.g. Unicode, but it should be possible to get to the semantic model info easily.

After discussion, we concluded this approach is suitable for XML based data formats, but probably won’t work for existing PDLs.

There was discussion as to whether a printer can implement all the characters of a given repertoire, or if it would need to be able to report specific characters and ranges it supports.  For new designs, it should be fairly easy to assemble characters to a spec.  This is an assumption to clarify in the documents.

Suggestions for other character repertoires to examine:

There was discussion about whether the repertoire supported should depend on font.  An assumption of the current design is that printers use a system font with fall-through so that any character will print even if not supported in the current font.  For CSS systems this is within the intention of CSS.  For legacy PDL systems, or strict WYSIWYG systems, this may not be what the user wants.  Need to clarify these assumptions in the documents.

Rod Acosta from Agfa Monotype missed the meeting due to a travel snafu.  But afterwards, he agreed to bring our draft list to their specialists in Asian markets.  Agfa will suggest preferred repertoires for various countries.

Action Items 

Given this discussion, I took the following action plan:

1. Create a document with a short list of names for preferred character encodings and repertoires.  This will be reviewed and discussed, and ultimately published as a PWG document, similar to the media sizes list.

2. Make a specific recommendation for adding to the Semantic Model a new attribute on the Printer object called "character-repertoires-supported", with its description coming from the above document.

3. Take this to the SM, UPnP, and PSI groups to work into their roadmaps.

4. (Lower priority) re-work the Implementor’s Guide into a background document for education purposes.  Make it explicit about assumptions we are making.

5. Discuss with the CR group the proper handling of implicit support, and Microsoft.