From: "John Hudson" <[EMAIL PROTECTED]> > Philippe Verdy wrote: > > > Interesting point. This would be an argument for the developement (out of > > Unicode) of some standard technical solutions to exchange these private > > conventions on PUA usage, including exchange of character properties, etc... > > > > Why not then within fonts -- namely in Opentype tables for fonts built with > > these PUA assignments? > > Fonts are collections of glyphs. It is important to keep this in mind when proposing that, > somehow, character properties be included in fonts.
This is not so stupid: a font made for PUAs will also deliver at least a definition URI record. The actual character properties tables need not be present within the font itself. So just suppose that a XML schema is created to reference or embed various character properties data tables, and made accessible at a known URL or with a well-defined URN, then such XML document could be sufficient to describe completely what is in a private user agreement for PUAs. This means that this document URI can be inserted within the font (I'm quite sure that the OpenType format already has such meta-tags table, in which a URI could be inserted to access those properties). The private user agreement becomes explicit when a user tags its document with this URI. So the document becomes interchangeable even if it uses PUAs. And already today, the language tags could convey this URI within plain-text files containing those PUAs, or within standard XML files or within Word documents (using custom document meta-tags), or within HTML files (here also using <meta> tags)... This does not break what a font is made for: it is still a collection of glyphs and of glyph positioning and substituing tables. What this meta tag inserted into the font will indicate is that the font was made according to an explicit user agreement. Some fonts may be made that will work with several PUA agreements... by listing multiple URIs in them. The URI is then exactly like an interface signature. This would have an interest: if some PUA convention starts becoming popular, the private definitions files accessed from that URI will already constitute a viable experimentation for future characters to standardize. The same mechanism would also be used to tag Unicode versions or subsets supported in standard fonts...
