> I'm with Jody here. The mapping algorithm has 2 > parts: a fixed, reproducible one, and a second part > driven by user preferences. We should not impose the > font alias preferences of the author of the document > in the reader of the document.
The way i see this is that there are multiple users and the creator user may want to specify differnt preferences to the recipient user. The recipient of course should be able to make their preferences take precedence, i am not suggesting the user be forced to accpet a layout just that whoeever creates the document could want the option to specify very detailed fallbacks. Assuming the font mapping mechanism is some how crap and flawed (no offense you cannot guarantee it will be perfect) the document creator might know better than than either the recipient or the automatic mapping system. (pehaps they know that some symbol, say a filled heart shape in one font is not available in another but that there is a close enough symbol, say an outline heart shape that could convey the same meaning but it would be unlikely that the two would have a direct mapping) I am trying to say that even though it this is highly unlikely and probably complicated but not totally beyond the realm of possibility. I guess that is reason enough not to bother, i was just got distracted/hung up on the suggestion that it would *never* be useful. "All generalisations are false including this one" It is a non issue so ill be quiet now. Sincerely Alan Horkan
