From: Mark E. Shoulson (mark@kli.org)
Date: Fri Jun 10 2005 - 15:27:49 CDT
Mete Kural wrote:
>>Similar questions apply to other <span>-formed effects (and the like).
>>What about a ligature (or some Korean jamos, for that matter) where one
>>character is bold and one plain? One italic and one upright? One in
>>48-point type and one in 5-point?
>>It's less clear if we're dealing with mandatory joining like in
>>Arabic, but I suspect that even in Arabic it's better to see both
>>characters there, unjoined in all their typographical hideousness, than
>>to have nothing at all (but a "missing character" box) to read.
>>
>>
>
>Yes the font change and size change cases are more problematic than color change (which I think should be relatively easy to handle). Although Mark, I think that displaying the medial shapes of the boundary characters is better than displaying the final and initial shapes of the boundary characters in case of a font or size change. They will still look ugly (two medial shape characters trying to join each other but missing each other in the air) but I think it is better than showing the final form of the previous character and the initial form of the next character which is bound to confuse the heck out of the reader as to what the text is trying to say since it will be interpreted as a seperate word in the case of Arabic.
>
Probably a matter of personal opinion, when dealing with such
distasteful options. Which is the least of evils? I guess if we can tell
that they should have been connected, then using the connecting forms is
best, even if they don't meet up. On the other hand, Arabic-speakers use
isolated letters in crossword puzzles, I am told. Presumably it is
possible to read even that.
Color-change is, indeed, the simplest case. An alef-lam ligature of
nastaliq and kufic, anyone?
~mark
This archive was generated by hypermail 2.1.5 : Fri Jun 10 2005 - 15:30:14 CDT