Dave Pawson wrote:
> Are the language elements not an illogical mix?
> font-name and font-name-asian might be illogical in one document?
>
> style:shadow and text:shadow?
>
> fo:margin and fo:margin-top etc?
>
> padding and padding-right (basically the shortform attributes and the
> longer forms)
>
> Doubtful if it will impact defaults, but writing mode tblr then
> setting some asiatic
> script that required tbrl etc?
Those are interesting points...
The script specific settings are not mutually exclusive. They allow to
select a font based on the script type, so the same paragrapg can use a
different font for Chinese (asian script type) text than it uses for
Arabic (complex script type). The overall feasibility of default fonts
however is something that I am not sure about.
There is no text:shadow, so I assume you mean fo:text-shadow, which is
in fact something different than style:shadow. The former controls a
shadow effect for characters, while the latter controls the shadow of an
object (which may contain characters with another shadow effect), e.g. a
page or a frame. The default for both of these should be to have no shadow.
The short forms probably should not get a default. Their default would
rather be reflected by the defaults of their individual components.
I am not sure about the writing directions, but I think our defaults are
OK as long as they are consistent. In general, whenever there is
potential for conflicting attributes we must make a choice and write
that down in the spec. From an interoperability point of view, an
application is free to disagree with out choice as long as it says so in
the files it produces.
Bests,
Lars
