Posted by Alan J. Flavell on 10/16/19 11:40

On Sun, 19 Feb 2006, Spartanicus wrote:

> "Alan J. Flavell" <flavell@physics.gla.ac.uk> wrote:
>
> >http://www.w3.org/People/mimasa/test/schemas/NOTE-ruby-implementation#css2-inline-table
> >turns out to be effective.
>
> Be aware that the "inline-table" value is not supported by Gecko or
> IE.

Thanks. The "simple Ruby markup" works pretty well, I'd say, in all
the browsers I tried (including those which only display the fallback
behaviour). IE implements it natively, and doesn't seem to let itself
be disturbed by the trick stylesheet. Mozilla, Firefox and Opera
produce quite a reasonable rendering, I'd have said.

But when we get to the "complex Ruby markup", it's a different story.
MSIE produces a total mess, and there is no fallback behaviour in
non-supporting browsers.

I guess it's more by good luck than by design that Mozilla gives an
impression of doing what was intended. The initially displayed result
looks strangely convincing, but any attempt to fiddle with it shows
that the rendering is highly unstable. Strange that the results on
Mozilla and Firefox are sufficiently different from each other to
suggest that they must have different code versions in their
rendering. Equivalent versions of Mozilla on Windows and on Linux
gave the same results as each other, though.

If you have any suggestion for improving the practical results, feel
free. Would you anticipate any benefits by trying to follow the ideas
which you set out here? -
http://www.spartanicus.utvinternet.ie/centered_image_gallery_with_captions.htm

I've more than run out of time, seeing that I've got a day job to
attend to, but don't hesitate to take a copy of what I've done so far,
if you feel tempted to hack it. (The copyright mark is an automatic
formality in this context AFAIAC.)

The actual ruby markup seems to be purpose-built for doing what
the O.P requested; if it's currently impractical to deploy it, at
least in its complex variant, that's a shame really. (I guess one
could process the Ruby through some kind of transformation which
produced actual table markup resembling what was suggested by Jukka,
so as to get something that would work in practice, but without having
to abandon the Ruby markup as the actual source.)

cheers

[Back to original message]