Re: Renderers and XSL-FO, plus other thoughts...

On Sun, 04 Feb 2001, David Carlisle wrote:
> > (HTML out of FO? What's the point?)
> 
> FO is intented to be a platform neutral expression of the style of a
> document, one reasonable platform that you might want to render the
> document in would be an HTML browser (HTML+css). This isn't totally
> unreasonable is it? Actually of course given XML+CSS one might be able
> to cut  out the HTML altogether and just CSS style the fo,\ but using
> HTML does perhaps offer some fallback behaviour for inconsistently full
> css implementations.
> 
> David

A minor (maybe not so minor :-)) quibble: I would regard the "expression of
style" as the mapping itself, i.e. the XSLT+XSL-FO stylesheet. The FO obviously
still retains some "style" characteristics, particularly when it comes to
content, but I would not equate the formatting picture described by FO to
"style".

That being said, I agree that the _XSL WG_ intends FO to be something that
could be rendered into any format. One realizes from reading the spec that they
think rendering FO into handheld-device markup languages is a good idea too.
Which is nuts. I take a hardline position - I think that the XSL WG
is offbase on this one, badly.

With reference to my GIF diagram, note that as far as any user is concerned,
there is one big black box that takes (XML + XSLT) in and spits out whatever.
Internally there are a number of paths and exit points. I think the logical
exit point for XML+CSS going to a Web browser is before FO processing, not
after it. Same for just HTML. If XML+CSS is produced as an input, and
_printing_ is intended, then a fuller-featured formatter (not just FO) can take
over.

I think HTML+CSS is eminently reasonable as a browser input - I just don't see
FO processing as a reasonable waypoint to arrive at that.

Regards,
Arved

Received on Sunday, 4 February 2001 14:26:49 UTC