Re: removing HTML flow objects?

[David Megginson <ak117@xxxxxxxxxxxxxxxxxxx>]

Frank Boumphrey writes:

 > >>I don't think that the first point was as much of a boon as one might
 > >>think: the HTML flow object set supported was nearly useless for
 > >>pre-CSS browsers (such as the still-widely-deployed Netscape 3.0)
 > >>since it didn't include most basic HTML element types like <UL>.
 > 
 > The figures that I have heard is that by the end of this year less
 > than 10% of users will have a 3 or worse browser. With the next
 > version of Opera supporting CSS are be really going to allow a
 > small percentage of users to eviscerate what was once a promising
 > new development?

I find that number surprisingly low (IS departments in big companies
are conservative about upgrades) but even if the number turns out to
be true, it doesn't invalidate my original point: the HTML flow
objects in the current XSL draft do not generate good, general-purpose
HTML.

Remember that HTML is supposed to scale up or down to different
devices and to different media (such as a voice reader for the
visually-impaired or for an automobile driver on the highway).  A
document full of <span> and <div> elements with CSS style attributes
will provide very little useful information for such processes.

Again, it would make much more sense to create an XSL-like
transformation language to produce _any_ structured output (XML, HTML,
or SGML) rather than a tiny and arbitrary subset of HTML 4.0.  This
task is, however, probably outside of the XSL WG's mandate.


All the best,


David

-- 
David Megginson                 david@xxxxxxxxxxxxx
           http://www.megginson.com/


 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list