I repeat that this philosophy has not be endorsed by the group,
sensible though it is (and Murray and I have expounded the same
view).
But it's also not applicable to HTML 3.0, which is now pretty big
and complex, with a spec running over 200 pages (probably would be
over 100 even if all the unnecessary info about how to process
SGML were deleted, as I urge Dave to do).
The HTML 3.0 table model was simple to begin with, but has been
acquiring CALS-ish bells and whistles. By now it's not simple
at all. As for editing by hand, I would be much less likely to
lose my place if my table header were in a header wrapper, in
the CALS manner, than in the present model, where undifferentiated
TRs may contain either header cells or body cells. This is a
large structural difference between 3.0 and CALS, which has
big implications for reuse of information. The issue is not
"minor simplifications to CALS filters" but rather, why should
we invent Yet Another Table Format when we could use a subset
of an existing one that is well established?
Regards,
-- Terry Allen (terry@ora.com) O'Reilly & Associates, Inc. Editor, Digital Media Group 101 Morris St. Sebastopol, Calif., 95472 occasional column at: http://gnn.com/meta/imedia/webworks/allen/A Davenport Group sponsor. For information on the Davenport Group see ftp://ftp.ora.com/pub/davenport/README.html or http://www.ora.com/davenport/README.html