Re: HTML table model suggestion

Murray Maloney (murray@sco.COM)
Fri, 31 Mar 95 10:13:39 EST

> Paul Grosso writes:
> > As Daniel points out, the CALS model has tfoot before tbody:
> >
> > <!ELEMENT tgroup - o (colspec*, thead?, tfoot?, tbody) >
> >
> > This is to allow implementations to "pick up" the tfoot without
> > having to scan a potentially very long table (I've seen CALS tables
> > that were 20 pages long) since one might want to present the tfoot
> > at the bottom of each page of the table.
> But do we need to worry about printed tables when we are defining a markup
> for an electronic publishing medium, i.e., there is no definition of a page?

There is nothing in the spec -- or anywhere else -- that says
that the use of HTML is limited to online browsers. I can
cite two uses of HTML which do have a definition of a page:
the ICADD applications which convert HTML into Braille and large print,
and SCO's online documentation system which allows users to
print out whole chapters -- a collection of nodes -- can both
make use of table heads and feet.

For online uses, I can forsee browsers which present a persistent
head and foot while allowing you to scan the contents of the table.
In this example, the window is the virtual page.
> I like having the table footer defined at the end of the table body because
> it seems more natural to me, especially when I am authoring a *small* table
> by hand or via an SGML editor.

If you are authoring with an SGML editor, then it will be up to
that editor to decide how it will accomodate your needs --
possibly by reordering the screen presentation to match your
view of a table foot as an object which occurs at the end
of a table. If you are editing by hand -- oh, well... I guess
that you may have to adjust your conceptual model of a table foot,
just as you would have to do for a footnote which may not
occur at the bottom of a page online.

> - Bernie