Comments: HTML3 at

Dan Connolly (
Wed, 8 Mar 1995 14:45:49 +0500

Daniel Glazman writes:
> > <!ENTITY % list "UL | OL">
> Once again, why do you separate UL and OL ???? This keeps compatibility with
> ancient GML styles but has no valid reason to exist any more.

You cannot make the installed base of HTML documents go away. Vendors
cannot responsibly remove features or break existing documents.

The OL and UL idioms are widely accepted and understood. The cost
of supporting them at this point is minimal. The cost of removing
them would be tremendous, and to what benefit?

> An ordered list
> is a list; an unordered list is also a list; difference between UL and OL is
> only an attribute (attribute can there be understood in its general meaning AND
> in its SGML meaning). I clearly see this as an intrusion of layout style in a
> SGML content.

The distinction between what should be an element and what should
be an attribute is arbitrary. (Technical (i.e. formal or emperical)
arguments to the contrary are welcome.)

<ul> is easier to type than <list unordered>. It costs less network
bandwidth. It's here today and it works.

> Why don't you use the Euromath DTD for mathematics ?????
> Or even ISO 12083 ?????

Because we need to display it interactively. Both of those DTDs
were considered (I believe).

> How can you convince LaTeX users that their zillions of formulas
> are convertible into HTML easy-going ????

We will write tools.

But keep in mind that LaTeX has a different set of design constraints
from those of HTML -- LaTeX is for beatifully typset books, and HTML,
while it may be printed, MUST be processed interactively.

The process of converting math documents from LaTeX to HTML will be
aided by tools, but it will not be painless -- converting a print
article to HTML is like converting a book into a screenplay. The
medium and the message are not inseparable.

> bra and kets (math meanings)
> are NOT ONLY a succession of a '<' (char) objects separated by a comma
> and a closing char '>'. A distribution is an OBJECT with a MEANING
> containing (SGML meaning) OBJECTS.
> * ....

I refer you to the CLASS attribute, which allows applications which
need to retain such information to do so.

> ========================
> > message digest such as MD5 for the linked object and is needed
> As we say in French, you want the butter and the money for the butter.
> You are designing a SGML dtd. Not a network protocol. No other comment :-(

But HTML is part of a distributed information system. Security and
reliability are an integral part of the system. Some applications will
require the ability to check the integrity of links. An MD5 attribute
allows applications to do that. This mechanism should be refined and
inhanced (e.g. to incorporate the information such as the DN attribute
required by S-HTTP), not eliminated.

> HTML is a SGML application. As a SGML application, it ****MUST**** be
> 100% consistent with the norm.

> How can you write such things in SGML ????

There is nothing in the SGML standard that specifies that SGML
document must represent "truth, beauty, and eternal understanding"
nor that representing information about physical characteristics
and layout are illegal.

On the other hand, practitioners in the field of technical documentation
have learned that documents are more reusable and long-lived if the
content is separated from the presentation.

Keep in mind, though, that HTML is not used exclusively for technical
documentation -- it's used for advertising, catalogs, informal
communication, and lots of short-lived information.

The author of the HTML 3.0 draft has weighed quite of evidence
carefully, and found a balance that he believes will support a wider
variety of distributed information applications on the web.
I think the draft needs some significant work, but it is in the
right direction.

Daniel W. Connolly "We believe in the interconnectedness of all things"
Research Technical Staff, MIT/W3C