Re: The remaining issues list

Terry Allen (
Tue, 21 Mar 95 18:21:41 EST

| Here is a summary of the issues I listed, together with diverse other issues
| that arose whilst discussing aforesaid & aforementioned issues.
| (Unless I missed someone; please don't waste time saying
| I left you out -- if your issue isn't here, raise it!)
| Someone else please take over, I'm gone for 2 weeks.

May we hear from the chair?

| | The <DIV> element.
| Joe English <>
| DIV element -- I believe there was consensus on this.
| Peter Flynn <>
| Stick with <div class="chapter"> etc for the moment. Going for
| fully-fledged <div0> <div1> etc is probably too much for the user
| community.
| [disagreements still - Lee]

DIV might be put off till after 2.1, I'd say. If DIV may contain
any block elements, and does not enforce hierarchy, what concrete
and immediate need is there for it (a query, not a challenge)?

| | Universal ID attribute.
| Joe English <>
| Of the three changes, this is the only one that would require
| modifications to existing browsers
| Liam Quin <>
| This means that it can't be part of 2.1, but could be part
| of 3.0, right?
| [disagreements still; but the attributes could go in the DTD if browsers
| were allowed to ignore them (which they are), right? Lee]

Yes, but if you parse as SGML, you have to put them in your local
version of the DTD. I'm all in favor of a universal ID att, but
is it going to be called NAME?

| | Tables - must have, work still to do
| Peter Flynn <>
| What's waiting here?
| Jon_Bosak@Novell.COM
| strong suspicion that the issues [...] aren't going to get
| completely resolved in the next few weeks, whereas tables can be and
| should be.
| Joe English <>
| For tables, the basic content model seems pretty solid.
| There was a question about whether <TH> cells should be
| allowed anywhere in a <TR>; I think this was resolved.
| There may be questions about the formatting attributes,
| wrt compatibility with other table models, ICADD in
| particular. Is the colspec attribute OK?
| Brian Behlendorf <>
| What's left? It seems mostly complete to me, the netscape extensions
| are all stylesheet-able.
| [still under discussion, no consensus yet aboutthe details - Lee]

Let's hear it from anyone who doesn't want 3.0 tables, as is today,
in 2.1.

| ...
| | Footnotes - no consensus
| Joe English <>
| should there be a separate FN element,
| or should these be specified with <NOTE ROLE=footnote>?
| (I say the FN element should be reintroduced.)
| [still under discussion, no consensus - Lee]
| [[I think they should be a style sheet option, as in Panorama!]

Would Footnote proponents describe what they want out of it that
<NOTE ROLE=footnote>, or even just <NOTE>, wouldn't supply?

| ...
| HyTime links - no consensus reached

I believe that as of right now, no proposal has been made, either. What
is desired on this point?

| ...
| | Unicode support - no consensus reached
| Gavin Nicol <>
| will probably not be resolved in any great hurry. I should have
| something to offer regarding Unicode and what not this week
| Rick Jellife has got ERCS online at the SGML Open WWW site.
| Terry Allen <>
| Let's say instead, "character set encodings" for purpose of discussion.
| This is one of the big items that was supposed to get into 2.1.
| (Albert Lunde)
| [...] it looks like ERCS is itself in draft status.
| Peter Flynn <>
| I thought this one had been put to bed.
| [Liam: I think `put to bed' means `resolved', not `forgotten'.]
| (Amanda Walker)
| our client already has Unicode support designed and ready to implement
| as soon as I see a consensus on how Unicode pages will be labelled
| and encoded.
| [issues remain to be resolved - no consensus - Lee]

I would like to enable the support of charsets other than ISO 8859-1
and Unicode, even if we recommend that Unicode is the way to go. We
should be able to get that far in 2.1.

| | FIG extension - OK
| Joe English <>
| <FIG> -- the content model needs to be fixed to eliminate
| the pernicious mixed content
| (Amanda Walker)
| FIG definitely looks good, and is a reasonable
| stopgap for client-side image maps for many purposes.
| [work to do still, but close to consensus? - Lee]

Just a tweak needed. In general, when adding new elements, there
is no reason to perpetuate deprecated content models, as was done
here. No need for a DEPRECATED version at all, as happened here
through use of the two-version %body.content;.

| ...
| | Revised content models for LI and DD
| Peter Flynn <>
| Do we get to allow users to put text as direct content of <li>,
| <dd> etc without the need for an internal <p>?
| [?? - Lee]

I hope Peter would now agree with me that things are Right as is,
no change needed.

Terry Allen  (   O'Reilly & Associates, Inc.
Editor, Digital Media Group    101 Morris St.
			       Sebastopol, Calif., 95472
occasional column at:

A Davenport Group sponsor. For information on the Davenport Group see or