Re: The remaining issues list

lilley (
Wed, 22 Mar 95 12:44:00 EST

Christophe ESPERT writes:

> In message <> 22 Mar 95
08:34:52, wrote:

> > Christophe ESPERT wrote:

> > > People keep saying that there is the backward compatibility problem.
> > > I understand this point of view but I do not think this is such a big
> > > issue. Some people want to make them HTML 3 with no effort and that's
> > > what they call the backward compatibility requirement.

> > For "Some people" I'd read 99% of the WWW users. We should consider
> > the "competition": whatever one may think of Netscape, to date they
> > have added their extensions in a way that usually does not totally
> > break documents/software ignorant of them.

> Do you really think that 99% of W3 users need the new features
> proposed by HTML 3.0?
> Do you think that 99% of the documents on the Web need to take
> advantage of tables for instance?

Yes. The user population is already voting with it's feet. That is why
putting out an HTML 2.1 with tables, fig, sup, sub and so on is a good
idea (provided it does not delay the release of HTML 3.0) - to head off
the march to incompatible formats.

Lots of pages already use align for images; quite a few are starting to
use tables. Look at all the pages that use GIFs of tables. I suspect
even more people would like to use these features but are hanging back
from making their pages browser specific.

2.0 represents the rough concensus of mid-94. A 2.1 that represented
the rough concensus of early to mid 95 and additionally was compatible
with HTML 3.0 would give a lot of people the freedom to go ahead and use
the features they have been asking for. But if it were agreed on tomorrow,
it would still take 3 or 4 months to get through the standardisation hoops.

But this just gives us a breathing space to get HTML 3.0 finished and out
while it is still relevant and still has a chance to be the standard that
is used, rather than the standard that is ignored.

> > If we define a "standard"
> > for HTML 3.0 that is cumbersome migrate to, the market may
> > vote with its feet for something else. I don't want a standard that
> > needlessly increases fragmentation in the web.

I concur. If we do put out an HTML 2.1, this should not (IMHO) be license
to take a year or two making HTML 3.0 an all singing, all dancing,
HyTime compliant, super everything, incompatible with all existing
documents, late, irrelevant ... hell, we could take time out to make
HTML 3.0 PREMO compliant as well if we really want to delay things.

I suspect HyTime has a lot to offer. But that will need to be discuseed at
some length and it opens up a big can of worms. Someone (sorry, don't have
the ref to hand) suggested that it might be appropriate to look at HyTime
for HTML 4.0, and I suggest this would be a suitable way to go.

Alternatively, if Mr Espert wants to move faster than this and backwards
compatibility is not an issue for him, perhaps influencing the text/sgml
content type would be a better way to proceed?

> As far as Netscape has done with their HTML "enhancements", I'd
> say that maybe the new element types they introduced were necessary
> but they shouldn't have done it this way.

We all say that. The point is not what they should have done, but that
they have done it and that people have enthusiastically taken up the

> Once again I don't see why you want to migrate all the documents to
> HTML 3? Only those documents that need the features introduced by
> HTML 3 need such a migration (e.g. a document in which a table should
> have been present in the first place).

I think that there are more of these documents than perhaps Mr Espert

Chris Lilley
| Technical Author, Manchester and North HPC Training & Education Centre  |
| Computer Graphics Unit,        |     Email:      |
| Manchester Computing Centre,   |     Voice: +44 61 275 6045             |
| Oxford Road,                   |       Fax: +44 61 275 6040             |
| Manchester, UK.  M13 9PL       |      X400: /I=c /S=lilley              |
|                 /O=manchester-computing-centre /PRMD=UK.AC /ADMD= /C=GB/|
|<A HREF="">my page</A> | 
|This is supposed to be data transfer, not artificial intelligence. M VanH|