W3C home > Mailing lists > Public > www-html@w3.org > October 1999

RE: Future version of HTML!?

From: Tim Bagot <tsb@earth.li>
Date: Sun, 24 Oct 1999 11:01:32 +0000 (UTC)
To: HTML mailing list <www-html@w3.org>
Message-ID: <Pine.LNX.3.96.991024095414.3836S-100000@c32.keble.ox.ac.uk>
On Sun, 24 Oct 1999, Keith Bowes wrote:

> When I read about new standards, one question always comes to  
> mind:
> What's the point of new standards if no one uses them?
> Hence, my proposal.

What makes you thinks that people are more likely to implement your
standard than the other ones?

> Here are some things I can't find a standard equivalent for.   
> Please tell me if there is some equivalent:
> accomplished with <BGSOUND> or <SOUND> in the <HEAD> or <EMBED  
> Autostart="true" Hidden="true"> in the <BODY>).

The OBJECT element ought to do this.

> *Scrolling and sliding text (like can be accomplished with the  
> <MARQUEE> tag).

This is a job for scripting, applets, or just animations. This approach
allows for far greater flexibility without making HTML horribly
complicated (and therefore more difficult to implement).

> *A specification of how much space between adjacent frames  
> (like can be accomplished with <FRAMESET Border=>).

Similar functionality can be achieved with the CSS box model. If your HTML
5 is to include frames and CSS, you will have to write a new CSS as well.
The decision not to include frames in CSS was not an arbitrary one.
Likewise the decision not to include them in the HTML 4.0 Strict DTD.

> >The id and class attributes are designed to do just that and  
> are exposed for all elements.
> What can class do besides apply stylesheet rules?  What can id  
> do besides identify an element (so you can change the  
> attibutes of only that element)?

What can "made-up" attributes do? What can they do that can't be achieved
with id/class?

> I'm talking about true dynamicness and animation.  Perhaps  
> you've noticed the rise in the use of Internet Explorer since  
> "Dynamic HTML" was introduced in version 4.  The reason is  
> undoubtably that people know what they want web pages to be.
> It would be pretty hard for the W3C to control everybody,  
> despite how much they want to.

DHTML is just scripting, and not standard across implementations. We have
ECMAScript and the DOM, both strongly influenced by what people were
already doing. The intention is that both authors and implementors should
know what is required and what is provided. As you say, the W3C does not
have any direct control over them; another standard will not change this.

> I don't agree with you.  My proposal is actually based on  
> ACTUAL use.
> Go around the Internet.  How many people use XML?  Not very  
> many (if any).  How many people use non-standard constructs to  
> achieve what's impossible with standard?  Most.  You'll see  
> that my proposal will probably be accepted better by the  
> majority than the current standard.

Not many people use XML. XML is still new. Nobody used CSS when it first
appeared. Constructing XML applications is not a job for the amateur; as
more are created more people will find uses for them.

Yes, people use non-standard constructs. Often, what they do is possible
within the standards, but fails in practice because the standards are not
everywhere (or, in some cases anywhere) well-implemented. Again, another
standard will not change this.

> I agree, Netscape has implemented their innovations poorly.   
> But Microsoft has applied, fixed, and expanded these  
> technologies so that they actally do look cool.  Personally, I  
> love it when people write and say they love my site, and it's  
> even more gratifying when someone asks me how I did it.

And how do other people implement them with any certainty of
compatibility? Much as they might like to be, Microsoft are not the
world's only providers of operating systems and web browsers.

> Wrong.  They're already implemented in the majority of web  
> sites.  Once again, what's the point of standards if no one  
> uses them?  The opposite approach would be more logical-  make  
> standards and implementations based on what's used.

What about user agents? Are your proposals implemented in exactly the same
way even in all of the major web browsers? Sometimes what is already used
is inadequate or problematic, and often not specified exactly. It is
better for standards to precede implementations, because then the
imlementors know what they should implement, and the authors know what
should be implemented. If earlier versions of HTML had been quicker in
appearing there would have been much more standardisation between
implementations, rather than a confusing mess of proprietary extensions.

> I'm not going to fight with those of you on the dark side.   
> With the force as our ally, we will win.  Besides, there's  
> more of us than of you.

Who are "us"? Who, indeed, are "you"?

> >Do you have any idea who the W3C represents?
> Yes.  A group of corporate monopolies that don't care about  
> what is actually used and wanted, are living in some sort of  
> "W3C" utopia, and actually have the money to have the say-so  
> in standards.

Have you read the member list? Or is this a new meaning of the word
"monopoly"? What evidence do you have that the member bodies have great
influence over the content of W3C recommendations? Most of them have very
little to gain from that in any case.

> I disagree, again.  Anything that thinks it'll be listened to  
> just because it proclaims itself authoritative is a couple of  
> sandwiches short of a picnic.

Indeed. The W3C does not think that.
> I doubt it.  I neither fear nor respect the "leaders of  
> industry".

So why should they respect you?

Tim Bagot
Received on Sunday, 24 October 1999 06:59:15 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:05:51 UTC