[whatwg] Web Applications Markup Language 1.0 is an XUIL?

[Ian Hickson]

I thought I'd send a quick reply to this e-mail, since it asked questions 
at a high level about the next step for WHATWG, namely some proposals for 
the Web Apps spec.

[All: If I haven't yet received to your feedback yet, don't worry. I will 
get there eventually.]

On Tue, 3 Aug 2004, Matthew Raymond wrote:
>
> I've noticed that Web Apps 1.0 has recently changed its name to Web 
> Applications Markup Language 1.0 (WAML1).

Actually that's been the name in that spec since before WHATWG was 
announced.


> Considering the contents of section "1.1. Requirements and ideas", it 
> would appear to be a partial XML User Interface Language (XUIL), with 
> perhaps the exception that it's technically SGML or tag soup when used 
> in conjunction with HTML.

It is much a "XUIL" as XHTML. More so, in fact, by design.

I don't really understand the term "XUIL", though, it's a bit like "UUIL" 
("Unicode User Interface Language") or "UILUET" ("User Interface Language 
Using English Tags") -- highlighting one minor aspect of the language to 
the exclusion of others.


> If WAML1 is indeed an XUIL, this raises a number of questions:
>
> 1) Will WAML1 be a part of "HTML5", or is it an independent technology that 
> can be used as a subset of "HTML5" in much the same way that XForms is used 
> by XHTML 2.0?

The intention is that what we have informally been calling "HTML5" is what 
is specified as HTML4 + Web Forms + Web Apps. So it is intended to be a 
part of HTML5.


> 2) While WAML1 elements require a namespace? If so, how will this be 
> handled in HTML user agents that don't support namespaces?

This will be handled identically to the way HTML4/XHTML 1.1 and Web Forms 
2.0 handle it.


> 3) Will "graceful degradation" be possible with WAML1 elements?

Again, the idea is that it will be the same as for Web Forms 2. (Whether 
this means the answer to your question is "no" or "yes" depends on whether 
you are Jim or not, I guess.)


> 4) Will WAML1 borrow heavily from existing web-based XUILs, such as XUL 
> 1.0, or will it be a complete reinvention?

It's an extension of HTML4. The primary concern is compatibility with HTML 
documents and UAs in the various ways that have been discussed on this 
list before. So it will be heavily influenced by HTML, and will of course 
be influenced by all the other languages that all the people contributing 
to Web Apps have experience with.


> If the latter, wouldn't that slow down development and adoption of 
> WAML1?

No more so than Web Forms 2's.


> If the former, wouldn't that make WAML1 largely a subset of an existing 
> XUIL, similar to my old XUL Basic/"Keymaster" concept?

It will be largely a superset of HTML4.


> I'm not saying this is the best solution, and I'm open to other 
> suggestions, but right now I think that the above track is the only one 
> we have time for. Microsoft's Longhorn is going into beta at the 
> beginning of next year, and if there isn't an effective standard for web 
> application UI by the time Longhorn hits the shelves, then this is all 
> for naught. HTML based widgets are fine, but if they lack the basic 
> functionality that exists in most native applications, developers will 
> eventually look elsewhere.

That's one of the main ideas behind WHATWG.

HTH,
-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'