[whatwg] Web Forms 2.0 - what does it extend , definition of same,

[Jim Ley]

On Tue, 11 Jan 2005 01:15:14 +1300, Matthew Thomas <mpt at myrealbox.com> wrote:
> On 10 Jan, 2005, at 9:36 PM, Jim Ley wrote:
> .... Which are two separate issues, because Gmail could be using a
> <title> etc despite being an application.
> 
> However, Gmail requires that you log in, so search engine indexing
> isn't an issue;

I don't belive search engine indexing to be relevant to
web-applications, web-applications simply aren't indexed, we don't
want search engines indexing our email, or our aggregrated RSS or ... 
The IMDB or Amazons product pages are not web-applications, they're
brochure pages, their a very different use cases IMO.

As I seem to have a very different idea on what Web Applications are,
could someone please help me out with the definitions being used by
the WHAT-WG?

> And what would be the benefit of that? Most authors don't care about
> semantics.

One of the main arguments I repeatedly see for the WHAT-WG stuff, is
more semantics in the mark-up, yet you're now arguing that authors
don't use it anyway, what's the point of having it if the authors
aren't going to use it?  The cost of developing this stuff takes
implementation time away from more useful things.  The reasons GMail
doesn't support minority browsers, is the weakness of their scripting
engines, maybe the time spent doing Web-Forms was invested in those,
we'd see more web-applications using it.  GMail is a good example of a
Web-Application supported in many user agents, almost all are written
for IE alone.

> > They could however carry web document semantics to aid non visual
> > understanding, the fact they don't isn't something that needs more
> > specs to help.
> 
> How are those issues related? The rest of the Web needn't grind to a
> halt while we wait another decade for Google to fix their markup.

GMail is an example, it's not IME a bad example, it's a typical
example of the web-application, so I think it's a very useful thing to
look at when looking at what features are relevant when building
specifications for future ones.   Now I realise use cases etc. for the
various features of Web Forms 2 and Web Applications are lost in the
pre-public portion of the WG, and no-one seems willing to re-visit
them, but I still like to think of the development in terms of what is
useful to web-application developers.

> It is quite impressive for you to have snipped my answer to your
> question and then to have asked it.

Unfortunately, that's not an answer to who the WE are, in terms of
web-applications, Amazon and IMDB are certainly not relevant, they're
not web-applications, they publish information, that's a very
different use case, and isn't a use case that Web Forms 2.0 or Web
Applications is addressing.  They're all in the interaction sphere
which is irrelevant to search engine ranking.  Equally the arguments
over search engine ranking and semantic mark up really help, IMDB puts
the title of the movie in a STRONG element, and Amazon in a B element,
not a heading as would seem appropriate.

So still we don't have examples of Web-Applications that use the
document semantics of HTML, even if we extend the definition of web
application to include amazon and imdb.  In any case using scripting
for behaviour (form validation etc.) does not prevent search engines
finding content, very few people use javascript to include content.

> > You're also missing one of the elements, non-arbitrary XML - For
> > example Bill McCoy's RSS reader won't be consuming arbitrary XML.
> 
> That's not relevant. We're talking about what applications are written
> in, not about what they consume. 

Are we?  I thought we'd been discussing Bill McCoy's excellent points
on how web-applications are put together, one of which is that
consuming appropriate XML semantics in your web-application is a good
thing - that's something that XForms and XBL and others allow, it's
not something that the work here is contributing too.

Jim.