W3C home > Mailing lists > Public > public-html@w3.org > April 2007

Re: About the Web Forms 2 proposal

From: Maciej Stachowiak <mjs@apple.com>
Date: Sat, 28 Apr 2007 22:58:01 -0700
Message-Id: <FE13E285-D984-4086-AE10-41D54FB29FE2@apple.com>
Cc: Dave Raggett <dsr@w3.org>, Matthew Raymond <mattraymond@earthlink.net>, public-html@w3.org, Sebastian Schnitzenbaumer <sebastian@dreamlab.net>
To: John Boyer <boyerj@ca.ibm.com>


On Apr 28, 2007, at 7:42 PM, John Boyer wrote:

>
> Hi Maciej,
>
> The analogy was exactly the right one.  Please try again.  The  
> hybrid approach in XForms is another technology for solving the  
> same problems more easily than using an imperative script approach  
> alone, just like the script approach is more desirable than writing  
> it in machine language.
>
> You've looked around and found that you don't see the need for the  
> better solution because, you say, no one is using the new solutions  
> right now in their existing solutions.

I don't think that's an accurate description of my remarks. My main  
point is: it appears that current forms use cases on the web would  
not benefit from a declarative expression language. As a secondary  
point, even current forms that might be able to benefit from more  
declarative features do not tend to use the ones available.

>  The existing solutions use just plain javascript because that's  
> all they have right now.

That doesn't seem to be the case. FormsPlayer and the XForms  
Transitional prototype are two alternatives. I'd like to see some  
success for the existing plugin- or library-based solutions before we  
decided that the technology is proven enough to go into the core of  
HTML.

>  We're trying to talk about a better way to do things that makes it  
> possible to do current things more easily and to do even harder  
> things that would be quite hard to do with existing challenges.   
> Just like using electricity and a light bulb is better than using a  
> torch, even for reading something as antiquated as reading a book.

I recognize you're trying to talk about that. but I don't think your  
proposed solution (declarative expression language) addresses the  
problems people are actually having on the web. To continue your  
analogy, a light bulb is better than a torch, unless your problem was  
actually to set things on fire, not to provide light. In which case  
it is over-engineered and unhelpful.

> And the use cases you listed here (http://esw.w3.org/topic/HTML/ 
> FormsUseCases) are pretty trivial.

They may sound trivial to you, but these are the kinds of forms that  
are authored every day and used by people all the time. And many of  
them don't work as well for users and developers as the best desktop  
application technology. Anything we can do to improve those kinds of  
experiences is extremely high leverage. So for you to be dismissive  
of these use cases makes me think you may be out of touch with how  
forms are actually used on the web.

> And yes people in many verticals (government, financial, insurance,  
> healthcare, etc.) do in fact want to have a more usable forms  
> technology, and their requirements really are more complex than the  
> use cases you are looking at.

It sounds like a specialized single-purpose technology like XForms is  
just thing thing for those verticals. But it doesn't sound like those  
same features are essential in HTML, the general-purpose language of  
the web. HTML should include basic building blocks, and specific  
solutions for common use cases. It shouldn't add special features for  
relatively rare use cases, and ones that mostly won't come up on the  
public web. Separate technologies are a better way to deliver special- 
purpose features.

Regards,
Maciej
Received on Sunday, 29 April 2007 05:58:22 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:15:53 GMT