[AccessD] Re: xform design tools

Gustav Brock gustav at cactus.dk
Thu Feb 19 02:37:09 CST 2004


Hi those of you interested in XForms.

I noted these comments and useful links on the present status of
XForms which might be of interest for some of you.
I'm quoting as the letter is not on-line yet.

<quote>

The SitePoint TECH TIMES #83                   Copyright (c) 2004
February 18th, 2004                                PLEASE FORWARD

EDITORIAL - - - - - - - - - - - - - - - - - - - - - - - - - - - -

THE STATE OF XFORMS

Back in the Tech Times #53 [1], I introduced XForms as the
latest candidate recommendation to come out of the W3C. I
explained the advantages it had over HTML forms, and even showed
a simple example of how XForms work. 

At the time, the media buzz surrounding XForms came from the
fact that Microsoft was building something called XDocs into its
upcoming Office 2003 software. Many attributed Microsoft's
refusal to endorse XForms to similarities that the
recommendation bore to XDocs.

Over a year later, Office 2003 has been released, XDocs is now
called InfoPath [2], XForms has become a full-fledged W3C 
recommendation [3], and the fine folks at x-port.net have just
released formsPlayer 1.0 [4], a free plug-in for Internet
Explorer that fully complies with the XForms standard.

So why aren't we seeing XForms springing up all over the Web?
After all, the most popular Web browser in the world now
supports them with a free plug-in!

Due to the stagnation of Internet Explorer [5], all the
ultra-keen Web developers who would normally jump on a
technology like this have moved to Mozilla [6] (and Firefox [7])
as their development platform of choice. To put it bluntly,
no one really cares what Internet Explorer can do now, because
there is so much that it can't do (like properly support CSS2
[8]).

So, the question is, what are the other browser makers doing
about XForms?

Mozilla is tracking requests for XForms support and volunteers
interested in working on it in bug 97806 [9]. The comments on
this bug, which dates back to the days when XForms was a working
draft, make interesting reading. There are many opinions on why
XForms may or may not be worthy of consideration for inclusion
in Mozilla, but the status quo is that it remains a relatively
low-priority feature request in need of good developers, despite
having nearly 500 votes from community members.

Apple (whose Safari browser has become a serious consideration
for developers) and Opera issued a combined statement [10] last
September in response to XForms becoming a proposed
recommendation. In it, they outlined a list of "substantial
issues" in the standard that they felt made XForms inappropriate
as a replacement for HTML forms. When pressed, they admitted
[11] that XForms was probably worthy as a platform for advanced
forms development, as long as HTML forms remained as the
mainstream choice in XHTML 2.0.

So why all this negative sentiment towards XForms? Is it really
so flawed?

Like any new technology, XForms has a rough spot or two, but the
productivity it offers to developers cannot be denied. Yet,
while XForms itself isn't all that complex, it relies on a host
of technologies that are fairly complex, and are not widely
implemented in today's browsers. Quoting from Apple and Opera's
statement,

"XForms has too many dependencies. In addition to XForms itself,
an XForms implementation needs to support XML with namespaces,
XML Schema, XPath, XML Events, DOM Events, DOM Core, CSS, a
stylesheet linking technology (e.g. the XML Stylesheet PI), and
a host language (e.g. XHTML or SVG). In particular, its
dependency on XML Schema is of great concern to us."

The Mozilla bug discussion suggests that XML Schema [12]
support is not strictly necessary for a basic implementation of
XForms, but XML Events [13] is a definite piece of the puzzle
that has yet to be built into Mozilla. Meanwhile, browsers like
Opera and Safari are even further behind the curve.

>From what I can tell, XForms has the dubious distinction of
being one of the first implementations of a number of XML
technolgies with mass appeal to Web developers. Though well
thought out, none of these technologies has had a compelling
reason to be buit into a Web browser before, and now, all of a
sudden, they would all have to be built at once in order to
support XForms.

Even with a free, fully compliant plug-in for Internet Explorer,
XForms has some serious hurdles to overcome before developers
can consider it as a serious tool for general Web development.
If you want to pitch in, I would definitely encourage you to
contribute your services to the Mozilla project, but it will
take some smart brains and a lot of work to bring XForms into
the spotlight.

[1] <http://www.sitepoint.com/newsletter/viewissue.php?id=3&issue=53>
[2] <http://office.microsoft.com/infopath/>
[3] <http://www.w3.org/TR/xforms/>
[4] <http://www.formsplayer.com/>
[5] <http://www.sitepoint.com/newsletter/viewissue.php?id=3&issue=68#5>
[6] <http://www.mozilla.org/>
[7] <http://www.mozilla.org/products/firefox/>
[8] <http://www.w3.org/TR/REC-CSS2/>
[9] <http://bugzilla.mozilla.org/show_bug.cgi?id=97806>
[10] <http://lists.w3.org/Archives/Public/www-forms-editor/2003Sep/0006.html>
[11] <http://lists.w3.org/Archives/Public/www-forms-editor/2003Sep/0017.html>
[12] <http://www.w3.org/XML/Schema>
[13] <http://www.w3.org/TR/xml-events/>

The Archives are located at:
http://www.sitepoint.com/newsletter/archives.php

</quote>

/gustav


> Strangely, just this morning I received this link from Novell:

> http://developer.novell.com/research/appnotes/2003/septembe/04/a0309046.htm#1849119

> Notice the headline and link

>   The Novell XForms Technology Preview (NXTP)

> I have, however, no idea of the cost for this tool.

> /gustav


>> Date: 2003-10-08 03:58

>> I saw both of you had posted ?'s asking about the existence of any xforms design tools at the w3c-forms site.

>> Have either of you found any?

>> I am exploring QLink and Liquid Office, both of which seem to be much more than xforms, same story with holosofx (now ibm)

>> Appreciate any help.

>> Gene McKenna




More information about the AccessD mailing list