Novell and IBM to Implement XForms in MozillaTuesday August 10th, 2004The Mozilla Foundation has announced a project to implement XForms in collaboration with Novell and IBM. The effort aims to add support for the XForms 1.0 Standard to Mozilla. XForms is the W3C's XML-based Web forms technology. It offers many more features than standard HTML forms, including support for input validation and integration with Web services. However, some critics maintain that XForms is too complicated for Web authors and difficult for vendors to implement (no major browser yet supports XForms). More information about the XForms implementation can be found at the Mozilla XForms project page and the W3C XForms site has more general information about the standard. Both Novell and IBM have developed XForms applications before (the Novell XForms Technology Preview and the IBM XML Forms Package). I don't claim to know much at all about XForms, but it sounds like they are very powerful. As for the comment on usability, If it were wrapped into an easy to use library (ie, php's HTML_QuickForms), that did not require the web developer to be fluent in XForms themselves, they could be used very effectively. Kudos to IBM and Novell for stepping up for this one. How does this fit in with the work of the WHATWG? who are working on simpler HTML-like forms which are likely to cover the most important features of xforms but in a simpler way? I'm fairly sure xforms support will be a compile time option just like mathml, svg, etc but the question is when it's ready will anyone use it, and if no one uses it should we enable it by default? Full MNG support was removed because they wanted to save space now we're in the vicious circle of no one uses it so it's not included and no one will use it because it's not supported - so I guess it's a hard decision to make XForms will be pluggable - you will be able to add it into Mozilla after its installed if you want. We're using the upcoming XTF feature to do so. ...which *might* have been worth mentioning either in the mz article or in the press release (if it could be done without sounding overly technical). It was mentioned in the Developer's Day slide show linked from one of the previous Mozillazine articles, tho, for those who happened to read that. I also notice it mentioned, albeit not prominently, on the Mozilla XForms project page. Sounds like an excellent example of working together with outside companies. (Presumably IBM/Novell developers are actually going to do most of the work, possibly by way of integrating components from their existing software.) As for the WHATWG thing - leaving aside my personal doubts at that project anyway, the press release mentions that XForms are/will be recommended for use with XHTML2. Consequently, if Mozilla is to fully support XHTML2 it would be advantageous to also have XForms support. XHTML2 is very much a 'future' standard (in case you haven't come across it already - it's not backwards-compatible and throws away all the HTML cruft in favour of much cleaner syntax), but it's still worth starting the work now. So anyway - you'd then have the WHATWG stuff to use with HTML 4, with XHTML 1 you could choose either WHATWG or XForms, and with XHTML2 you'd use XForms. If I guess correctly. --sam n/t Then get cracking on this one: http://bugzilla.mozilla.org/show_bug.cgi?id=161463 :-) ...or any significant markets that I can see. Let's be realistic here. XForms may be used in intranets, or required by a few "extranet-ish" government or industry sites -- but any sites blazing this trail are effectively requiring IE and a plugin such as formsplayer, right now. Mozilla supporting XForms or XHTML2 will not change the economics of the web, which favor backward compatibility and at most the kind of incremental, cross-browser evolution that the WHATWG is fomenting. So optional (one-click download-and-install, eventually) XForms support in Mozilla products will help us compete with IE+formsplayer in a few niches. Good news, but nothing revolutionary. /be "cross-browser evolution" I think the cross browser solution will be HTML 4.1 still for many years, little will change unless Internet Explorer settles on another standard or if its market share dramatically shrinks. For most of the big web this will just be fine. What is in my opinion interesting is the small web, the many intranets, small and large, they demand more than what HTML 4.1 is and they are in the position to introduce new technology, they don't have the constraints of the big web. Technology which gets accepted by the small web will drive the big web in future, that's how I see it. Just don't know what Technology this will be, XForms, XHTML2, XUL or others? Would be interesting to know. karl I think it's been self fulfilling prophecy to say it's a niche, since there isn't a viable target for X-Forms deployment. Yes, there are "designers" who only know, and will probably only ever know HTML. But once a browser supports it fully, you'll see an enormous amount of growth because there are a large number of real application developers targeting the browser that would give there left mouse button for a sane forms model. And with the developers comes the world. -Andrew You're probably right, but if IE supports XHTML2 (a very big if, but not impossible) then there's no reason it might not gain significant use quite quickly. Look at how eager some people are to adopt the newest W3C standards even when they have no current benefit and can't even be used properly at present (XHTML1.1, I'm looking at you). XHTML2 has significant benefits in terms of allowing much nicer code and being simpler in various ways. CSS wasn't backwards-compatible; it was a complete and radical change from the previous method of layout (sticking random attributes on things in html), much more different than XHTML2 from HTML. It's now used very widely indeed, even sometimes for layout, across many major commercial sites. The real reason for this is that eventually people woke up to the fact that CSS was a good way to go and that all browsers in serious use actually supported it quite well. Give XHTML2 browser [and server] support, let it settle for a few years, see what happens... IMO you'll likely see the same effect. On the other hand, if IE doesn't support XHTML2, then nothing's going to happen. --sam Microsoft is a real question mark. Though if XHTML2 and Xforms proofs to be a real benefit on the small web (the many intranets and non IE users) then Microsoft might sooner or later be forced to support those standards and technologies. Who knows. karl Microsoft is a real question mark. Though if XHTML2 and Xforms proofs to be a real benefit on the small web (the many intranets and non IE users) then Microsoft might sooner or later be forced to support those standards and technologies. Who knows. karl #25 Re: Re: XHTML 2 is not about to sweep the web...by Endzeittraum Friday October 29th, 2004 8:32 AM >On the other hand, if IE doesn't support XHTML2, then nothing's going to happen. I think you are wrong in this point. In my oppinion the IE will loose it's rank of the no. 1 browser in the next few years. This for many reasons each of us should know of. Even if the guys at MS fix a lot of bugs, make the IE realy more secure and getting their good reputation in this point back they will loose. As they do not support new technologies like XHTML 2 and CSS 3. As you wrote XHTML 2 got many advantages and people will want to make websites using this technology. And in my opinion it is just a matter of time until the users recognize that all the fancy and cool sites require a better browser. I think it's been self fulfilling prophecy to say it's a niche, since there isn't a viable target for X-Forms deployment. Yes, there are "designers" who only know, and will probably only ever know HTML. But once a browser supports it fully, you'll see an enormous amount of growth because there are a large number of real application developers targeting the browser that would give there left mouse button for a sane forms model. And with the developers comes the world. -Andrew I really think XForms is an awesome way to go. It'll cut down significantly on bandwidth between client and server. You can do all XForms does using client side JS already - all this does is make it more accessible (in theory). Andrew: there's no demand for XForms on the web, and even if Firefox had 5% market share by use (possible, we intend to do better), it wouldn't cause content authors to sink scarce site-owner dollars into XForms content, unless they have a zero-cost way for it to work in IE (they don't). So wasting time and footprint on XForms is the definition of shooting our own feet. Making sure Firefox kicks IE's ass for the next two years, including by shrinking its download and runtime footprints, is more like shooting the other guy's foot. Which we aim to do. /be Well, I can only speak from personal experience.. and I'm sure you're looking at it from a more informed position, but as a consultant to both government and corporate development, mostly directed towards ASP.Net/Web development, I've seen a lot of interest in firefox from the developers, but very little from end users. So guess what the developers target... even though they personally use firefox. It's not because they don't like it. But tabbed browsing and small footprint doesn't provide their end users with a significantly better experience and it doesn't cut down their own development effort (in fact supporting it when developing in ASP.Net can be quite painful). And if you try to tell corporate developers to drop their IDE and hand code XUL and javascript they'll laugh. (I know some of those issues are being addressed, I'm looking at it from where we are at today). Developers pushed IE back when 4 came out, just as much as Microsoft did by including it in their OS. Why? Because it had sane support for DHTML. It wasn't mom and pop's and static web pages that needed DHTML back then.. it was developers of applications. When applications were built around IE's DTHML support, and they only worked in IE, IE gained market share. Now if you have an advance in development capabilities as profound as from html to DHTML as you have between DHTML and XForms, and you get your developer base to start targetting it (with tool support built into their IDE.. something I'm very interested in working on personally), firefox will gain market share across the board. I wouldn't sacrifice firefox's benefits in order to include xforms support.. that's not the argument. Firefox is perfectly suited to support xforms for application development as it is by using an optional component. The value isn't in not requiring an install on the client, it's with the web application deployment model with a sane and efficient forms model. Anyway, I know I'd push it from where I sit.. and people like me who would do the same is a signficant market share boost. Knowing that it's being backed by IBM and Novell makes me feel even better.. I appreciate the work of both the core firefox development and great additions like xforms and svg. Thanks! -Andrew I totally agree--developers are the ones that need to be won over. I too was a developer that switched to programming for IE when MS released a better model for client-side scripting, and now I'm finding that Firefox has quite a few enticing features, with XForms support at the top of my list. Think about it...with the help of an XML-aware database (or other persistence system) and an XML-aware programming language (ASP.NET, Java, PHP, Perl, etc.), we could be writing one design spec for data using XSD, which would then be used in all three tiers of the application. And all three tiers could potantially validate the data, which would eliminate quite a bit of hand-written validation code. Think of how many times we have to handle data parsing in strongly-typed languages like Java or C# and then imagine if the data model in each layer was actually consistent! Complex data types (like a mailing address or even a tax form) would especially benefit. If developers start taking advantage of these opportunities, XForms will eventually become as commonplace as Javascript after IE4. So why bother with the WHATWG, if IE doesn't implement it, its shoting in our own feet working on that rather than making gecko faster. I disagree. If WHATWG can get cross browser support without a 3rd party download (you can find out more about how they plan to implement IE support using HTC's if you're curious), then there is benefit to the world in general.. And there's fairly limited functional changes necessary to support Web Forms 2 compared to XForms.. I hope to see a lot of good stuff come out of the work of the WHATWG, but how equal browser support will be the major key to it's success in my mind. Unfortunately WebForms 2 doesn't go as far as I think we need in the web application development world and I don't think it can/should do much in the way of increasing mozilla's browser share. But what do I know. -Andrew That's pretty much what governs most developers in IT shops. If it takes weeks to learn this technology they're not going to fool with it. Are there applications that allow you to design and maintain xForms? Are there tools for allowing easy integration? New technology such as this is great, but if it's out of the reach (from a time standpoint) of mainstream developers it will have a tough time getting adopted. Yes, indeed: "tempus fugit" when delving into new technologies :) Novell exteNd has since late last year had extensive XForms support for both development and deployment: The Designer includes a visual XForms and pageflow designer. The exteNd "Composer" creates web services out of legacy back-ends and wizards generate (modifiable) XForms from WSDL or databases; the XForms can be deployed as portlets in the exteNd "Director" portal product. Of course it still is early days but a number of other vendors are also working on XForms support as is OpenOffice.org: http://specs.openoffice.org/appwide/xforms/GUI_spec_part1.sxw http://specs.openoffice.org/appwide/xforms/GUI_spec_part2.sxw
I'm pretty sure that XForms will once be usable in the public internet, but it will take a long, long time until then. Even today there are a few people surfing the net with Internet Explorer version 2. and a lot other people use IE 4, 5 or 6 that don't even support standards that are more than 6 years old, like CSS 2. The ammount of people wich use a standards web browser such as Firefox, Mozilla, Netscape or Opera is still quite small and this won't be changing very soon IMO. Even if Microsoft would release a version of their browser that would support XForms, CSS3 and XHTML2.0 web developers still couldn't develop sites that use Xforms for the public, until at least 80% of the world's web surfers use an XForms supporting browser. Until XForms can be used, another 5 years could pass by. //max They showed a clip of the "first lady" giving a speech on the whole "Stem Cell Research" issue, and how she said it was wrong of the political ads to try and sway voters with relatives that have alzheimer's by saying that the cure is just around the corner.. because that's just not true.. it'll be years before we'll have a solution... then they cut back to Jon Stewart, and he said something flippant (as usual) along the lines of: "So why bother starting now?" Build the tools, build the support, build the knowledge base.. now. Don't just sit back and wait for the cure.. -Andrew |