Trouble Brewing at the W3C?

An anonymous reader writes "A breakaway faction of the World Wide Web consortium (W3C) called WHAT-WG, or the Web Hypertext Application Technology Working Group--which includes Apple, the Mozilla Foundation and Opera--is threatening to revolt over electronic forms standards. WHAT-WG has announced its intention to submit the draft to the W3C, posing the potentially awkward possibility of the consortium advocating two conflicting avenues for Web forms. The fate of a standard could also determine whether the order form could be accessed in any standards-compliant Web browser, or if it would be available only to users of a particular operating system--an outcome that has browser makers and others worried about the role of Microsoft."

Re:Is it really a Battle of the Browsers?

[R.Mo_Robert]

Ummmmm ... is the combined market share of ALL browsers outside of "Apple (aka Safari), Mozilla, and Opera" and IE even close to 1%?

You haven't kept up with the stats lately, have you?

WebSideStory this January has Firefox alone at 5%, and IE is about 90%. Nobody I know is reporting anything below 1% for the "other guys". Besides Firefox, Opera was at 2.1%, Netscape (maybe including the Moailla suite as well) was at 2.6%, and I don't know about Safari or others.

OneStat even has IE below 90% at 88.9%, and that was last November.

So, yes, the combined non-MS marketshare is certainly more than 1%, which you could have easily found out with a little research.

Storm in a teacup?

[Hortensia Patel]

Sound to me as if someone either missed the cluetrain, was having a slow news day and decided to invent a crisis, or swallowed some Microsoft FUD without checking his facts.

From the Web Forms 2.0 draft spec [whatwg.org]:

"This specification is in no way aimed at replacing XForms 1.0 [XForms], nor is it a subset of XForms 1.0.

XForms 1.0 is well suited for describing business logic and data constraints. Web Forms 2.0 aims to simplify the task of transforming XForms 1.0 systems into documents that can be rendered on HTML Web browsers that do not support XForms."

The Web Forms proposal is hugely important precisely because it can be implemented for IE using a "standard library" of client-side script. It won't be quite as nice as native implementations, but it'll work. It's the first evolutionary proposal I've seen that actually makes allowance for the festering carcass of IE holding everybody else back.

Re:w3c sucks

[W3bbo]

a) Using CSS instead of the tag actually uses less code if you need that "font style" a lot in a page or website.

b) XHTML is based of XML, therefore all XML rules (including code termination) must be met. If you've got a problem with this, go back to SGML based HTML4.01 which allows this

c) In CSS you can have a

as wide as its content, use: "width: 0; overflow: visible;"

Full height-sidebars? "height: 100%;"

d) Attribute="value"s have to be completed in order to comply with XML spec, as I said earlier, SGML-based HTML4.01 is more flexible....And you forgot to close your element properly ;)

The W3C Standards exist for a reason and many are devised by people who, lets face it, are waaaay smarter than both you and I. If you've got a problem with this, then join one of the W3C Working-Group Mailing lists and ask them yourself.

XUL? XAML?? Flash???

[idlake]

XUL and XAML are general markup languages for GUIs. And Flash is a complete runtime.
The notion that XUL and XAML are substitutes for a forms standard makes about as much sense as saying that a C compiler is a replacement for a web browser: just add a little bit of code yourself. I guess we should count our blessings that at least they aren't proposing to use Java.

XForms is specifically for forms: things you fill in and submit. XForms also has facilities for off-line filling and mailing of forms. We need a standard like that.

Having said that, I find neither XForms nor Web Forms 2.0 particulary persuasive. XForms suffers from second system effect: there is just too much of it. And Web Forms 2.0 seems like a mess; reliance on JavaScript is a no-no.

Thanks, but not thanks: everybody should go back to the drawing board. Maybe in another few years, they'll come back with something reasonable.

Re:What's the difference??

[0x20]

IE doesn't have over 90% of the market. Currently the number is somewhere around 60-80% and falling. Still too much, but the situation is improving.

Source 1: http://www.w3schools.com/browsers/browsers_stats.a sp/ [w3schools.com]
Source 2: http://www.upsdell.com/BrowserNews/stat.htm/ [upsdell.com]

Re:w3c sucks

[Anonymous Coward]

Don't use that <font>... you can do it in CSS and it only takes 3 times as much code!

I call bollocks on that. Separation of style and content is very important for machine readability, being able to change the style of your site without trawling through every single page and editing every tag and defining the font-family ONCE for an entire site in ONE file makes for a lot less lines than in every single page on your site.

All tags must be terminated explicitly. Because it happens so often that you need to nest a <td> directly within another <td>, with nothing between them.

Again, crapola. All tags must be terminated explicitly so that HTML4 is XHTML1 compliant - allowing browsers to lose a lot of bloat that is otherwise needed to correct mistakes made by numpties - and the more stupid the page designer, the more bloated the browser needs to be to compensate.

CSS can do any styling you want! Unless you want a centered div that's only as wide as a content... or a full-height sidebar... then you're just screwed. (hint: use tables instead)

Hint, use properly nested DIVs and a little imagination and CSS can do a lot more elegant styling than huge bloated tables filled with 1 pixel shim.gifs

col and colgroup: Bug 4510

[tepples]

Firefox ignores backgrounds on <col> and <colgroup> even though the standard explicitly allows them.

Bug 4510 happens because a cell inheriting style from both a parent <tr> and a pseudoparent <col> would inherit conflicting information. Should it follow the style of the <col> or of the <tr>? CSS specifies no way to resolve multiple inheritance. What is the Right Way, with documentation?

What trouble?

[mnemonic_]

A breakaway faction of the World Wide Web consortium (W3C) called WHAT-WG

According to the WHAT-WG page [whatwg.org], "Many of the members of this working group are active supporters and members of the W3C..." So it seems they themselves do not see WHAT-WG as a "breakaway faction."

And if they actually rejected the W3C, why are they planning to submit their proposal through the standard W3C pipeline? Why not attempt to bypass or ignore it? If WHAT-WG are against the W3C, they would not be planning to cooperate with them.

It looks like this WHAT-WG is just another group submitting another proposal to the W3C. Yes, that proposal conflicts with an existing W3C one. But that doesn't indicate anything about turmoil in the W3C. It's just another potential standard that happens to have the same goal as another. Competition of standards in the W3C is nothing new.

Re:It's ABOUT FSCHK'N TIME

[Anonymous Coward]

What the fuck are you on about?

A well designed website will work fine in Lynx, and look FUCKING great in Moz, and even better, still work for disabled people.

Why? CSS, data seperation. HTML == Define data, give it names (with id/class). CSS==looks.

Re:col and colgroup: Bug 4510

[Anonymous Coward]

Backgrounds aren't inherited anyway, they're just layered over each other. The <tr>'s background takes precedence.

XForms is backwards-compatible, including IE

[mdubinko]

I happen to have recently surveyed [xml.com] XForms engines, and at least two of them under development run entirely within the client, in the style of gmail, Google Maps, Flickr, etc.

Modern browsers are up to this, it just takes a (one-time) engineering effort, treating JavaScript as a full programming language.

Of course, if browsers like Mozilla natively support [mozilla.org] XForms, all the better. -m

Re:Give me a break

[Anonymous Coward]

Yawn. Yet another anti-W3C troll. I swear you guys are getting more and more obvious. Take this blatant lie, for example:

And they apparently won't even consider taking any of Microsoft's adaptations to the standards into consideration

Then how do you explain display: inline-block; [w3.org], which started out as an Internet Explorer extension to CSS? If their stance on Microsoft is so clear, then surely you'll be able to point me to a statement a W3C member has made saying such a thing?

So really, why should Microsoft give any credibility to these standards and the people behind them

Microsoft helped develop both HTML and CSS (check the credits in all the specifications), and have given a lot of money to the W3C over the years. You are quite simply ignorant.

Re:What's the difference??

[robmv]

See XForms working on Explorer [novell.com]

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Wait a minute...

[Anonymous Coward]

scripting is a terrible way to handle form input. It just doesn't scale and you have the form in one location and the code scripting in another place, so if you change something you have two separate locations to update everything in.

Have you looked into how Remote XUL works? A server sends a bunch of stuff to a client (xml, css, and javascript). The client uses the script you send to handle the workload appropriate to the client end. Using a programming language for this task gives you a lot more flexibility than pre-adjudicating what clients should and should not do.

Scaleable? Sure, you can pass off as much of the load as appropriate to the client, leaving your server to do only what it needs to. Why do you say not scaleable? The toughest part would be doing database replication on the server end if you need to load balance across multiple machines, but that has nothing to do with what we're talking about. Two separate locations to update? No, all xml, css, and javascript is maintained centrally. Application updates happen for all clients whenever you change what you send them.

P.S. - Microsoft was involved in the XForms standardization process (or should I say lack of process/progress). This effort has been mischaracterized elsewhere as being the brainchild of a small cadre of browserless anti-Microsoft folks. See http://www.w3.org/TR/xforms/sliceI.html to see a full list of everyone involved. It is interesting to note, though, that the finished product doesn't list Microsoft as a participant, but as a former participant. I would love to know more about Microsoft's contribution, or foot-dragging lack thereof, to the successful completion of this standard.

WEB FORMS 2.0 DOES NOT REQUIRE JAVASCRIPT!!

[DeanEdwards22]

The idea that Web Forms 2.0 requires JavaScript is a fallacy.

JavaScript may be used to provide legacy support in the client (browser). However, Web Forms 2.0 is intended for implementation by browser manufacturers. Because it is based on existing HTML forms technology it is potentially implemented quite quickly. No Web Forms 2.0 application should ever assume that the browser supports WF2. There should always be proper validation for legacy browsers. This is being a good web-developer anyway.

Re:You know the saying -

[TheRaven64]

Not to mention the fact that it completely bypasses the browser's forms assistance features. My local cinema uses Flash for online booking forms (doubly obnoxious since they charge around 50p/ticket for the privilege of booking online). Every other site I visit with a form of this nature gets my name, address, etc autocompleted from my personal vCard. This one doesn't. It is spectacularly irritating for the end user.

Re:What's the point?

[l0b0]

For XForms you already have a Flash viewer [packts.net] and a server-side [sourceforge.net] XForms to HTML + JavaScript implementation. There are lots [w3.org] more, but those are the only two I've tested which implement a good deal of the standard and are good-looking..

Re:Browser Statistics?!

[MortimerGraves]

Averaging might not be all that useful as different sites have very different audiences. IE usage at Slashdot will be very different from that at cnn.com and different again from Sourceforge, etc.

Trending may be more useful; looking at stats for the same sites, the same user population segment, over time

January stats from TheCounter.com, which looks like it might offer reasonable stats for a range of fairly "general" sites, shows IE6 + IE5.x = 88% This is down from IE usage mid-2004 at the same site (93%), a usage level that had been fairly consistent for the 6 months previous.

Looking at the trending for the past three years on W3Schools, which its more technical user population, shows a drop in IE usage from 86.8% in Jan 2002, through two years of relatively little change to 84.1% in Jan 2004 and then a steady drop over 2004 until 69.7% in Jan 2005.

Sourceforge shows a drop from 74.8% in Jan 2004 to 58% in Jan 2005.

IE usage is falling at different rates (and from different heights) at different sites, but the overall trend is downward. There may have been a time when IE had 95-99% use on some sites, but that appears to be in the past.