Only 4.13% of the Web Is Standards-Compliant
Death Metal writes "Browser maker Opera has published the early results of an ongoing study that aims to provide insight into the structure of Internet content. To conduct this research project, Opera created the Metadata Analysis and Mining Application (MAMA), a tool that crawls the web and indexes the markup and scripting data from approximately 3.5 million pages."
It is very simple http://validator.w3.org/
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
W3C's validation tools
Normally I'd go on my own rant but I'm feeling lazy today and recently I read a good article at A List Apart that sums it up. As for the W3C, I like this list they compile:
W3C's Pros & Cons
Pros:
Cons:
You should read that article, it's pretty spot on for this subject.
My work here is dung.
For example, xhtml-strict does not include support for "target" attributes in links. What kind of idiotic decision was that?
A very good decision, there are two main uses for the "target" attribute:
/Mikael
Greylisting is to SMTP as NAT is to IPv4
MAMA didn't validate against a single document type. They validated against the document type that each individual document claimed to be. So all the ancient HTML 2.0 pages out there will correctly be identified as valid in they are, in fact, valid HTML 2.0.
Bogtha Bogtha Bogtha
There aren't degrees of validity. A document is either valid or it isn't. You can't be "more strict" when validating something, if a tool offers you an option like that, then it is doing something other than validating, it's probably linting as well. There's at least one widely-used "validator" that doesn't actually validate at all.
Yes you do. If you didn't close them, your pages wouldn't work in any browser. What you mean is that you don't explicitly close your paragraph and line break elements. And you don't have to. The closing tags for <p> elements are optional and the <br> element type is empty. Those are not errors.
Bogtha Bogtha Bogtha
Are there degrees of strictness?
Yes. HTML 4.01 and XHTML 1.0 each have two DTDs: a "transitional" DTD that allows presentational elements and a "strict" one that disallows them. The trouble is that a couple structural elements and attributes got removed by mistake in the strict DTDs along with the presentational ones, most notably the value attribute of the li element. For this and other reasons, most valid HTML that I've found has used a transitional DTD.
It's up to me as a user to choose where a url opens, especially since we are all using the tabbed paradigm now.
User agents currently do not allow the user to submit a form into a new window or tab. This is the nearly nine-year-old bug 17754 on bugzilla.mozilla.org with 99 votes.
If you go to the source of the research, you will for example notice that the last time a similar study was done (in June 2006), only 2.58% of the tested pages validated completely. A 1.5% increase might not seem to be all that much, but it's definitely indicative that we're on the right way. (And of course, perfect validation is never the final goal in itself, but merely an easy first step for people en route to writing better, semantically-meaningful, universally accessible websites.)
Not really a lot of point to it, though -- savvy users will simply middle-click on the link if they want it in a new tab/window. If they don't, that generally implies they want it right where it is, and your attempt to open a new tab/window is going to be annoying.
But hey, at least using a target for that is better than linking to a javascript: URL. A lot of sites are even worse -- they add an onClick event, and they set the link href to #, or to javascript:void(), meaning that middle-clicking on it inevitably does something unexpected.
My preferred method (if I ever need to force a new window) is to use a plain old link, and progressively enhance it with Javascript to open a new window. That way, if people middle-click, it does exactly what they want.
Don't thank God, thank a doctor!
Does using "blink" make my code non-standard?
Yes, because blink is not defined as conforming in any standard. However, it is possible to make a page containing blink (or any other element or attribute you like) pass validation by providing a custom DTD or an internal subset.
But note that the claim that "4.13% of the Web Is Standards-compliant" isn't quite accurate. The study only used the W3C markup validator, which is only able to detect a subset of the machine checkable conformance criteria. It's trivial to create a non-standards compliant page that passes validation.
By reading this signature, you hereby agree with the content of the above comment.
Here's Opera's article, with actual numbers:
http://dev.opera.com/articles/view/mama-markup-validation-report/
Standards do help, but the problem is the leading browser doesn't follow the standards too well.
My primary browser is Firefox, so I do 90% of the development in there. The last 10% is spent adjusting little bits and bobs for IE6/7. It doesn't matter that my pages validate, and that I'm pretty damned good at making my spaghetti code spew proper XHTML - IE is still a retard.
Where I work, we guarantee only one browser: IE6. Anything else costs extra, because of the testing involved. If someone wants us to guarantee their site will run perfectly in Opera or any other browser, they have to cover our time spent testing and massaging the markup. Sometimes that can amount to another 2-3 days of billable time on a big site/app.
I don't use Konqueror at all for web browsing. Frankly I wish it would just embed the Gecko engine and be done with it.
-Billco, Fnarg.com
There is also a Web Developer plug-in available at the FireFox Add-ons site.
http://www.accountkiller.com/removal-requested
XMLHttpRequest is now a standard since everyone decided to "ignore standards" and use it anyway.
http://en.wikipedia.org/wiki/XMLHttpRequest
"The World Wide Web Consortium published a Working Draft specification for the XMLHttpRequest object's API on 15 April 2008."