W3C home > Mailing lists > Public > www-validator@w3.org > June 2009

Re: proper use of validation icons

From: Michael A. Peters <mpeters@mac.com>
Date: Sun, 07 Jun 2009 10:12:59 -0700
Message-id: <4A2BF51B.6040806@mac.com>
To: "Jukka K. Korpela" <jkorpela@cs.tut.fi>
Cc: www-validator@w3.org
Jukka K. Korpela wrote:
 > Michael A. Peters wrote:
 >
 >>> Despite what the validator says, the document is in fact not "valid
 >>> XHTML 1.1", to the extent that such an expression makes sense. The
 >>> document is valid (markup conforms to the DTD specified) but it does
 >>> not use a DTD specified in XHTML 1.1 specifications.
 >>
 >> It does, just has some additions added to it - the X part of XHTML
 >
 > No, you either use the specific DTD or you don't. If you don't, you
 > might still call your markup XHTML (or HTML or XML or whatever you
 > like), but it isn't XHTML 1.1. The "X" in "XHTML" is just hype, though
 > technically it means that the syntax is based on XML, the strongly
 > simplified version of SGML. HTML is just as extensible as XHTML is.

I do use the specific DTD.

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" 
"http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd" [
<!ATTLIST input autocomplete (on | off) #IMPLIED>
]>
<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en">

And no - HTML is not just as extensible.
To add an attribute you have to define a new DTD. You can't do the above 
with html. You can't even do it with xhtml sent with html mime type 
(well, you can, but then you get a ]> display artifact).

Also, you can't (legally) define namespaces, or include other xml 
specifications - such as MathML. XHTML allows you to do that stuff.

 >
 >> I like pages with the icons on web pages because when there is a
 >> problem rendering, it allows me to quickly and easily see if the
 >> problem is a bug.
 >
 > How many of your _visitors_ are interested in debugging your pages?

Not very many.
I know with my browsing behavior, I don't even bother with a page that 
has trouble where the author hasn't given me an indication he cares 
about web standards. I just go elsewhere. When a web master indicates he 
cares, I am willing to send him notification that his page is not compliant.

People who don't care at all just don't care at all, no damage done.
I suppose you could argue a few kilobytes of wasted bandwidth sending 
the icon, but that's why browsers cache images.


 > And
 > do you mean that the icon that _claims_ the markup to be valid is useful
 > for _checking_ whether it is? There's something odd in that idea, but
 > the fact is that many of the "Valid HTML" icons make a false claim,
 > often an intentionally false claim (that is, a lie). That's one of the
 > reasons why they should be used: people have no grounds for trusting on
 > them.

They don't guarantee valid markup.
All my pages are authored via DOMDocument and I actually had a function 
written that validated the page server side - logging the error if it 
didn't validate, displaying the icon only if it did, but it was too slow.

They do however indicate the web master cares about web standards, even 
if that wasn't the original purpose.

Fish in the market labeled "wild caught" is very often actually farm 
bred (they can tell by chemical analysis). Labels are abused, that 
doesn't make them worthless.

W3C doesn't go after people who incorrectly use it, so it's not a 
guarantee the page is compliant. However, few people who really don't 
care are going to intentionally put it on their page. So it does serve 
the purpose of notifying users you care about standards compliance.

 >
 >> So displaying the badge serves a purpose there, I'm sure I'm not the
 >> only one with that philosophy about reporting errors.
 >
 > What about the 99 % of surfers who have no intention of reporting markup
 > errors?

It doesn't cause them any harm. Other than a few kilobytes of bandwidth.

 >
 > And I thought the idea was that an author who adds a "Valid HTML" icon
 > has himself taken care of validation.

Yes. However, dynamic content can break from something as simple as 
updating a module or library.

 > If the icon is meant to say "I
 > want to use valid markup, please help me with that!", then it should
 > really say that and not something rather opposite.

What the icons display is the version of the DTD and a check mark. I 
think the alt tag W3C suggests may say "Valid", I'll have to check, but 
the icon itself indicates the version of the document and has a check on 
it, the icon itself nowhere uses the word "valid". When you click it, 
the check takes place.
Received on Sunday, 7 June 2009 17:13:44 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:35 GMT