dynamic page validating without scripts?

you know you love my stupid questions

[Don_Hoagie]

Seeing all these folk with their "XHTML" and "CSS" validation stickies at the footer of their pages, I came up with this stupid question:

Is everyone using some kind of server script to send that particular page to the validator, or does the validator allow you to use some generic JS command or something that says "send this url" so you can have your pages validate on-the-fly? Surely these sites with hundreds of pages are not submitting each one manually and copying the link into the footer.

And stupid question #2:

Since I usually don't have to look at the bottom half of the W3C validator(looked at it just now to see if they said something about on-the-fly validation)... what the heck is that gold thing in the picture at the bottom? Cinderella's slipper? Would it kill ya to change those pictures once every decade?

[kaled]

The picture is thought to be a banana - maybe.

A generic link is all that is required since the server script that validates the page uses the referrer data (i.e. it knows what page requested validation). Of course, referrer data can be switched off - it's sent by the browser.

Kaled.

[encyclo]

This is actually one of my pet peeves. ;) The most common links to the validator are using the following URI:

http://validator.w3.org/check/referer

With this the W3 server validates the page given in the

HTTP_referer

string from the browser. The advantage with this is that the publisher does not have to personalize the link for each page. The disadvantage is that the link is broken to every self-respecting webmaster who actually disables referrer-logging in the browser (easy to do in Firefox and Opera). It is also broken to the unsuspecting IE user who is unaware that their adblocking program (ie. Norton Internet Security) has removed the referrer string.

Your second option is to use a server-side scripting language (PHP, ASP) to echo the page URI and append it to a link to the validator:

http://validator.w3.org/check?uri=http://www.webmasterworld.com/

Obviously it would be a waste to enable scripting just to do this.

The third (and usually best) option is not to bother with the links at all. You could argue that they play a role on a web designer's portfolio site, but outside that very narrow niche they add nothing to the page, offer no useful information to the end user, and are nothing more than clutter.

[kaled]

My site validates (I think) but I don't bother with these silly links/graphics. I validate major updates direct from my PC and rely on a local validator and/or a keen eye for minor changes.

Kaled.

[Don_Hoagie]

great info guys... more than i expected. Nothing says code validation like a banana peel.

And yeah, I agree encyclo- I too wonder who is enamored by seeing such links- I suppose it's sensible for now though, in these "infancy" stages (at least as far as the rest of the web is concerned) of standards and CSS-heavy sites. Car manufacturers boast about such and such option (now with seatbelts!) until everyone has it and it becomes obvious... same effect here i guess. And when everyone sees those footer validation links, it can only bolster the argument for designing with standards. Ok, so i've changed my mind in mid-paragraph. I like the links now.