Re: Invalid Tiki Wiki page

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Joseph Reagle <reagle@w3.org> wrote:

>wonder why people aren't linking to:
>http://validator.w3.org/check/referer
>as the validator page directs?

Possibly because "Personal Firewall" products have this nasty habit of
chopping off the Referer header under the assumption that mangling the basic
functionality of HTTP is doing users a favor... :-(


>And I wonder if there be an easy way for us to check those people using
>the icon from our site, to ensure that if they use the icon, their site
>is actually valid?

Well, I've taken the position -- for a number of reasons, and not claiming
this is the only valid position -- that use of the badge constitutes an
assertion by the _publisher_ of the page that it conforms to the relevant
Recommendation; and _not_ an assertion (by the publisher or the W3C) that the
W3C has certified the page to be conformant.

IOW, the W3C logo on the badge merely indicates that the Recommendation the
publisher asserts conformance with happens to be published by the W3C. As far
as I can see, any other stance on this would place the W3C under obligation to
provide some form of certification program that I do not think would be in
anyones interest for it to do.

But, yes, it certainly can be tested automatically to a reasonable extent if
you want to go down that road.


>Or better yet -- those this would be very processor intensive -- we
>could even return a "broken" icon to those sites that aren't valid
>when they request the icon!

This has been on the todo list for ages and the main reason it hasn't been
implemented yet is due to that very concern. As I understand it, the Valid Foo
badges are the most requested resource on the w3.org servers (with the
possible exception of the home page) and "generating" each of them on the fly
would a) require significant overhead in itself, b) increase the load on
validator.w3.org since it would have to serve all the badges instead of
handing that job off to the www.w3.org round-robin setup, and c) increase the
overhead by an order of magnitude by requiring that each including page be
validated every time it is loaded by a browser (instead of every time someone
clicks on the badge).

Implementing this is a non-issue; it's a five minute job for a first cut (IIRC
from the last time I looked at it) and certainly could be included in the next
major revision for a neglible amount of extra development effort (as other
changes in the next major revision happen to make this even easier to
implement).

- -- 
By definition there is _no_way_ any problem can be my fault. Any problems
you think you can find in my code are in your imagination. If you continue
with such derranged imaginings then I may be forced to perform corrective
brain surgery... with an axe!                            -- Stephen Harris

-----BEGIN PGP SIGNATURE-----
Version: PGP SDK 3.0.2

iQA/AwUBPtgp/qPyPrIkdfXsEQJTzQCePbq8VFWNJXno0Fb8bqBOX0w4STgAoN/8
ffiiLJgSoUwRVU9vyGGgCap0
=eDzM
-----END PGP SIGNATURE-----

Received on Saturday, 31 May 2003 00:05:30 UTC