On 2011-07-07 15:05, Jordan Clark wrote:
To Ian Hickson / whom this may concern,
I second what DriedFruit says in his email, "ISSUE-118 CP 3, rel="start"
and friends, rant":
http://lists.w3.org/Archives/Public/public-html-comments/2011Jul/0000.html
I too, was surprised to find that my previously valid web page suddenly
had four errors according to the W3C validator, all of them relating to
values found in REL attribute of A or LINK elements – two of which are
microformats: rel-home and rel-licence)!
I put it down to tinkering with the validator; after all, HTML5 support
is currently experimental, but after reading this list, it looks as if
this is going to become an official "feature"!
I also believe that the range of values for the META element's
HTTP-EQUIV and NAME are much too restrictive. Why can't you just allow
arbitrary values for the attributes mentioned above? To not do so will
not only break backwards-compatibility with *existing* technologies,
it's also restrictive for *future* technologies too.
I hope that you take the time to reconsider this.
...
With respect to link relations, you should follow
<http://www.w3.org/TR/2011/WD-html5-20110525/links.html#other-link-types> --
in theory you can make these relations valid again; and the process for
doing so should be tested.
I *do* agree that it's not clear whether it makes sense to make
unregistered link relations invalid, though. But IMHO the whole HTML5
approach to validity is questionable anyway, so I have kind of given up
on this topic for now.
Best regards, Julian
