Ben,
I'm not immediately convinced that it isn't it a relationship. NSFW
would formalise the fact that document A:
1) contains a link to document B
2) document A's author considers document B "not safe for work" by
their own standards
at best you could make the argument that rev="nsfw" is appropriate
within the semantics of HTML (rev is the reverse of rel). That's how
votelinks work - rev="vote-for", the rev attribute capturing the
sense that
"this document or a substantial part of it has the relationship with
the destination of this link as being a vote-for it" (yes that's
tortured)
So by analogy you might argue
rev="nsfw" "means"
"this document or a substantial part of it has the relationship with
the destination of this link as being an observation that its content
is nsfw" (but I feel that is really pushing at least two aspects of
rev, in particular the document (or substantial part of a document)
level at which it works).
(One of the misleading aspects of both rel and rev is that while they
are encoded on links, they apply to inter-document relationships))
But I certainly don't think even at a stretch you could make the
logic of the rel attribute work sensibly in the case of rel-nsfw -
the same objections as the rel="vote-for" I think apply, and then some.
Perhaps a (wildy off topic) suggestion to the WhatWG and or the W3s
new HTML WG which emerges from the discussion of rel and rev is to
consider providing attributes that enable link level assertions -
that is a mechanism for typing links themselves, perhaps in a manner
similar to rel and rev, so that this is extensible via profiles or
convention. FWIW over a decade ago I implemented a hypertext based
system that had user extensible typed links, and that feature was
widely used by the (relatively small) user base. But that's really
out of the scope of ufs eh?
happy new year to all
j
John Allsopp
style master :: css editor :: http://westciv.com/style_master
blog :: dog or higher :: http://blogs.westciv.com/dog_or_higher
Web Directions North, Vancouver Feb 6-10 :: http://
north.webdirections.org
_______________________________________________
microformats-discuss mailing list
microformats-discuss@microformats.org
http://microformats.org/mailman/listinfo/microformats-discuss
