Re: rel="nofollow" attribute

> really [stupid|shocking|untrue] but I don't even want to link to it'? Or 
> sites that refuse to link to competitors?)

I think that refusing to link to competitors goes deeper and would
require rel=competorsite as a specific relationship, not that authors
would make the distinction (users might find it useful to build lists
of competitors but authors might not want that!).  Naming competitors
is rare in print media, except in knocking copy.

I think "nofollow" itself was established from the use in meta elements
to inhibit all links on the page.  This was a bad choice even there, but
is consistent with authors preference for verbs over adjectives and nouns,
which is basically the same as the preference for presentational over
structural (e.g. verb "display as red", versus adjective "very important" [A]).
In rel, which has tighter semantics than meta, it is even worse.

If I understand the issue properly, whilst the business problem for
Google may be link pollution, the actual problem, from a search engine
user point of view, includes all of the third party text in the blog 
page, so what is really needed is some way of marking off whole blocks
as unsafe.  At the moment, the only real way of doing that is with
objects with HTML content, but support for these is poor.

If bquote were used properly and only properly, it would probably
meet the immediate search engine requirements, although it wouldn't
distinguish between vetted and un-vetted quotes, which would be relevant
for controlling indexing of words in the material.

[A] Even though class="red" appears to be an adjective it is really a 
command, in the author's mind.

Received on Friday, 21 January 2005 08:33:44 UTC