Gunnar Hjalmarsson wrote:
rich@xxxxxxxxxxxxxxx wrote:
6) I would suspect, although I have little objective evidence, that
Your typical webring pages are simply ignored by search engines.
They May consider them spam. Actually, they are spam. ...
A Ringlink ring with say 50 sites generates at least 50 different
variants of the list page (not to mention the 'offset' cases, which
problem was previously addressed this way:
http://arc.ringlink.org/ringlink-open/msg02866.html).
One radical change would be to include
<meta name="robots" content="noindex,follow" />
on all list pages whose URL includes a query string.
Correction: I meant whose URL includes a query string with a specified
site ID.
To the extent the search engines adhere to that META tag, it would
result in only one variant of the list page being indexed, and nobody
could reasonably consider a Ringlink webring spam any longer.
If the changes I have in mind in this respect would be accomplished, the
*only* Ringlink generated pages that would be indexed by the search
engines would be one system wide page and two ring specific pages. Examples:
http://www.gunnar.cc/cgi-bin/ringlink/main.pl
http://www.quotationring.net/cgi-bin/list.pl?ringid=quotes
http://www.quotationring.net/cgi-bin/stats.pl?ringid=quotes
I for one get more and more inclined to believe that this would be good.
What do others think? Would it carry us *too* far?
/ Gunnar