scareduck - 1:31 pm on Feb 4, 2002 (gmt 0)

First, I wanted to say, "Thanks, Brett" for a great article. Our site has made many of these suggested changes already in the last month, and for the first time, we're seeing Googlebot crawls on hundreds of thousands of pages, as opposed to the tens of pages we were getting. The good news, if there was any, was that the relatively small number of pages we had in Google had excellent pagerank. But on the subject of virtual servers:

>any one could tell a virtual server to a dedicated server Not virtual "server" but rather virtual "host" where many domains are on the same ip address. Some spiders have historically had a hard time with http 1.1 and have not used "domain" http headers. There are also se's that run their own dns servers and are fraught with slow updates. eg: I don't trust them to get it right 100% of the time. It's not worth the risk.

Er, which ones? We never see this problem with Google, and I can't think of another spider that does. Our site does many, many cobranded versions of our main site, all operating out of the same IPs. Given IP address space is slowly evaporating, this strikes me as really bad advice from a purely ecological point of view.