I use a HTTP Header Viewer on my sites. Sites that are perfectly
fine in HTTP Version 1.1 - producing a 200 message.

However, where I get totally different results when
running the same test on HTTP Version 1.0 or 0.9.

I am concerned that if a site produces a 200 message in version 1.1
like this:

HTTP/1.1 200 OK
Date: Wed, 21 Jan 2004 21:03:19 GMT
Server: Apache
Connection: close
Content-Type: text/html

But produces a 401 message in 1.0

HTTP/1.1 401 Authorization Required
Date: Wed, 21 Jan 2004 21:04:17 GMT
Server: Apache
WWW-Authenticate: Basic realm="Webteam"
Connection: close
Content-Type: text/html; charset=iso-8859-1

Or give information about the virtual hosting company when I use GET
(in Version 1.0 or 0.9) that Google and the other engines may not be
able to pick up on the pages.

These are in the logs so I assume these engines are using 1.0 as indicated.

HTTP/1.0 ia_archiver 
HTTP/1.0 FAST-WebCrawler/3.8+ 
HTTP/1.0 Mozilla/5.0+(Slurp/cat;+slurp@inktomi.com;+http://www.inktomi.com/slurp.html)
HTTP/1.1 Mozilla/4.0+compatible+ZyBorg/1.0+(wn.zyborg@looksmart.net;+http://www.WISEnutbot.com)
HTTP/1.0 Googlebot/2.1+(+http://www.googlebot.com/bot.html) 

Does anyone know if Google and/or the other engines use 1.1
exclusively or if they are seeing the 1.0 information.

Thanks, 

Terry Mickelson

HTTP/1.0 only works with websites that have a one-to-one mapping
between hostname and IP address. I believe what the engines are
using is a backward-compatibility hack, which works for both 1.0
and 1.1 servers. It's a 1.0 request with the "Host:" header coming
through in the next line. You'd be able to see it with a network
sniffer (it won't show in the logs).