On Sun, Apr 03, 2005 at 01:58:56PM -0400, Joshua Slive wrote: >... > BrowserMatch "Mozilla/2" nokeepalive > BrowserMatch "MSIE 4\.0b2;" nokeepalive downgrade-1.0 force-response-1.0 > BrowserMatch "RealPlayer 4\.0" force-response-1.0 > BrowserMatch "Java/1\.0" force-response-1.0 > BrowserMatch "JDK/1\.0" force-response-1.0
Over a two-week period on blogspot, the hit percentages for above: Mozilla/2: 0.20% MSIE 4.0b2: a single hit RealPlayer: no hits Java/1.0: no hits JDK/1.0: 0.000076% I would recommend tossing ALL of them. The middle three for obvious reasons. The JDK because it is really, really insignificant. The Mozilla/2 is a bit tricker. The problem is that the match string is *way* too loose. It is matching *way* more browsers than I bet it was intended to match (I found 201 different User-Agent strings). For example, consider this User-Agent: Mozilla/2.0 (compatible; PlanetWeb/1.011 Golden; SEGA Saturn; TV; Do we really know if the keepalives are broken on that client? It certainly doesn't seem to bear any resemblance to the Mozilla that I'm guessing that BrowserMatch was trying to detect. Or maybe this one: Mozilla/2.0 (compatible; MSIE 3.0; Update a; AK; Windows 95) via proxy gateway CERN-HTTPD/3.0 libwww/2.17 Given that it appears to be proxied, then I'm betting the keepalive is totally fine. Cheers, -g -- Greg Stein, http://www.lyra.org/
