On 18 Dec 98, at 19:26, Susan Duncan wrote: > Check your site to see if *anywhere* in your html files you put your full > path and not just a relative URL. What I have seen on our hitlist reports > is that the domain we are running it on ends up being the top referrer > because in some instances we are using full URL's. We exclude our own > viewing by eliminating our IP addresses, so it's not us. I would think that the vast majority of websites are their own top referral. I do not see how filtering the url of the site whose logs you are anlayzing will affect the user sessions ( as I understand them as defined by Webtrends -- and I have not changed the default definition). I see no way that a user session could ever begin with your site as the referrer. NO REFERRER is nearly always, always tops. Yet here is the response and my reply to techsupport at webtrends (note his patience in defining the terms for me): On 18 Dec 98, at 17:14, [EMAIL PROTECTED] wrote: > Just in case: No referrer means that people either typed directly your url > or they had it bookmarked. I am afraid thre is misinformation in the KB. > The user sesisons will be affected if youfilter by referrer. I don't think there is anyhing "wrong" in the KB. I think it perhaps is not fully explained. The KB does indeed state that if you filter your url they you will influence your user sessions. I am looking for WHY. Your explanation does not make sense. > BTW: A referrer is the URL of the HTML page which a visitor clicks to > access a page on your site. Your own domain or IP address will usually > appear in the Top Referring Sites and Top Referring URLs because only the > initial hit to your site will log a URL outside of your site as the > referrer. Some web servers will log the gif file sof your web site as the > initial hit to your web site and not the html page. This gif file will > have your own site as the referrer. Thus we see this as the referrer. The referrer value is passed in the headers sent by the browser. They are not always necessarily accurate and some spiders will spoof referrers. I am quite familiar with the referrer from a cgi point of view as my job is webmaster/programmer. Please tell me the name of just one web server that would somehow record the request for a url before the request of the html file in which the reference to the gif exists. That is very interesting. I am using Apache/1.2.5 PHP/FI-2.0 I have been trying hard to think of a way that using my url as a filter could influence the number of user sessions but I cannot. Perhaps it is the 30 minute rule. User with IP #X stays on the site for more than 30 minutes. His next request after 30 minutes is with the same IP number but the referrer is my site, naturally. Well, if I am filtering my url then that would not be considered the start of a new session. But if I am NOT filtering my url then perhaps it could be considered the start of a new session. I would appreciate an EXACT why filtering the url whose logs we are talking about would influence user sessions. If as you said about the odd gif before html recording, and my version of Apache is one of those, then enough said. But if it isn't then ... gee... what do I know? Thanks, Peter -------- The enduring goodness of the American military character over the past two centuries does not automatically derive from our nation's nutritional habits or from a good job benefits package. This character must be developed and supported, or it will die. -By Daniel J. Rabil http://www.wbap.com/marknote/note111298.html ____________________________________________________________________ -------------------------------------------------------------------- Join The NEW Web Consultants Association FORUMS and CHAT: Register Today at: http://just4u.com/forums/ Web Consultants Web Site : http://just4u.com/webconsultants Give the Gift of Life This Year... Just4U Stop Smoking Support forum - helping smokers for over three years-tell a friend: http://just4u.com/forums/ To get 500 Banner Ads for FREE go to http://www.linkbuddies.com/start.go?id=111261 ---------------------------------------------------------------------
