Print

[marian]

Could someone explain how the Coppermine view count works? It has always been down on the display image count in the server logs and it is getting worse. It now is consistently about 50% of the server figure.
I realise that things like this might have something to do with it http://images.google.be/imgres?imgurl=http://www.bymnews.com/photos/albums/userpics/10006/thumb_HJ6A3167.jpg&imgrefurl=http://www.bymnews.com/photos/displayimage.php?album=random&cat=&pos=-1040&h=100&w=150&sz=5&tbnid=ZGcfD47CCTsJ:&tbnh=60&tbnw=90&hl=nl&start=7&prev=/images?q=louis+vuiton+cup&svnum=10&hl=nl&lr=&sa=N but I dont think such glitches could account for such a high discrepancy. Sunday's are always down on views but 14 hours into today the server log is showing 781 and the Coppermine figure is 344 and I'm sure there haven't been that many referrals from the Google Image bot, though I cant be certain since the software I'm using doesn't recognise the image spider as being a search engine.

[Nibbler]

It doesn't count repeat views from the same user in a certain browsing period. If you want it to count all views I already posted how to do this.

[marian]

It doesn't count repeat views from the same user in a certain browsing period. If you want it to count all views I already posted how to do this.

Do you think that's the cause? I randomly look in at the "live" views and I haven't spotted any evidence that suggests people look at pictures more than once.

[Nibbler]

It also ignores hits from the 'lasthits' meta album, and from users who are not accepting cookies.

[marian]

It also ignores hits from the 'lasthits' meta album, and from users who are not accepting cookies.

I wasn't sure what you meant by lasthits meta album, but a search revealed it. I dont see how anyone would be accessing that, I certainly wouldn't know how!
By the way, I noticed in one of your posts about most recent hits, you suggested that awstats might do things coppermine cant. Actually awstats results are not worth a light. We have three freebie analysers on our server, awstats, analog and webalyzer and none are accurate. Analog is such a mess as to be unusable and awstats records less than 50% of visitors, webalyzer is best of the three recording about 2/3rds. These comparisons are with Webstats, Mach5Analyzer and WebExpert all of which gaive EXACTLY the same visitor count when we tested them. We settled for WebExpertLite, which gives all the info we need and is free!