sell aircraft with Aircraftbargains.com
aircraft for sale
sell aircraft
aircraft for sale

Advanced Search
New Listings
Forums
Dealer Login
Services
Contact
Home

corner

corner

LIST AIRCRAFT BY:

FORUMS:

ADVERTISING:

CONTACT:

SERVICES:


  AVIATION TOP 100 - www.avitop.com Avitop.com

Welcome to Aviation Forum Sign in | Join | Help
in Search  

Top 100 different from server logs

Last post 11-01-2001, 2:01 AM by anonymous2. 1 replies.
Sort Posts: Previous Next
  •  11-01-2001, 1:19 AM 1555

    Top 100 different from server logs

    *** Posted by anonymous ***
    Hi;
    I'm curious to know how you guys come up with the "Avg Per Day" count - it doesn't seem to come close to what our logs tell us on our system...
  •  11-01-2001, 2:01 AM 1556 in reply to 1555

    Re:Top 100 different from server logs

    *** Posted by run ***
    Hi,

    I apologize if any of the following information is trivial to you, but I don't know your background, so here we go:

    Server logs are normally different from the statistics you see on the Aviation Top 100 list, and there are many reasons for this.

    The Aviation Top 100 list counts (average unique users/day) so if a user comes back to your site several times during the same day (On our server a day follows Eastern Standard Time) he only counts as 1 unique visitor, but can create several hits.
    We use cookies to determine a unique user. Some of our competitors use IP addresses. We think cookies are a more accurate measure because people will often be assigned a different IP address every time they log on to the internet, so they would then count as a new user even if they have already visited the site one time. Different users could also be assigned the same IP address if they log on different times during the day, so all in all IP addresses are a lousy measure for unique users.
    The downside to using cookies is that some people have cookie-fobia and have cookies disabled.

    When talking about server logs there are also different kind of hits. Some statistics programs will use IP address and some use cookies to determine unique users. Then you have page hits which is pretty much the amount of pages served and finally server hits which is every little files that is pulled from the server.
    A page like http://www.avitop.com/pictures/galf16.htm creates around 50 server hits just to load one page.

    If, considering the above, there is still a big difference between your logs and the Aviation Top 100 list, the problem is probably one of the following:

    Problem: Some of your pages don't display the Aviation Top 100 code.

    Solution: Put the logo on all pages or accept the difference in statistics.

    Problem: The Aviation Top 100 code doesn't load before the user clicks on to a different page.

    Solution: Several solutions to this one:
    1. Place the code/logo high on your pages. Then it will load quickly and be more accurate.
    2. Optimize your pages. A webpage with pictures, banners etc shouldn't be much more than 50 Kb in size. Not everybody has cable.
    3. The way stuff loads on your pages has a lot to do with your layout (Tables/frames etc). There are a couple of articles on the subject different places on the internet.
    Bottom line: If the code loads a visitor is counted. The Aviation Top 100 server is far from overloaded. Our dual P800 with 1 GB RAM and 2xSCSI3 drives is averaging 2-4% CPU power and 4-6% during peek hours.
View as RSS news feed in XML
Aircraft Wanted Engines and parts Avionics Employment Partnerships

Advanced Search
New Listings
Forums
Login
Services
Contact
Home



©Copyright 2004 Aircraftbargains All Rights Reserved
For more information feel free to Contact Us