|
Access - Page view |
|
Count HTML pages only
This is far more accurate than the hits as it's so easy to add 'pixel' images into your HTML file to increase hits (many HTML editors use 'pixel' images
to build complex layout)
|
|
Requests - Hits |
|
Count all file requests
Just to give an idea about your stats. The ration Hits/Access give you
the number of files loaded by the user when viewing one page.
|
|
Visitor |
|
Unique IP address
Most stats package use 'visitors' to count the number of different
IP addresses. I don't want to use this because this number is not
very accurate. Is a spider a visitor ? Different visitors can use the
same IP if they use a proxy or share the same internet connexion ...
and on the other side, the same user can visit your site many times
if he use dynamic IP. So many visitors --> one IP and same visitor -->
different IP !
Session is more accurate and can make the difference but it's a little
bit more hard to compute !
|
|
Hosts - Site |
|
IP address
This number can give you a rough estimation of your visitors. But
don't forget this is not very accurate due to proxy or ISP works.
Local hosts are machines inside your domain name, external hosts come
from outside. It allows you to get stats also within your domain.
|
|
Traffic |
|
The bandwith used
It could be a very useful data to count the number of Mb people are
downloading from your site. Detect the files which take most of your
bandwith.
|
|
Proxy |
|
Check requested from a proxy
When people use a proxy, most requests come from this proxy which check
if the file requested is uptodate. If yes, the proxy send the file to the
user and no request is made from your website (except about asking if
the page is still uptodate). If not, another request is made to get the
file requested and this request is counted as a hits. So this number is the proxy
requests about data check.
|
|
Not found |
|
Request for a missing file
This is the number of requests for a file which have been deleted or
misspelled from your website. try to get this number as low as possible.
Check the error stats and the bad links section to rectify.
|
|
Forbidden |
|
Request for a protected file
Some area of your site may be protected against unsollicited people. When
your server answer 'you are not allowed to get this file', this number is increased.
|
|
Scripts |
|
Requests about CGI
If you are using some dynamic pages on your site, you may use CGI. When
URL is based on a POST method, script arguments could be splited and
some stats can be computed.
|
|
Internal |
|
Single level directory
Only statistics for a specific directory is printed. If you want to
count also sub-directories, you should use the 'Tree' section.
|
|
Tree |
|
Multi level directories
Statistics including all sub-directories are counted. if you have a root
directory for each language (/uk/, /fr/...), you could easily see
which language is used the most.
|
|
Session |
|
Visitors number
There is no session on Internet but it can computed with some basic rules.
People don't stay too long on the same page to read it (30 minutes is the
limit) and robot don't scan your website during the whole day. Then session
can be extracted from the data. This is not the number of hosts but it
should be close (many people can get the same IP over time (ISP) or the
same IP can be shared by many people (proxy))
|
|
Robot |
|
Web spider
Some site like Altavista, Inktomi, Google use their own robot to
scan the web to feed their database search engine. The first request
from a robot is for a file called 'robot.txt' which give the rules
(permission) to access your website. If no file is found, the robot
assume it can scan everything. So when this file is requested, the
session is assumed to be a robot one. People which use tools to
download a whole website use also robot session.
|