> > "forcing" cookies
>
> I'm not sure that's ALL that likely. But the following are very likely:
> 1) People refuse cookies. Unless you won't let them into your site (which
> is also arguably skewing your statistics) you will not count them
> properly.
> 2) People clear out their cookie cache. (jar??)
People who refuse cookies won't be counted, so the question about
*minimum* estimates wouldn't be affected by this. You only have to be
concerned when someone gets counted twice- which can happen if they remove
the cookies from their computer.
Depending upon the epiration set for the cookies though, I doubt it
would cause any big overestimation. It'd make up for some of the "lost"
people who didn't accept cookies. Gotta love statistics! ;)
However, it *is* pretty easy to write a script that hits a server,
grabs a cookie, hits something else, and clears out the cookie before
repeating the process. It's a 10 line perl script with LWP. With a little
more work you could forge/change your IP address each time and make it
harder to track.
The question didn't specify what purpose this tracking served. If it
involves money, people *will* do things to skew stats. There was a
customer of ours who runs a fairly popular site that was told by a *major*
ISP/host that they [wc]ould hook up a "load generator" to the network and
drive traffic levels up to make the stats look good for the sponsors. The
favor was declined, but I doubt this sort of thing is uncommon. Probably
not an everyday act either though.
-=Jim=-
------------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe analog-help" in the main BODY OF THE MESSAGE.
List archived at http://www.mail-archive.com/[email protected]/
------------------------------------------------------------------------
