IP addresses and personal information (was Re: [LINK] Fwd: On Line Opinion - 16 February 2007)

[rchirgwin at ozemail.com.au]

Irene Graham wrote:

>On Sat, 24 Feb 2007 07:46:53 +1100, rchirgwin at ozemail.com.au wrote:
>[...]
>  
>
>>Irene,
>>
>>Without getting into the tit-for-tat over Hitwise's techniques, there
>>are other approaches to Web stats which bother me more. 
>>    
>>
>
>There are also other approaches that bother me more.
>
>  
>
>>In particular,
>>there's the pratcise of launching an invisible redirect using Port 443 -
>> that is, while the page is loading, you get flick-passed to an SSL
>>session to the ad counting company.
>>
>>This, I presume, is to help distinguish 'unique visitors' - but how
>>does the technique work?
>>    
>>
>
>Sorry, I'm not sure whether I know about that. Are you referring to things 
>like dgmaustralia and imrworldwide? But my recollection is they run via 
>javascript in the pages, so I guess that's not what you mean.
>  
>
Irene,

I think the two are either similar or the same - here I will admit that 
I'm dancing on the edge of my technical grasp, so if I'm shot down in 
flames, I'll probably sign off with Spike Milligan's epitaph - but:

Whether it's via Javascript or something else, I have seen Websites 
launch Port 443 (ie, SSL) sessions in a redirect when you visit. Roughly 
it went like this:

User > visits site foobar.com
Foobar.com > redirects user to hitcounter site on Port 443
Hitcounter site > redirects user back to Foobar.com
User sees Foobar.com load in the browser.

Because no page loads, the redirect isn't visible to the user unless:
a) you're running the sort of firewall that has a live monitor (Tiny 
Personal Firewall was nice for this); or
b) the browser settings pop up a "you are about to enter / leave a 
secure site" each time SSL is invoked.

And yes, imrworldwide.com was one of the sites I observed using this; 
Telstra is or used to be a customer.

Now, this doesn't preclude the Javascript; the JS may simply be 
activating the Port 443 communication instead of the browser; here I'm 
beyond my expertise!

And finally, because my observation of this is out of date (I have 
posted to Link but it was some time back), it may be that this is an 
obsolete hit-count technique. I would be interested to know if it 
persists...

RC

>Irene
>
>
>
>
>_______________________________________________
>Link mailing list
>Link at mailman.anu.edu.au
>http://mailman.anu.edu.au/mailman/listinfo/link
>
>  
>