1 |
On Thu, 26 Jan 2012 12:56:49 +0000, Mick wrote: |
2 |
|
3 |
> > They can track a lot more than IP addresses, your browser can provide |
4 |
> > a lot of information, not just user-agent but installed fonts, plugin |
5 |
> > information and much more. There is enough to do a damn good job of |
6 |
> > identifying you even when your IP address changes. It is certainly |
7 |
> > simple to see if you are one user or two. |
8 |
> |
9 |
> Not necessarily without making some broad assumptions. For example two |
10 |
> different users could be using the same machine and OS and browser; or |
11 |
> same user could be using same machine, but different browser; or |
12 |
> different users using different machines with same OS & browser, etc. |
13 |
|
14 |
There is actually a huge amount of information available, giving a high |
15 |
level of pseudo-uniqueness. There was a web site that showed you how |
16 |
much it could glean from even an anonymous session, but I can't remember |
17 |
where is was. Somewhere like the EFF. |
18 |
|
19 |
Of course, two people using the same browser on the same computer as the |
20 |
same user would be indistinguishable, which is as good a reason as any to |
21 |
not let anyone else use your browser. |
22 |
|
23 |
> So extrapolating the user profile from browser headers is unreliable. |
24 |
> Of course Google may only be interested in getting right most of the |
25 |
> time in which case they may use such info - although I have not found |
26 |
> any references that they actually do. |
27 |
|
28 |
Agreed on both, I was only saying that it can be done, not that it is. |
29 |
|
30 |
Not that Google's profiling of individual's information is that hot |
31 |
anyway. Last year they approached me about a job for which I am |
32 |
completely unqualified - and not just because it meant getting out of bed |
33 |
before 9am :-O |
34 |
|
35 |
|
36 |
-- |
37 |
Neil Bothwick |
38 |
|
39 |
Men who have playful kittens shouldn't sleep in the nude. |