From: Bjorn S. on
On Fri, 29 Jan 2010 17:32:55 +0000, za kAT wrote:

> On Fri, 29 Jan 2010 14:40:05 GMT, Franklin wrote:
>
>>> I tried it again and got this...an additional .01 bits...wonder what
>>> changed since I last tested it a few minutes ago.:
>>
>> You think your browser is indestructible? It wears out a little, every
>> time you use it. :-)
>
> I die a little, every time I read a Bottoms post :(

People died by the droves every time he dropped off a load of dope or a
ache of guns.
--
All the best,
Bjorn S.
- I only post via <news.individual.net> or <news.broadpark.no>!
From: Wheel on
UnsteadyKen wrote:
> Wheel said...
>
>> Firefox: unique - 191,279 > 17.55
>> Iron: unique - 191,755 > 17.55
>> IE: unique - 191,828 > 17.55
>> Avant: unique - 192,592 > 17.56
>> Opera: one in 207 > 7.69
>>
> Chrome: -Unique 203,975 > 17.64
> Safari: -Unique 204,135 > 17.64
> IE8: -Unique 204,235 > 1n7.64
> SlimBrowser: -Unique 204,313 > 17.64
> Off by One: -Unique 68,288 > 16.06

Thanks for giving some more numbers, but it's making less
and less sense the more I try to make sense of it.

I take it you meant:

Off by One: -One in 68,288 > 16.06
From: Spamblk on
=?UTF-8?B?wrtRwqs=?= <boxcars(a)gmx.net> wrote in
news:20100128200352.17e480ed(a)bellgrove.remarqs.net:

<snip...>
> I don't trust the numbers. It says my fingerprint matches 1 in
> 82,517. (Yesterday it said 1 in 59 K or so.) But my fingerprint is
> much more likely unique. My U-A string alone has "Gecko/20100108" (coz
> that's the day I last compiled it) and a vendor string I made up to
> replace "Gentoo".

It detected my Browser as a Linux browser, so for a second time I tried
http://panopticlick.eff.org after I created the Useragent string in Firefox
about:config "general.useragent.override" to "Mozilla/4.0 (compatible;
MSIE 6.0; Windows NT 5.1" as described at http://dotdoh.com/?p=80 And I was
then advised that only 1/247K browsers have this value?

Given all the other data a browser puts in http requests, however, the EFF
might still have a valid point it seems.
From: »Q« on
In <news:20100128200352.17e480ed(a)bellgrove.remarqs.net>,
»Q« <boxcars(a)gmx.net> wrote:

> In <news:hjt9sn$c5s$1(a)news.eternal-september.org>,
> Craig <netburgher(a)REMOVEgmail.com> wrote:
>
> > I tried out panopticlick <https://panopticlick.eff.org/> and was
> > presented with:
> >
> > > Your browser fingerprint appears to be unique among the 160,945
> > > tested so far.
>
> I don't trust the numbers. It says my fingerprint matches 1 in
> 82,517. (Yesterday it said 1 in 59 K or so.) But my fingerprint is
> much more likely unique. My U-A string alone has
> "Gecko/20100108" (coz that's the day I last compiled it) and a vendor
> string I made up to replace "Gentoo".

Now that they have more data, my browser matches less of it.

Firefox 1: 257,152 (unique)

These are with out-of-the-box settings, AFAICR:

Dillo 1: 51,450
links -g 1: 257,283 (unique)
links 1: 257,318 (unique)
elinks 1: 257,346 (unique)
lynx 1: 128,682
Opera 1: 257,392 (unique)
SeaMonkey 1: 257,449 (unique)
Conkeror 1: 257,504 (unique)
Konqueror 1: 257,530 (unique)
w3m 1: 51,512

At this point I became too bored to continue.

To understand the importance of the number of bits takes some
information theory,
<https://www.eff.org/deeplinks/2010/01/primer-information-theory-and-privacy>.
Part of the explanation which is in non-jargon says,

So for instance, if we know someone's birthday, and we know their ZIP
code is 40203, we have 8.51 + 23.81 = 32.32 bits; that's almost, but
perhaps not quite, enough to know who they are: there might be a
couple of people who share those characteristics. Add in their
gender, that's 33.32 bits, and we can probably say exactly who the
person is.

(Well, less jargon than the log calculations, at least. ;)

ISTM much depends on who's looking for the person. If the searcher is
someone who has a database with the identities, birthdates, and genders
of everyone in the zip (or who is willing and able to get such a
database, *then* the person can probably be uniquely identified.
Similarly, how much to worry about your browser's fingerprint depends
a lot on who might be looking for you and what resources they have.

Enough of that. Now everybody go freak out about
<http://whattheinternetknowsaboutyou.com/>.
From: UnsteadyKen on
Wheel said...

> Thanks for giving some more numbers, but it's making less
> and less sense the more I try to make sense of it.

Same here.

> I take it you meant:
>
> Off by One: -One in 68,288 > 16.06

Yep.

Just tried Off by One again and got this...

Within our dataset of several hundred thousand visitors, only one in
64,596 browsers have the same fingerprint as yours.

Currently, we estimate that your browser has a fingerprint that conveys
15.98 bits of identifying information.

It seems to be getting better, or worse?
Perhaps a bit fell off Off by One putting the numbers off by one or 2
bits.


--
Ken O'Meara
http://www.btinternet.com/~unsteadyken/