Google?

[Rtep]

....."Google Adsense
Authorized third parties such as Google Adsense place cookies as described in our Cookies Policy. We use this information to serve you certain advertising content.

Google Analytics
Linux Mint uses Google Analytics for pages which URL starts with linuxmint.com/start/. The information collected is anonymized and is used by Linux Mint to have a rough idea of how successful each particular release is."

Where I found it? In LINUX policy. [...] Google is the company who spying on everybody (except me) who using anything Google!
People, wake up! Use VPN service!

[karlchen]

<mod>
Hello, dear folks.
As can be noticed some posts have been removed, because 2 of them only held personal attacks and counter attacks.
Here is your second chance of discussing about Google in a civilized manner.
Even though this is "Open Chat", please, stick to verifiable facts. Do not resort to personal attacks, because you have got no arguments to support your point.
</mod>

[Hoser Rob]

Why single out Google? It's hardly just them.

I'd suggest reading Dawn of the New Everything by Jaron Lanier (actually all his books I've read are pretty interesting). It's mostly about virtual reality, which he describes as the only thing you can raise capital for in Silicon Valley without promising to spy on everybody.

In fact, I know people who are always complaining about Google spying, yet the first thing they do when they fire up the machine is log into Facebook and then they leave it logged in in a tab until they shut down the machine. Who scrapes you worse than FB?

I don't like it much either but the whole business model of the Web is just wacky. No one is willing to pay for Web content for anything other than streaming media. Yet everyone expects these people to spend a billion $$$ or more to build and maintain the server farms needed to host the web apps and not charge for them. For example, if you install Google Earth, you're just installing a front end. The software that's doing all the work is on one of those multi billlion $$$ server farms.

If you have any ideas for these companies on how to pay for those server farms without charging users or collecting data from them and selling it, I'm all ears.

[t42]

Oh boy, Google knows where I live, and more of that ...I'm paying for bread in our corner shop using Google Pay. Should I switch to the Garmin Smart Watch payments until its too late?

[MikeNovember]

....."Google Adsense
Authorized third parties such as Google Adsense place cookies as described in our Cookies Policy. We use this information to serve you certain advertising content.

Google Analytics
Linux Mint uses Google Analytics for pages which URL starts with linuxmint.com/start/. The information collected is anonymized and is used by Linux Mint to have a rough idea of how successful each particular release is."

Where I found it? In LINUX policy. [...] Google is the company who spying on everybody (except me) who using anything Google!
People, wake up! Use VPN service!

Hi,

It is very simple to avoid ads, and to avoid to be tracked online, by Google or any other. The use of "/etc/hosts" as an IP address filter, and of some browser extensions arrive to this result.

I have explained it here (in a tutorial about chromium flatpak): viewtopic.php?f=42&t=358979.

You can compare the results before / after at EFF "cover your tracks": https://coveryourtracks.eff.org/

Note that the results concerning tracking are good, while the results about browser fingerprinting are ridiculously wrong: any browser will have some 18 bits of identifying information, and will be unique among 250,000, including your smartphone browser (Safari running on my iPhone 6S is unique! ). This is simply because Cover your tracks is not so much used, and so its statistical sample is too small.

Regards,

MN

[gittiest personITW]

Bare in mind that Google and 'others' in general may still only have enough computing power to tailor information to groups (those who buy bread in their corner shop with googlepay for instance) rather than looking at you as an individual. Obviously the more of those 'groups' you are in, the more tailored the advertising will seem to be.
Soon (if it hasn't already happened) it will be more personal but I really don't think that any organisation at the moment has the computing power to look after 3 billion people on an individual basis. I may be wrong - this is just opinion/hope.

For those with smart televisions who suddenly get adverts about pink kettles straight after talking about pink kettles may or may not be correct in the supposition that their telly is looking out for them. Or it could just be conincidence. Ask your telly.

[Pierre]

it's very doubtful, that it will ever get that personal,
and it would be impossible to track everything that everyone does, at any given point.

as already mentioned, Google et al, can only generalise their "target"
and most cases, their advertising is not very relevant .. despite what they are telling their advertisers.

[Rtep]

This year (2021) over 500 million Google accounts all over the world was hacked! Not my! I ended everything Google about 20 years ago., never used Facebook, twitter. and other so called social media. My life with computers, phones and tablets is much easier. I wonder ....nobody answer my original question: why Linux collecting anything about me using Google? If is for "getting better" listen to people criticism! Criticism will make what are you doing better, not collecting anything from people. That's called spying! Period.

[t42]

I ended everything Google about 20 years ago.

Gmail (and Gmail account) appeared 17 years ago as an invitation only service. And cookies were first used by Google in 2007.

[t42]

This year (2021) over 500 million Google accounts all over the world was hacked!

Nothing like that it was 533 million Facebook users' phone numbers and personal data have been leaked onlinehttps://www.businessinsider.com/stolen- ... ine-2021-4

[xenopeek]

why Linux collecting anything about me using Google?

Assuming with "Linux" you mean Linux Mint, the parts you quoted earlier from the privacy & cookies policy https://linuxmint.com/privacy.php already sum up the why, and the policy explains in more detail, but it boils down to that anonymized data is collected on the start page with Google Analytics to get a rough idea of how popular each release is, which gives information like this https://blog.linuxmint.com/?p=3953, and Google Adsense can be used for advertisements on the websites to help fund the project. To the best of my knowledge Google Adsense is currently only used on the forums. The privacy & cookies policy explains how you can opt out of personalized ads. You can instead also use an ad blocker which I guess you will already be using. If that doesn't answer what you find unclear about the policy you can address your concerns to the data privacy contact person.

[Aztaroth]

People, wake up! Use VPN service!

That may just shift your data to another company which, if successful, will be bought by Facebook, Amazon or Google itself.
The Tor browser offers better privacy and I found also very easy to install on LM a Parrot/Kali tool called Anonsurf which redirects all your traffic through the Tor network (even if you're using Firefox).

[Hoser Rob]

....nobody answer my original question: why Linux collecting anything about me using Google? ...

Did you really think that using Linux would stop this? It's a function of the browser, not the OS. You're still subject to spying and browser hacks.

[Aztaroth]

Note that the results concerning tracking are good, while the results about browser fingerprinting are ridiculously wrong.

No, they aren't. You're just mixing privacy and anonymity.
I'll use a simple example to illustrate :
Imagine you have a nice home with a garden surrounded by a fence. We can say you've got a good privacy because you only let in or out who you want. But as your neighbours, when you're in the garden, can see if you're taking a nap, reading a book, cutting the lawn or playing with the kids, your anonymity is about zero.

Trackers and ads are related to privacy and your add-ons surely give you a good one.
Fingerprinting is very bad because of ... these very adds that protect your privacy.

Illustrating again :
Looking for John Doe in a crowd is a very difficult task, but if you know that John Doe has installed the blue-eye add-on, the grey hair add-on, the collar beard add-on, speaks German, wears jeans, ... you'll find him within seconds. Transpose with LM20.2, Firefox 95.0, uBlockOrigin, CookieAutoDelete, HTTPS Everywhere, Decentraleyes... The most add-ons you pile to protect your privacy, the less anonymous you become.

The Tor Project / Browser uses exactly the opposite path : they try to make you a John Doe and discourage you to use any add-on and to make any change to the browser's configuration to keep you such a John Doe.

I tested many Firefox config with add-ons, VPN's, through Tor tunnel with anonsuf ... and the best I could achieve is "nearly-unique" configuration instead of "unique". With Tor I'm always not-unique if I keep using English instead of my country language. The reason is the one given by Hoser Rob just above : Firefox by design leaks some fingerprinting info, while the TOR Project set their modified Firefox to transmit the same data whoever the user is (we are 8 billions to live in TimeZone 0, to have a 1000x800x24 screen, to have unknown system fonts...)

Also, and because that post of mine may also be related to the OP, there's no access denial from the LM Forum or official site when using TOR as it can occur with other sites, but I just did it as a test because it's not really sensible to try to be anonymous on a site where you log in with an alias and a password, meaning you agree to be unique on that site.

[MikeNovember]

Note that the results concerning tracking are good, while the results about browser fingerprinting are ridiculously wrong.

No, they aren't. You're just mixing privacy and anonymity.
I'll use a simple example to illustrate :
[...]

Hi,

Here are test results justifying my "Note that the results (of Cover Your tracks) concerning tracking are good, while the results about browser fingerprinting are ridiculously wrong".

Tests are done with EFF Cover Your Tracks https://coveryourtracks.eff.org/

Here are test results, using EFF Cover Your Tracks https://coveryourtracks.eff.org/

Latest TorBrowser 11.0.2 on Linux Mint, as downloaded, NoScript enabled :
Some protection against tracking
Blocking tracking ads: Partial protection
Blocking invisible trackers: Partial protection
15.21 bits of identification
nearly-unique fingerprint, 1 among 37,939 tested

Latest TorBrowser 11.0.2 on Linux Mint, NoScript disabled:
Some protection against tracking
Blocking tracking ads: Partial protection
Blocking invisible trackers: Partial protection
15.47 bits of identification
nearly-unique fingerprint, 1 among 43,355 tested

AnonFirefox 95 on Linux Mint, with uBlock Origin:
Strong protection against tracking
Blocking tracking ads: Yes
Blocking invisible trackers: Yes
16.79 bits of identification
Unique fingerprint, 1 among 113,374 tested

Chromium 96 on Linux Mint 20.2 Mate, with uBlock Origin (fully customized, with a total of 19 extensions):
Strong protection against tracking
Blocking tracking ads: Yes
Blocking invisible trackers: Yes
17.79 bits of identification
Unique fingerprint, 1 among 226,742 tested

Safari on iPhone 6S, iOS 14.8.1, with AdGuard, Privacy and Firefox Focus content blockers (8 blockers):
Strong protection against tracking
Blocking tracking ads: Yes
Blocking invisible trackers: Yes
17.79 bits of identification
Unique fingerprint, 1 among 226,792 tested[/u]

Safari on iPhone 6S, iOS 14.8.1, without any content blocker:
Some protection against tracking
Blocking tracking ads: Partial protection
Blocking invisible trackers: Partial protection
16.79 bits of identification
nearly-unique fingerprint, 1 among 113,406 tested

Comments:

- Comparison between TorBrowser, with or without Noscript, shows little impact of NoScript on trackability. (The use of NoScript is not realistic, most of websites need at least javascript). Latest revision of TorBrowser, 11.0.2, includes only two extensions, HTTPS Everywhere and NoScript; TorBrowser is no longer protected against tracking by PrivacyBadger, which explains the poor result in resistance to tracking.

- Comparison between TorBrowser and AnonFirefox shows that, speaking of browser fingerprinting, TorBrowser has a lower fingerprint than AnonFirefox (15.21 or 15.47 identification bits, instead of 16.79). However, TorBrowser has a partial resistance to tracking, while AnonFirefox fully resists to tracking (AnonFirefox is protected against tracking by PrivacyBadger and Easy Privacy list in uBlockOrigin).

- With 16.79 bits of identification and unique among 113,374 AnonFirefox is said having a unique fingerprint, while Safari on iPhone without content blocker is said having an nearly-unique fingerprint, with the same results as AnonFirefox: 16.79 bits of identification and unique among 113,406 --> tests global appreciations are not coherent with test results.

- Adding a protection against tracking, ads, malware etc. using a browser extension (uBlock Origin) or content blockers (Adguard, Privacy, Firefox Focus for Safari on iPhone) increases the number of bits of identification by some 1 to 1.22 bits, but gives a strong protection against tracking. To avoid tracking is mandatory: as soon as there is a tracking cookie in your browser you are identified with 100% probability; it is preferable to avoid tracking than to have an increased fingerprint, since fingerprint increase just slightly increases identification probability, below 100%.

- Safari on my iPhone 6S is said being unique among 113,406 tested. This shows that Cover Your Tracks has a small comparison basis, not enough to give reliable browser fingerprinting results (my iPhone 6S has nothing unique).

So:
- use Cover Your Tracks to see if your browser resists to tracking,
- use Cover Your Tracks for its advice to reduce browser fingerprinting,
- don't give any confidence to the figures (number of bits of identification, browser unique among...) resulting from Cover Your Tracks tests.

Regards,

MN

PS: if you want to avoid to be tracked by browser fingerprinting you should use the most commonly used browser: Google Chrome with English language, without any customization, used within Windows 10 with English language

[xenopeek]

Tor network isn't very trustworthy. It's probably fine to bypass geo restrictions of legal websites you visit as guest, but don't use it for shady websites or websites you log in on.

Since at least 2017, a mysterious threat actor has run thousands of malicious servers in entry, middle, and exit positions of the Tor network in what a security researcher has described as an attempt to deanonymize Tor users.

Tracked as KAX17, the threat actor ran at its peak more than 900 malicious servers part of the Tor network [at least 10% of the Tor network servers]

https://therecord.media/a-mysterious-th ... or-relays/

Since January 2020, a mysterious threat actor has been adding servers to the Tor network in order to perform SSL stripping attacks on users accessing cryptocurrency-related sites through the Tor Browser.

The group has been so prodigious and persistent in their attacks, that by May 2020, they ran a quarter of all Tor exit relays

https://www.zdnet.com/article/a-mysteri ... g-attacks/

[Aztaroth]

Tor network isn't very trustworthy. It's probably fine to bypass geo restrictions of legal websites you visit as guest, but don't use it for shady websites or websites you log in on.

Yes, of course. I may have added that my use of TOR is for anonymity but not for illegal purpose. An example, if you search info on diabetes or cancer because one of your folks has such a disease, a de-anonymized browser can leak that info and you may personally be harmed the day you want to take a loan or an insurance policy because you will be assumed as having those diseases.

The group has been so prodigious and persistent in their attacks, that by May 2020, they ran a quarter of all Tor exit relays

It was only for profit through Bitcoins but you're right to point out a possible security breach. However there's a feature in TOR forcing connexion through TOR bridges (obfs4 being the best), which can be considered as certified entry relays. I'll have to make some homework to see if this certification is or can be expandable to the 2 remaining relays.
Anyway, thanks for pointing this to me.

[Aztaroth]

Latest TorBrowser on Linux Mint, as downloaded:
Partial protection against tracking
Blocking tracking ads: Partial protection
Blocking invisible trackers: Partial protection
15.47 bits of identification
Partially unique fingerprint, 1 among 43,355 tested

Here's what my Tor Browser's CYT returns (never changed anything to the intial installation as recommended by the Tor Project). So anyone may have the same if he doesn't play with his system.

Blocking tracking ads? Yes
Blocking invisible trackers? Yes
Protecting you from fingerprinting? Your browser has a non-unique fingerprint

Your Results
Within our dataset of several hundred thousand visitors tested in the past 45 days, only one in 1889.93 browsers have the same fingerprint as yours.
Currently, we estimate that your browser has a fingerprint that conveys 10.88 bits of identifying information.

We may eventually guess that all those experiments you're making on that machine of yours and posting here and there definitively compromised your anonymity and privacy. This casts some doubts if not disqualifies your comments.

- With 16.79 bits of identification, AnonFirefox is said having a unique fingerprint, while with the same number of bits Safari on iPhone without content blocker is said having a partially unique fingerprint --> test results are not coherent.

They are coherent, but you're mistakenly assuming the number of bits being directly correlated to the fingerprint evaluation.
It isn't because all the bits don't have the same charge. An example to illustrate :
Java enabled uses one bit (true/false). So that bit has 100% of the fingerprinting charge of an item.
User agent uses 127 bits so each bit has 100/127 (about 0,8) % of the fingerprint charge of an item.

About the algorithm used to get those bits, it is a very complex formula based on a field of the information theory called the Shannon entropy, which is serious enough to be used in genome sequencing and data lossless compression.
For the curious ones, here's the report, as a PDF, on how the formula has been implemented :
https://coveryourtracks.eff.org/static/ ... ueness.pdf

And here lies the problem. Most of us, if not passionate by the subject, could have a really good life without knowing of (that's why I didn't mention it earlier). But exception has to be taken for those who want to recommend and find those recommendations based on false premises because they just don't understand how things work. This is particularly relevant because you're recommending a mix of yours (Anonsurf exists, Firefox exists, but not that brew you're trying to promote, at least in an approved version).
Maybe I should follow the path of others and leave you in your world (in fact, I already did - just wanna prevent other users from falling into traps you set, even not voluntarily as may be the case of your own computer with the CYT report). So, I will close that debate even if you're free to answer because it's been all said from my side.
However, Cover your Tracks is reliable if you have confidence in EFF's algorithm or are able to understand the results.

[MikeNovember]

Latest TorBrowser on Linux Mint, as downloaded:
Partial protection against tracking
Blocking tracking ads: Partial protection
Blocking invisible trackers: Partial protection
15.47 bits of identification
Partially unique fingerprint, 1 among 43,355 tested

Here's what my Tor Browser's CYT returns (never changed anything to the intial installation as recommended by the Tor Project). So anyone may have the same if he doesn't play with his system.

Blocking tracking ads? Yes
Blocking invisible trackers? Yes
Protecting you from fingerprinting? Your browser has a non-unique fingerprint
Your Results
Within our dataset of several hundred thousand visitors tested in the past 45 days, only one in 1889.93 browsers have the same fingerprint as yours.
Currently, we estimate that your browser has a fingerprint that conveys 10.88 bits of identifying information.

Hi,
I have not played on my system, and not played with TorBrowser.
Comparing your results and mine are very interesting. I have used latest TorBrowser revision, 11.0.2; in this revision there are only two extensions, HTTPS Everywhere and NoScript; PrivacyBadger is no longer here to protect TorBrowser users against tracking.
I have made tests with and without NoScript; the use of NoScript slightly decreases the number of identifications bits, but its use is not realistic: most of websites need at least javascript to function.

We may eventually guess that all those experiments you're making on that machine of yours and posting here and there definitively compromised your anonymity and privacy. This casts some doubts if not disqualifies your comments.

Are you really serious ? Cover Your Tracks is a test website, not an attack one! And each browser (TorBrowser, AnonFirefox, Chromium) has its own profile, not located in the same place. And after each test (or each time I close a browser), I delete all the cache, cookies etc. And, of course, tests done with iPhone are not done on my Linux computer.

- With 16.79 bits of identification, AnonFirefox is said having a unique fingerprint, while with the same number of bits Safari on iPhone without content blocker is said having a partially unique fingerprint --> test results are not coherent.

They are coherent, but you're mistakenly assuming the number of bits being directly correlated to the fingerprint evaluation.
It isn't because all the bits don't have the same charge. An example to illustrate :
Java enabled uses one bit (true/false). So that bit has 100% of the fingerprinting charge of an item.
User agent uses 127 bits so each bit has 100/127 (about 0,8) % of the fingerprint charge of an item.

Read carefully what I posted: both AnonFirefox and iPhone had the same number of bits of identification, and the same unicity.
If I understand what you write, identification bits are on an arbitrary scale, while "unique among" is a relative value, calculated from the number of tests done. Test results with iPhone show that this number of tests done at Cover Your Tracks is not enough to be confident in the figure (it is not EFF responsibility, just the fact that CYT is not used enough).
Note that Java has been disabled for all browsers for years and should not taken into account.
And AnonFirefox user agent is a pseudo Windows one (I used a setting in "about:config", I will add this in my tuto viewtopic.php?f=42&t=358889).[done]

About the algorithm used to get those bits, it is a very complex formula based on a field of the information theory called the Shannon entropy, which is serious enough to be used in genome sequencing and data lossless compression.
For the curious ones, here's the report, as a PDF, on how the formula has been implemented :
https://coveryourtracks.eff.org/static/ ... ueness.pdf

Thanks, I will have a look.

And here lies the problem. Most of us, if not passionate by the subject, could have a really good life without knowing of (that's why I didn't mention it earlier). But exception has to be taken for those who want to recommend and find those recommendations based on false premises because they just don't understand how things work. This is particularly relevant because you're recommending a mix of yours (Anonsurf exists, Firefox exists, but not that brew you're trying to promote, at least in an approved version).

I don't know what is an approved version (and who approves it: yourself?). I use Tor (official), Firefox flatpak (official), the rest is just settings; anybody is able to understand the impact of these settings, with some thinking and some readings.

Maybe I should follow the path of others and leave you in your world (in fact, I already did - just wanna prevent other users from falling into traps you set, even not voluntarily as may be the case of your own computer with the CYT report). So, I will close that debate even if you're free to answer because it's been all said from my side.
However, Cover your Tracks is reliable if you have confidence in EFF's algorithm or are able to understand the results.

I don't set any traps. You don't just understand. (or don't want to understand).

Regards,

MN

[Rtep]

Why is this discusion drafted to Google?? Question still is : Why is Linux Mint "collecting" ANY data from users? That's SPYING. Period. If they telling....because we want to serve you better, doesn't go OK with me. If Linux want serve me better, they should listen to me. My criticism! My complains! They are not taking any criticism verry well.