Trackers blocked in the browser - Firefox vs Edge

If I block Zopim.com in Edge:



Edge does not block any Zopim trackers:



It is a demonstration that this tracker is blocked by the browser.

_________________________________________________________

With Firefox (unless someone explains it to me) I'll hazard a guess.

In UBO,Zopim.com is by default allowed.

So Firefox would block Zopim,UBO would not block it.

To me there is like an overlap of intervention that overrides the right action of Firefox.

I hope this wild guess of mine is completely wrong because those who use this browser should be very worried.

 

I think it's not blocked by uBO, because it has a legitimate use for people that want chat functionality.
(Google search)
See e.g.:
https://github.com/StevenBlack/hosts/issues/1625
https://support.knowledgehook.com/h...54849151-Domains-to-whitelist-on-your-network

It's not in any list of uBO, as far as I can see.
'widget-mediator.zopim.com' is a websocket.

https://dl.dropboxusercontent.com/s/29jojueeselkvkr/screenshot_zopim.png

 

OK.
You know Firefox much better than I do.
Look at the hopefully wrong conclusion to the previous post.

I am very tired today.
My wife is calling me and almost threatening divorce.
It has been a very busy work week for me.
And I think I will not get on the Internet tomorrow as well.

I wish a good weekend to all forum members.

 

Haha
I wish you a good weekend too.

 

I'm tired as well.
But before you go, I just noticed that Firefox blocks the tracking cookie from 'widget-mediator.zopim.com' (when disabling uMatrix)

If you want the nuclear option regarding websockets:

Code:

*$websocket

And add exception rules for sites you trust, e.g.:

Code:

@@||discord.com^$websocket,1p

Sleep well.

 

Thank you both.

Instead, I got little sleep.
I was wondering:

"If the tracker was not blocked by both the browser and the adblocker where did it go?"

Then an insight.

UBO must of necessity have stopped something upstream.

It could be this CORS request:



That as you can see there is no error in the absence of UBO:



Unfortunately, browser/adblocker engineering is certainly not my subject.
So a certainty requires the right person.

 

See @gorhill's comment here:
https://old.reddit.com/r/uBlockOrig...ng_domains_are_not_blocked_by_ublock/im5b4fn/

https://github.com/gorhill/uBlock/wiki/Dashboard:-Settings#important-note

 

Yes,we know that prefetching is not working in chrome-based browsers.
That's why in the UBO thread I have often recommended more anti-tracking lists than Edge.

But this behavior in FF + UBO I suspect is caused by something else,also because unchecking the specific rule doesn't change anything.

 

For those who like competitions , however, it is possible,in the case of Firefox, to use the test below:

https://d3ward.github.io/toolz/adblock.html

Code:

Testing urls => Ads | Number of tests => 25
Testing urls => Analytics | Number of tests => 14
Testing urls => Error Trackers | Number of tests => 6
Testing urls => Social Trackers | Number of tests => 26
Testing urls => Mix | Number of tests => 9

Considering that Firefox's built-in anti-trackers protection has a specific entry for Social Trackers:



,the difference is blocked by UBO lists (directly or indirectly).
In my case these are:

Firefox = 10/26

UBO = 16/26

Much more arduous to do the calculation with Edge, which does not separate the items.
If any forum member wants to try it.

P.S.

It is interesting to consider that Firefox with UBO disabled does not block 4 Social Trackers groups in the test.
While Edge with UBO disabled does not block 5 Social Trackers.

Both browsers with UBO enabled block all Social Trackers.

So I always suspect that Firefox and Edge are not using common sources.

 

The source is the same (Disconnect).
But it may be, that they use different whitelisted items.

 

Yeah,your conclusion is certainly accurate.

Lists that are the same but ultimately lead to different results.

I am waiting for some forum member who is able to replicate the test and have the same results (otherwise we violate the basic rule of the scientific method).

 

Just tested in a Windows 10 VM snapshot with clean new profiles of Firefox and Edge; with uBO as only extension (default and updated lists). Hosts file disabled.
Both browsers with strict protection enabled.

Score on the test site:
Edge:
uBO on: 93% (76 requests blocked)
uBO off: 80%

Firefox:
uBO on: 94% (116 requests blocked)
uBO off: 83%

Note:
You can get slightly different percentages when refreshing...

 

Last test and conclusions.

In the test:

https://d3ward.github.io/toolz/adblock.html


, the activation of UBO or its deactivation is irrelevant.
Both browsers block the same total number of trackers:

Firefox = 45 (it is necessary to count blocked trackers)

Edge = 41 (the total number of trackers blocked is in evidence):



So in the case of Edge, the AdBlocker must block a higher percentage of trackers.

So although the original lists are identical. (Disconnect) (thanks nicolaasjan).
Microsoft subjects the trackers to other controls:

https://docs.microsoft.com/en-us/microsoft-edge/web-platform/tracking-prevention


The result is a lower ability to block trackers compared to Firefox.

 

With the test (although with all its limitations) it is possible to understand whether the anti-trackers lists (UBO,AdGuard...........) especially in Edge are sufficient.
They must ensure a higher number of blocks than the lists in Firefox's UBO.

In my case both tests in the 2 browsers have the identical score (99%).
The anti-trackers lists in Edge are therefore perfect.

______________________________________________________

My work is therefore concluded.
I hope it has been interesting and not boring.
In case my apologies.

 

I did the test with Android Opera:




Browser blocked tests = 44/80

Tests blocked by DNS (OISD Full) = 26/80 (I have noticed that this type of blockage is not constant)

Unblocked tests = 10/80

 

I noticed a decrease in percentages on the d3ward test.
Probably the testing websites are too old and out of date.
So its reliability has certainly decreased.

 

I see the linked page is updated periodically.

This development is worth keeping an eye on:

This will no doubt be an imporvement to Edge's Tracking Prevention.

 

Interesting.
TH.

 

The test site seems to have issues. It doesn't work properly on Firefox for me. Specially when uBO/Adguard is enabled.
Also on Firefox, uBO is prioritized over the built-in tracking protection. uBO blocks earlier. It's the opposite of Edge & Brave.

 

One could use the website suggested elsewhere by @Jan Willy:

https://www.fastcompany.com/

The problem is to determine with certainty at any given time how many trackers and ads are present.

 

A LOT of trackers according to webbkoll:
https://webbkoll.dataskydd.net/en/results?url=http://www.fastcompany.com/

And cookies too: 118 third-party.

 

https://whotracks.me/ counts the number of trackers.

 

Guys need to take a point of reference.
Because our purpose is to check the performance of our browsers + adblockers and in case take action.

I would say consider (by necessity the ads as well) otherwise it is not possible.
Because in the count of our adblockers there are by necessity also ads.
Moreover we need to consider as 1 (those trackers/ads) that the adblocker blocks maybe 2/3 times in the web page.

It seems that the total web trackers/ads in fastCompany is 69.



I would say take Firefox as a reference point.
Because by default it has third-party cookies set to on with total Cookie Protection.

To me Firefox,now, blocks 19 types of Trackers.

69 -19 = 50 (trackers/ads that need to be blocked by UBO,AdGuard....).

Could any forum members verify this?

I did the verification with UBO = 55



You have to scroll the web page all the way down.

From this count you need to consider (in my opinion) that some ads/trackers are blocked by UBO multiple times so the right number of blocked trackers/ads is less than 55.
And then there are other types of blocking (which in my case are not there because I have eliminated a lot of non-ads/trackers lists that some of you use instead.

So to do as fair a test as possible it would be best to remove the anti-malware-phishing-**.......... lists from the adblocker......

It is not easy......unfortunately.

 

A very interesting benchmark is the number of connected domains. See https://github.com/gorhill/uBlock/w...ser-interface#the-number-of-domains-connected

 

I decided to switched Quad9 DNS (DOH) in Edge with NEXT DNS (DOH) because the total number of blocked trackers/ads was lower than in Firefox.

I also did a check in Firefox with Next DNS (DOH).
NEXT DNS lists work,although unfortunately direct verification is not possible.
The blocks in both Firefox and UBO are in fact lower:

Firefox = 13 trackers
UBO = 47 trackers/ads

I therefore advise forum members who get in FastCompany a total number of blocked trackers/ads much lower than 69 to take action.