🌐 Website Issue | Firefox about:config network.cookie.cookieBehavior #1704

Open
opened 2020-02-12 06:27:15 +00:00 by zer0byt · 11 comments
zer0byt commented 2020-02-12 06:27:15 +00:00 (Migrated from github.com)

Description

In the new versions of Firefox, there are five options to manage cookies. The website explains just three of them (number 0 to number 2).

The two other options are:

  • Cookies from unvisited websites (number 3).
  • Cross-site and social media trackers (number 4).

Screenshots

Screen Shot 2020-02-12 at 9 53 26 AM
## Description In the new versions of Firefox, there are five options to manage cookies. The website explains just three of them (number 0 to number 2). The two other options are: - Cookies from unvisited websites (number 3). - Cross-site and social media trackers (number 4). ## Screenshots <img width="549" alt="Screen Shot 2020-02-12 at 9 53 26 AM" src="https://user-images.githubusercontent.com/60427375/74308589-83b38180-4d60-11ea-86ca-f274a1fb0202.png">
Mikaela commented 2020-02-12 13:09:32 +00:00 (Migrated from github.com)

Do you have a source for 3? It seems to be wrong according to Mozilla developer documentation;

0 = accept all cookies by default
1 = only accept from the originating site (block third party cookies)
2 = block all cookies by default
3 = use p3p settings (note: this is only applicable to older Mozilla Suite and Seamonkey versions.)
4 = Storage access policy: Block cookies from trackers

I don't know what p3p settings are apparently for protocol obsolete for around 18 years, but I understand it to not apply to Firefox and 4 seems experimental and possibly shouldn't be recommended yet?

CC: @Thorin-Oakenpants

Do you have a source for 3? It seems to be wrong according to Mozilla developer documentation; > 0 = accept all cookies by default > 1 = only accept from the originating site (block third party cookies) > 2 = block all cookies by default > 3 = use p3p settings (note: this is only applicable to older Mozilla Suite and Seamonkey versions.) > 4 = [Storage access policy: Block cookies from trackers](https://developer.mozilla.org/docs/Mozilla/Firefox/Privacy/Storage_access_policy) * https://developer.mozilla.org/docs/Mozilla/Cookies_Preferences ~~I don't know what~~ p3p settings are apparently for [protocol obsolete for around 18 years](https://en.wikipedia.org/wiki/P3P), but I understand it to not apply to Firefox and 4 seems experimental and possibly shouldn't be recommended yet? CC: @Thorin-Oakenpants
zer0byt commented 2020-02-12 13:56:55 +00:00 (Migrated from github.com)

Do you have a source for 3? It seems to be wrong according to Mozilla developer documentation;

In the browser's privacy preferences, there are four options. From the list, select "Cookies from unvisited websites" then go to network.cookie.cookieBehavior in the about:config and check the value. It's 3. (Checked on Firefox v73.0)
Screen Shot 2020-02-12 at 5 17 22 PM
Screen Shot 2020-02-12 at 5 25 01 PM

> Do you have a source for 3? It seems to be wrong according to Mozilla developer documentation; In the browser's privacy preferences, there are four options. From the list, select "Cookies from unvisited websites" then go to _network.cookie.cookieBehavior_ in the about:config and check the value. It's 3. (Checked on Firefox v73.0) <img width="656" alt="Screen Shot 2020-02-12 at 5 17 22 PM" src="https://user-images.githubusercontent.com/60427375/74341144-21c63c80-4d9f-11ea-89ff-3d2c1723bb52.png"> <img width="713" alt="Screen Shot 2020-02-12 at 5 25 01 PM" src="https://user-images.githubusercontent.com/60427375/74341258-4cb09080-4d9f-11ea-9f8d-ffa00d039a63.png">
Thorin-Oakenpants commented 2020-02-12 13:57:03 +00:00 (Migrated from github.com)
  • 0=Accept cookies and site data
  • 1=(Block) All third-party cookies
  • 2=(Block) All cookies
  • 3=(Block) Cookies from unvisited websites
  • 4=(Block) Cross-site and social media trackers (FF63+) (default FF69+)

^^ these are the actual words used in the UI

I'm not (edit, left out the word not 🤦 ) sure how much you should trust that MDN page, even if it was last updated Feb 7th 2020. e.g

  • network.cookie.lifetimePolicy values have changed (see next point)
  • network.cookie.lifetime.days is deprecated here's the proof - hence (see above values have changed)

Up to you guys what you want to do: no-one is saying you have to list all the values, and value 3 is a waste of time IMO and will just confuse people. I wouldn't be surprised if it got removed and I can't see the point in such a setting TBH.

* 0=Accept cookies and site data * 1=(Block) All third-party cookies * 2=(Block) All cookies * 3=(Block) Cookies from unvisited websites * 4=(Block) Cross-site and social media trackers (FF63+) (default FF69+) ^^ these are the actual words used in the UI I'm **not** (edit, left out the word not :facepalm: ) sure how much you should trust that MDN page, even if it was last updated Feb 7th 2020. e.g - `network.cookie.lifetimePolicy` values have changed (see next point) - `network.cookie.lifetime.days` is deprecated [here's the proof](https://bugzilla.mozilla.org/show_bug.cgi?id=1457170) - hence (see above values have changed) Up to you guys what you want to do: no-one is saying you **have to** list **all** the values, and value 3 is a waste of time IMO and will just confuse people. I wouldn't be surprised if it got removed and I can't see the point in such a setting TBH.
Thorin-Oakenpants commented 2020-02-12 14:04:16 +00:00 (Migrated from github.com)

PS: this (cleaning up descriptions etc) is already slated as part of #1430 which is now been sitting waiting for some action for 3 and a half months - rather than ping me (edit: for things already on PTIO's webpage), how about getting #1430 under way .. just saying /sorry-for-being-grumpy :)

PS: this (cleaning up descriptions etc) is already slated as part of #1430 which is now been sitting waiting for some action for 3 and a half months - rather than ping me (edit: for things already on PTIO's webpage), how about getting #1430 under way .. just saying /sorry-for-being-grumpy :)
blacklight447 commented 2020-02-12 17:21:01 +00:00 (Migrated from github.com)

Hey there!
It true, the issue has been hanging around for a while, but its the next thing on my list to work on after im done writing our new COI and whistleblower policies :)!

Hey there! It true, the issue has been hanging around for a while, but its the next thing on my list to work on after im done writing our new COI and whistleblower policies :)!
Thorin-Oakenpants commented 2020-02-12 17:26:32 +00:00 (Migrated from github.com)

and 4 seems experimental and possibly shouldn't be recommended yet

heh. it's the default :)

> and 4 seems experimental and possibly shouldn't be recommended yet heh. it's the default :)
Mikaela commented 2020-02-12 17:44:42 +00:00 (Migrated from github.com)

For future reference, what is a source for documentation about these flags that can be trusted? 😕

For future reference, what is a source for documentation about these flags that can be trusted? :confused:
Thorin-Oakenpants commented 2020-02-12 17:50:18 +00:00 (Migrated from github.com)

^^ the source code

^^ the source code
dngray commented 2020-10-07 10:28:15 +00:00 (Migrated from github.com)

and 4 seems experimental and possibly shouldn't be recommended yet

heh. it's the default :)

That's still the case in 81.0.1.

Up to you guys what you want to do: no-one is saying you have to list all the values, and value 3 is a waste of time IMO and will just confuse people. I wouldn't be surprised if it got removed and I can't see the point in such a setting TBH.

I think we might fix this by removing the recommendation. We could put a suggestion, there for option 1 (with a warning), but that's really going to be the only useful option, imho

> > and 4 seems experimental and possibly shouldn't be recommended yet > > heh. it's the default :) That's still the case in 81.0.1. > Up to you guys what you want to do: no-one is saying you have to list all the values, and value 3 is a waste of time IMO and will just confuse people. I wouldn't be surprised if it got removed and I can't see the point in such a setting TBH. I think we might fix this by removing the recommendation. We could put a suggestion, there for option 1 (with a warning), but that's really going to be the only useful option, imho
paulo-erichsen commented 2021-04-15 22:05:53 +00:00 (Migrated from github.com)

note that since firefox 86, we can also set network.cookie.cookieBehavior to 5

To disable dynamic storage partitioning for all sites you can use the network.cookie.cookieBehavior pref:
5 | Reject (known) trackers and partition third-party storage.
4 | Only reject trackers (Storage partitioning disabled).
0 | Allow all

it would be great if we could get some some direction on whether it is better to set this setting to 5 or 1

note that since firefox 86, we can also set `network.cookie.cookieBehavior` to `5` > To disable dynamic storage partitioning for all sites you can use the network.cookie.cookieBehavior pref: > 5 | Reject (known) trackers and partition third-party storage. > 4 | Only reject trackers (Storage partitioning disabled). > 0 | Allow all it would be great if we could get some some direction on whether it is better to set this setting to 5 or 1 - https://blog.mozilla.org/security/2021/02/23/total-cookie-protection/ - https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Privacy/State_Partitioning
rusty-snake commented 2021-04-16 06:03:20 +00:00 (Migrated from github.com)

If you have FPI enabled 1 is better (5 will be downgraded to 4 AFAIK).
If you don't use FPI 5 (TCP/dFPI) is better otherwise you would have no isolation.

If you have FPI enabled 1 is better (5 will be downgraded to 4 AFAIK). If you don't use FPI 5 (TCP/dFPI) is better otherwise you would have no isolation.
This repo is archived. You cannot comment on issues.
No Milestone
No Assignees
1 Participants
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: privacyguides/privacytools.io#1704
No description provided.