yokoffing / filterlists

Collection of blocklists to fill in the gaps
1.17k stars 36 forks source link

Question about LennyFox's blocklist in Privacy Essentials #144

Open sillyjaybird opened 1 month ago

sillyjaybird commented 1 month ago

I see you use Lenny Fox's Medium Mode blocklist in Privacy Essentials. How do you incorporate it given that LennyFox no longer maintains it. Just wondering if you essentially have taken over updating it for your purposes or otherwise? TIA. BTW, love your work on this project. It's an excellent resource! :)

yokoffing commented 1 month ago

I just incorporated it and gave credit to him in the list back in June 2023. Then I removed redundancies found in EasyList, EasyPrivacy, and Privacy Extended (AdGuard Tracking Protection) lists.

I haven't checked it since, so I didn't know he archived the project. I should probably comb through it again and see what's made its way into EasyPrivacy.

sillyjaybird commented 1 month ago

Thanks for your reply. One reason I ask is that I use the W3C Annual Most Used Survey List from https://github.com/Kees1958/W3C_annual_most_used_survey_blocklist in a very slim tracking prevention list in µBO. On his repo, Lenny Fox directs users of his archived lists to Kees1958's lists as a replacement. I try to avoid redundancies in my lists but I like Privacy Essentials. I'll keep an eye on any necessary changes you might make to it.

yokoffing commented 1 month ago

Previously, there was a only a few false positives I removed out of the Lennyfox list. I'll use the list from Kees and see if I run into any issues.

I'll compare the two eventually. If it's better or more current, I may just remove the Lennyfox list from PE and recommend users to use the Kees list. We'll see.

Kees1958 commented 1 month ago

@yokoffing and @sillyjaybird

When there are any fasle positives, please let me know and I will remove them. For your purpose the Most used Mv3 version and addendum to Firefox and Edge are probably most suited.

Regards

Kees

yokoffing commented 3 weeks ago

After reviewing logs, I'm not encountering sites that trigger either LennyFox or Kees list. So I will just removed LF from PE for now. https://github.com/yokoffing/filterlists/commit/1ceee6d6ad7a600640e9bb2b9a6ec3da918b8d92

Kees1958 commented 3 weeks ago

Just surf to CNN with my list only and you will see heaps of blocks. Adding a condensed list to a bulk list defeats the idea of a condensed list.

For fun compare the blocks of your largest blocklist against using my Mv3 list only.

:-)

yokoffing commented 3 weeks ago

Unique entries on CNN.com:

bitmovin.com^$3p
optimizely.com^$3p
srdn.com^$3p

Three is pretty good for an independent list (i.e. not Easylist or AdGuard).

I'm not sure about blocking Optimizely. uBlock filters unbreak has

! To counter `optimizely.com` in Peter Lowe's list
||optimizely.com^$badfilter

but that's in a first-party context.

For fun compare the blocks of your largest blocklist against using my Mv3 list only

Any other sites you recommend?

Kees1958 commented 3 weeks ago

@yokoffing

Hal, thanks for reporting back

Especially for you (and for fun because I update my list yearly and am preparing for the more generous increase from 5000 to 15000 maximum dynamic rules in ManifestV3) I added a World most used list (also with RU, JP and BR networks sometimes seen on 'western world' websites) with a whopping (at least for me, for Easylist and EasyPrivacy it is peanuts) total of 9675 third-party blockrules.

Re: Optimizely Optimizely is an innocent A/B testing tool. Because of 'block count mania' most adblockers started to block website optimizing and user experience optimizing tools, applying the logic that A/B testing does track users on websites. I started to include them around three years ago, because some of these A/B testing tools also followed the user when they responded to third-part content.

It is a mystery to me, why uBO discharges (badfilter) just one of the many A/B testing tools. I can imagine that some users complained missing out any promotions related to A/B testing. In theory one should discharge (badfilter) all A/B testing tools, to prevent missing out on promotions related to A/B testing.

In the 'block count mania' adblockers even started to block debugging tools (because they also, to some extend, track user interaction). That is why a slider like adustment (like uBO light, Ghostery and Brave have) on a per website basis is the best approach (using a minimal or standard set for trusted websites and a maximum for unknown)

https://github.com/Kees1958/W3C_annual_most_used_survey_blocklist/blob/master/World%20most%20used%20advertising%20and%20tracking%20networks.txt

RE: other websites Just try a few you use a lot (I am curious also to see how World most used performs)

Kees1958 commented 2 weeks ago

Thanks, I removed the FP's