hagezi Blocklist

[Dasilo 0]

• Name: Multi ultimate
• Description: Ultimate Sweeper - Strictly cleans the Internet and protects your privacy! Blocks Ads, Affiliate, Tracking (+Referral), Metrics, Telemetry, Phishing, Malware, Scam, Free Hoster, Fake, Coins and other "Crap".
• License : MIT license 
• A raw URL which our system can fetch from periodically in order to build the list: https://raw.githubusercontent.com/hagezi/dns-blocklists/main/domains/ultimate.txt
  
  You can find more of his List [Even some that dont block so hard here: https://github.com/hagezi/dns-blocklists#ultimate

Edited ... by Dasilo
Removed Link

[JQ0P5ZCY&b 1]

Seconded. I'm using the pro++ https://raw.githubusercontent.com/hagezi/dns-blocklists/main/domains/pro.plus.txt as it seems a better compromise for my usage.

[veep1 0]

I would also really like to see Hagezi's lists included in AirVPN (at a Minimum the 'Lite' and the 'Pro++' lists) these are some of the best meta-lists at the moment, actively maintained, and promptly updated. Comparable to but more comprehensive than the OISD lists.

https://github.com/hagezi/dns-blocklists

direct links to the lists:
Light https://raw.githubusercontent.com/hagezi/dns-blocklists/main/adblock/light.txt
Pro++ https://raw.githubusercontent.com/hagezi/dns-blocklists/main/adblock/pro.plus.txt

Edited ... by veep1

[OpenSourcerer 1363]

2 hours ago, veep1 said:

Comparable to but more comprehensive than the OISD lists.

More comprehensive != better. Anyway, I will probably test some of these lists myself.

[OpenSourcerer 1363]

I've been testing the Ultimate list plus Threat Intelligence Feeds as per the recommendation.
The network-wide block percentage rose to ~15%. The most blocked domains were queried by an iPhone: Most of it was Apple telemetry with a bit of app-specific telemetry mixed in.

During regular operation the only annoyances I've found were link shorteners and homepages of some analytics software being blocked.
Blocks of websites with internal trackers were very rare, but occured: There was a short time when www.kobo.com was on the blocklist, which, among Tag Manager, Polyfill and other external trackers, includes a web-vitals.js script from static.kobo.com. I guess, some list included the AdBlock equivalent for *.kobo.com, leading to this false positive. Frequent updates are therefore recommended.

[veep1 0]

On 9/19/2023 at 1:02 AM, OpenSourcerer said:

More comprehensive != better. 

True, neither more or less comprehensive lists are objecively better. But having *the option* to use a more comprehensive or more aggressive list (or less aggressive/more conservative list) is objectively good.

When it comes to blocklists there is no objective best option, Its about finding the list that works best for your context and your priorities (e.g. for network wide blocking or multi-user contexts a more conservative minimal blocklist that is tuned towards minimizing breakages or false positives is preferable whereas on my personal devices I like a blocklist tuned towards more aggressive blocking at the risk of slightly increased changes of breakages or false positives).
 

Quote

I've been testing the Ultimate list plus Threat Intelligence Feeds as per the recommendation.
The network-wide block percentage rose to ~15%. The most blocked domains were queried by an iPhone: Most of it was Apple telemetry with a bit of app-specific telemetry mixed in.

I suggest you take a look at this analysis and discussion for some context and statistics on why the "Light" and "Pro++" lists tend to be the most recommended options. The ultimate list is a good option for some people, but its not the recommended option for most people and has the highest risk of false positives and breakages. So if Hagezi lists do get included, it should not just be the Ultimate version.

In my somewhat-informed opinion. If just one Hagezi list were to be added it should be Hagezi Light, if just two are added they should be Light & Pro++, if three than the supplemental full Threat Intelligence feed should be added, and if four Ultimate list should be included.