Hey, thank you for these awesome blocklists!

I started using the beta blocklist file and recognized that bag.itunes.apple.com is needed for the iOS app store to be able to load.

It took me some time to figure out how to add blocklists to Pi-hole. After you know it it pretty obvious and straight forward but you need to know where to look first :)

Hi! Why is the file empty? https://github.com/MoralCode/pihole-antitelemetry/blob/main/telemetry-domains.txt It doesn't make sense to block nothing... Thanks

Found two issues via trail and error:

1. xp.apple.com prevents me to upgrade from 14.4.x to 14.5 via iOS beta preview developer profile
2. humb.apple.com prevented me afterwards enabling my Apple Watch for unlocking my iPhone 11 Pro, when wearing a mask

By the way, running python3 update-lists.py didn't remove the two domains in question from telemetry-domains_beta.txt, albeit it inserts them into telemetry-domains_borked.txt.

there is a pdf (https://www.dropbox.com/s/qaazwyaj2ihj4qa/apple_google_additional_material.pdf) linked in the first article to more information about the connections made, which may contain more domains than are in the initial article.

Would like to have these included in domains.csv too

Would like to add information about what each domain is used for/what information is collected according to the results of the paper. This can help people understand which domains they want to blocked and what else may be affected.

as mentioned in the UPDATING file, there are some domains in the list that represent the same domain but differ by a commonly confused character (like dashes, em dashes, and en dashes). Theres also a few that look like two characters smushed together (I think there's one like that for fi in a couple places). These are likely due to copy-paste errors or from parsing the PDF, however I want to be sure that these different domains have no chance of being valid or in-use domains at all before removing them as this may be a tactic used by google/Apple .etc to evade domain-based blocks using common confusable characters.

For this issue I would like to:

• [ ] look into the validity criteria for a domain name to determine what unicode/funky characters are or are not allowed by the spec
• [ ] replace these characters with the canonically correct version(s) and update the id field for any groups of entries that represent the same domain and are merged with an existing category

This will likely be necessary to have complete before #10 will be able to work properly as that relies on having a consistent domain list

the UPDATING guide in this repo details some quirks and weird workarounds regarding how the CSV should best be edited.

It may be helpful to create an additional python script that allows a domain name and list name to be passed as a parameter so it can automatically update the list of every instance of that domain to the new list.

Related: #9

What other blocklist formats are out there that people would like to see supported?

This is primarily a voting issue for people to leave a comment to mention their preferred format and upvote ones they would like to see added.

I think it would be both interesting and useful to replicate some of the methods used from the linked papers as a means of validating domains in the beta and borked lists to gather some evidence to promote them to the main list.

That said there are other ways to help improve these lists, namely just by adding the beta and/or borked lists to your own pihole installation and report any domains that are in a list that they should not be in (i.e. domains in beta that break apps and services).

Since this repository is mainly intended as a community resource to gather information on these domains, any contributions you can make will ultimately benefit other pihole users and blocklist creators.

as seen by #4, the domains.csv is a spreadsheet of all the domains that were either part of the original paper or were submitted by users to this repo or on Reddit. Maintaining it in a CSV seems a bit cumbersome, so I'm looking for alternative ways to maintain this list as a community resource, ideally under version control.

Looking for ideas for how to do this. Using a shared, public-comment google doc seems pretty counter to the goals of this repo (and Id prefer an open, or at least minimally harmful-to-users platform anyway). Currently thinking of using a GitHub Wiki for this but maintaining it as a markdown table doesn't seem very portable/usable to other blocklist maintainers.