Attempt towards tracker- and profiling-proof RSS
Albert S
3a723b9440
Fetchers claim to be a certain client. They try to send the same headers as the original client. That's better than a simple curl request with a fake user agent, because curl doesn't send the other headers like the original client and therefore its traffic stands out. |
||
|---|---|---|
| fetchers | ||
| fetcherslist | ||
| randrss | ||
| README | ||
| TODO | ||
What is randrss? ================ Normal RSS readers usually fetch all feeds more or less at the same time at constant intervals. This leaves a signature for anyone who observes internet traffic. By looking for this pattern, you are more easily identified. randrss fetches all your feeds at random intervals at an random order over a certain period of time. The feeds will be downloaded and all you need is a local webserver. The added benefits of this approach are that you don't have to worry about how your client deals with cookies etc. Also, by having only one client fetching the feeds and your readers pointed to the randrss's downloaded feeds, you avoid certain trackers that may identify you across devices (google's feed proxy, cloudflare), because it's very likely that the combination of feeds you read are unique. As your feeds are on a single server now, you can isolate your RSS reader to its own network container so it can only contact that server. This is probably what you should do to be sure your client does not contact the feed servers in any way. In Thunderbird, set browser.chrome.favicons to false. randrss fetches the feed using Tor. The only drawback of this approach is that the time you get new feeds is delayed, but that should be acceptable. Usage ===== First, tweak the shellscript a bit, if you like The input file has the following format per line: url:output file:(optional parameter for the sleep time, in format x-y) An optional user agent file contains the user agents we will randomly use per feed. One user agent per line. randrss [input file] [user agent file]