Don't assign incremental IDs to scan results

[ChargingBulle]

https://privacyscore.org/site/ *

By returning an incremental scan result ID you expose various information about your users.

If someone submits multiple pages to scan they get adjacent IDs => identifiable
The order of the scan requests also expose spread of your service.

For example the first scan result is of uni hamburg (www.uni-hamburg.de/)
https://privacyscore.org/site/1/

followed by various other pages (including subdomains and printing services of uni hamburg, apparently you have studied CS there).

Some companies (i. e. Facebook) have elaborate social graphs that could reconstruct the exact flow of webservice recommendations because of the icnremental ID.

[ChargingBulle]

thinking of which it might make more sense to have

https://privacyscore.org/site/github.com/

as URL instead of

https://privacyscore.org/site/18489/

[muelli]

yeah, I'm in favour of that approach. We don't mind the exposure part though.
Anyway, sounds like #18, in parts at least.

[ChargingBulle]

PS implements an open data policy which is great so exposure of all datasets isn't so bad.
Yet it doesn't favour the privacy of the users if you can see (some domain only known to a few) ; pornhub ; (some domain only known to a few)

Many selfhosters have their own domain which means
knowledge of domain=>has connection to the hoster