They will do it the way they do now.The IWF "block list" is loaded at the ISP level and looks for IP &/or URL - try to access one of those and "nope". Similar tech is used for the court order sites (torrent etc).
The ISP level filters (currently opt in/out) are looking at URLs mostly - but once they start doing traffic identification and content inspection, it's a whole new ball game.
HTTPS (assuming no MITM certificates) will prevent a lot of HTTP content inspection, however then domain/subnet level blocking can be enacted.
VPN usage will go up, until that's blocked - "Sorry, we can't decrypt your traffic..."
You missed the point. Assuming the "viewer" passes the "age verification" step. Who would go about unblocking what and where ??
URL blocking of gov. defined "bad sites" is a constant, there is no opt in for things like mega*.* or whatever its called this week..
Content inspection - good luck with that. One image where the algorithm detects some skin tone - is it just a picture of someone on their holls, some medical picture on the NHS web site, a person wearing a beard or one or more people having some fun. Now change the location, change the race of one or more people, now change the camera angle, light level etc.
What if its just a picture of a crowd of people. The tech for "block porn" doesn't seem to exist just in the same way that the definition of "porn" doesn't seem to be too easy to define either. Its no different in many ways to the definition of virus or malware - none of those technologies are 100% accurate either.
So, back to the original point, when there is opt-in, there can be multiple people behind the same home firewall or there can be the same person on different devices or someone checked into a hotel or a coffee shop. The ISP can only block on source or target address type rules. The remote site can do all sorts but I can't see every web site owner world-wide doing this given that it seems to be a UK initiative.
Biting the hand that feeds IT
