[p2p-research] Never Mind the Policy: Can Filtering Technology Stop p2p?

[Tomas Rawlings]

Source: 
http://blog.catbot.org/content/never-mind-policy-can-filtering-technology-stop-p2p

There is another story in the media about the ongoing debate about what 
(if anything) the government should be doing in response to online piracy;

A rift has opened between music's creators and its record labels, with a 
broad alliance of musicians, songwriters and producers fiercely 
criticising the business secretary Lord Mandelson's plans to cut off the 
broadband connections of internet users who illegally download music.

What I find interesting about the coverage is that there is little to no 
discussion about if such measures are possible. By that I mean, is the 
technology there to accurately stop piracy – so, assuming you had robust 
legislation giving a clear definition of what was to be done – is the 
tech there to do it? There is no doubt that some technologies exist, but 
I suspect that there are serious limits to what it can achieve. For 
starters, any technology solution would need to have some or all of the 
following to be practical:

* Easy to implement – The network ecology that this technology is coming 
into is dynamic, thus if the technology takes time to implement, then 
the conditions may have changed and rendered the tech obsolete.
* Low bandwidth – if the process of checking all the traffic is taking 
too much out of the network, it is going to impact on the service that 
people will have paid for – as well as penalising all users, regardless 
of if they are indulging in non-legitimate uses.
* Discriminatory - Able to discriminate between legitimate and 
non-legitimate traffic accurately. If it targets the wrong people, it is 
going to cost time, money and reputation for the ISPs involved.
* Decryptory – There has been marked rise in encrypted p2p technology of 
late. This means that any system will need to deal with not only 
encrypted peers, but also encrypted packets of data.

Now those are the 'easy' points – the obvious ones that the technology 
would need. From the p2p research I've been doing, I would suggest that 
there are a couple of other ones of note too:

* 100% Effective – One key thing about internet ecology is that it is 
very, very easy for users to change p2p software. Users can also 
communicate virally, so can pass information on about holes in any 
system rapidly. The exploitation of the loop-holes will also be viral. 
This means a 99.9% effective system upon implementation will rapidly 
become a 0% one.
* Psychic – The other issues is that once a flag goes up, there will be 
many, many people trying many many methods to by-pass the system. This 
will be continuous and will consist of lines of attack from new 
directions. The system implemented need to anticipate and head-off these 
attacks to stay in-the-game.

I am being a little provocative in the points above, but I feel they are 
valid. I think you see what I am saying – stopping online piracy is 
very, very difficult. Some may point to the Great Firewall of China as a 
successful example of internet control – forgetting that it is backed by 
an authoritarian state and might not be effective for all we know – 
became its failings are not going to be publicised either by the 
government there or the users who've found a way though.


-
Tomas

-----------------------
Tomas Rawlings
Development Director, FluffyLogic Development Ltd.
web: www.fluffylogic.net
tel: 0117 9442233 
-
Also see:
blog on film & interweb: www.plugincinema.com
blog on p2p, media ecology & evolution: blog.catbot.org
tweet: www.twitter.com/arclightfire