The Liability Circle of Copyright Filtering

I feel for Microsoft and the other companies trying to integrate the co-called ‘copyright filtering’ (a term which makes it sound nice and simple like looking for naughty words in a text file) technologies.

My time at Kazaa trying to integrate these tools gave me a rich appreciation of how messy the process is.

• The problem is a hard one to solve. How does a machine know a collection of bits is copyrighted
• The technology is difficult to integrate into a real work flow because it takes a lot of CPU cycles and time to perform the filtering process. This is probably what the intrepid Craig Rubens was seeing in his testing when it took Soapbox days to filter the file. Of course Microsoft could just stop publishing files until the copyright filtering job has run, but is it a good product for people who want to upload their personal videos if it takes days for the video to show up

Of course these are just technical issues and can be solved in time.

The real problem is, these days there is a good chance you will get sued if your solution is less than perfect. And who will take that responsibility Liz Gannes’ report on NewTeeVee encountered this problem:

“We called Microsoft to ask what was going on. You should talk to Audible Magic, they said; our system is only as good as their index. We called Audible Magic, who essentially blamed Microsoft, for only implementing the audio version of its software.”

This is a technologically challenging problem that is clouded by lawyers designing the software either directly or vicariously via contracts and settlements. Everyone else is busy thinking about how to best cover there ass instead of building copyright filtering solutions that work.

Filed under Internet TV |