it’s obviously a sticky situation and different people draw the line at different places. some would probably argue that they should; i personally wouldn’t.

where i draw the line (and where i personally see a clear distinction), is when the service itself is being utilized to facilitate the infringement. in the examples you cite, the services (playstation network, itunes) are not involved in the infringement, they just happen to be tied to devices on which the infringement is taking place. i think it would be a different if case people were freely streaming (paid) movies from the playstation store or itunes; i feel that’s more analogous to what’s happening here because the connectivity that comes with the service is tied to the product being stolen.

i guess it’s sort of an odd-line to draw, but that’s how i see it.

also, to clarify on my point from above, i didn’t mean to imply that the cost to develop/produce the good ia necessarily a major factor, nor should the fact that it be from the same entity that runs the service; $100MM to microsoft is no-different than a couple thousand to an indie dev and, in both cases, it’s in the service provider’s best interest to prevent theft directly facilitated by their network, and i'd say they’re right to do it.

Well, do either of your examples cause the experience of other networks users to degrade, when you've modded the content you're playing?

Yes. Sony can do what they want within their own walled garden.

Now whether or not such a thing would be feasible ... is another story.