Setting up shop at a Lagrange point is a whole lot more interesting and likely profitable. Unless you really want little green men.
See http://www.cellebrite.com/mobile-forensics. Every Apple store has Cellebrite phone forensics software and so do a every police agency who can afford it.
There is plenty of bandwidth in Comcast's last mile cable infrastructure if they are willing (and provide resources) to light it up. After that Apple has it's own "internet" to play with. I think it's ingenious because back in 1995, I did a hypothetical business case based on what was then called The Microsoft Network. MSN 1.0 was a pre-internet alternative to AOL, but used a lot of Microsoft presentation level protocols as well as internet protocols. Back the, the ISPs and content providers were often the same.
At the time, I argued that with a slightly larger investment than they eventually made, Microsoft could have had more users connected to their "private internet" than were on the public nets or AOL. That would have lead to a "proprietary" internet and stunted the grown of the open network - at least for business and consumers. MSN 1.0 was delivered as a seamless backend to Windows 95. The last mile in 1995 was mostly dial-up, and you could only have one connection at a time. Consumers gravitated to the connection with the most value, but eventually the low cost of internet connections overcame the value of single providers. MS didn't invest enough and they lost ground (not visionary enough?) and the rest is history.
An Apple/Comcast "private IP network for media" that Apple/Comcast may be working on doesn't have the disadvantages of dial-up, but it does provide a protected, predictable, and well provisioned network for those that simply want something to work (no stuttering, no drop outs, and access to lots of stuff). If they invest enough, they could make this work. People will pay for value and they are barely getting it from cable providers now. Comcast needs something and given the short term thinking that pervades public companies, this could pay off if the alternative is wire cutting and churn.
Yes, this is an aggressive "land grab" by Apple, but with the consolidation in the ISP market, how else would you move ahead with re-inventing and delivering these kinds of services. It's gotten to be an ugly take-no-prisoners landscape. Read about what the railroad barons did back in the late 1800s if you want to see the outcomes of similar moves.