Instead it is much cheaper and hassle-free to use the firehose connection that already exists into the brain: the eyes, for which even protocols for communication are established. Screens in contact lenses are my bet on where the future of computers are headed. That will with time largely remove the need for physical objects that transfer information.
The interesting part is how Apple in 2020 will spin that they ten years earlier named a display far to bulky to fit into your eye "Retina display".
There's also the possibility of tapping in at the ears, the spinal cord, hell, it's entirely conceivable you could cut off a hand and attach an implant there. When you think about the amount of bandwidth that your mind has between your hand and your brain (seriously, move your hand a bit and think about how much data that is in absolute terms) there's a whole lot of bandwidth in established connections, especially for someone who has been maimed and has brain pathways lying dormant for controlling a limb.
[Update: If it helps rescue me, I found a page about the difference between rendering TS1 and TS3 in their respective times:
http://watchingapple.com/2009/09/pixars-blistering-rendering...
It turns out the original Toy Story took an hour per frame to render, whereas Toy Story 3 can render in real-time (or 24fps, at least.) Given that's probably on a server farm, my claim is still hyperbole, unless you expect cloud processing to be mainstream on all devices by then ;-)]
Assuming a single machine (which is obviously not true, but I can't find any details about the size of the render farm), that'd mean the entire film would take something like two hundred and thirty seven years to render.
Also, note that the WSJ article you linked to is talking about Toy Story 3D which is a re-rendering of the original film in stereoscopic 3D.
However, it was all worth it for your great citation. Thanks!