1: The ocean is the only place where it's feasible. The continents have much, much thicker crust (40-70km) than the ocean basins (7-15km). Areas of exposed oceanic crust (e.g. Iceland) have much thicker than normal oceanic crust.
Beyond that, it's (slightly) easier to "drill" through water than through rock. There are a still a _lot_ of challenges involved in deep-water riser drilling. You don't just have a what's basically a steel rod going down (as you would with riserless drilling), you have to have a system that's capable of returning fluids and maintaining high pressure going all the way down to the seafloor (and below, once you case the well). However, it's still faster to deploy the riser than it is to drill through solid rock. At any rate, I'm getting sidetracked. The locations where they're looking to drill to the moho are in relatively shallow water (3-4 km) compared to the water depths that the Chikyu has already drilled in.
2: You're misunderstanding how drilling works. :) In some sense, what you've described is exactly how it works. (Just think of the drillbit as the robot.) However, you have to overcome the pressures at the depth you're drilling at (otherwise the hole will collapse, among other things) and you need an efficient way to return the cuttings (waste material) to the surface. Thus: drilling mud. You circulate pressurized (and very dense, so that the pressure increases with depth) fluid that flushes the cuttings back to the surface and keeps the borehole open. (It's an oversimplification, but that's the basic idea.) In some cases (e.g. coil tubing rigs), this fluid is even what drives the drillbit.
At any rate, hopefully that helps illuminate things a touch. :)
That aside, they're very good and very reasonable questions.
Low bit rate as in < 30 bits per second (As of when I interned for a major oil services company this past summer)
I too don't know much about drilling.
1- I would not do it in the ocean
Lots of reasons, from the potential to damage an ecosystem to the unknowns that could be catastrophic in so many ways. Of course, cost is another reason.
Doing anything in the ocean --particularly deep-- is incredibly difficult and expensive.
2- I would not use a ten mile long shaft with a drill at the end of it.
I just don't think that this is the way to do it. Again, I am ignorant when it comes to drilling of this sort.
What I would do would be to design a drilling robot that would drill the hole and burrow itself down the hole as it works. To continue with the overly simplistic view, the only thing this robot would need would be power (lots of it).
You select a candidate site and build a large solar array, large enough to power the drill-bot.
A starter hole of the required depth and diameter would be drilled via conventional means. The robot would then be inserted into it. It would expand and clamp itself to the hole to begin drilling deeper. If the broken-up dirt and rocks must be removed either the robo-drill has the capability to collect some amount of it and traverse back to the surface or it comes back and a "cleanup" robot would go down to get the stuff out.
It'll take a long time? Is that a problem? The operation could probably have a good degree of automation.
Probably a horrible idea. Don't know.