You don't need to continuously teleoperate a rover. That's' why they are made semiautonomous. And they are getting smarter--once again I cite the example of UAV's. In just the last decade they've gone from experimental to ubiquotous.

I won't argue that the most effective use of a manned mission to mars would be to teleoperate from orbit (or phobos). There could be several landers then, and immediate teleoperation which would be more efficient. I don't think NASA could put together a manned Mars program for less that 1 trillion dollars right now. There are just too many scientific alternatives for the use of that kind of money for me to be able to see how to recommend it.

I believe baby steps must be taken to find ways to lower cost, and most of those baby steps should be taken by automated probes and landers. There are no missions in the short term worth the cost of a manned approach. I WANT to do manned missions, but the cost must be feasible. As I've said numerous times, the road to feasibly costed manned missions is more likely through private means than NASA.

Until NASA gets data on low gravity effects on humans and radiation issues and mitigation, I think it is irresponsible to talk about manned missions. Yet, in spending 200 billion in ISS and shuttle missions do we have any data beyond microgravity? NASA is currently too incompetent to handle a manned program if they can't come up with any convincing path to sustainability or even a plan to find the answers for sustainability.

What I continue to offer is the path and the plan towards sustained presence off planet. And that doesn't happen to include some gung ho ill concieved and premature manned missions that waste more 100's of billions.