In sci.space.shuttle Jochem Huhmann wrote:
"John Cody" writes:

I'm not wholly against the idea of a crewed Mars orbital mission
(particularly if it includes flybys/landings on Phobos and/or Deimos as a
bonus). It was the mention of 'photo reconnaissance of the Martian surface'
as the primary aim (as opposed to Phobos science or the real-time
teleoperation of Martian robots) that confused me. Is there *really*
anything useful we could learn about Mars that could be obtained by the
early 21st century equivalent of an astronaut pointing a Hasselblad at one
of the LM windows?

When you're flying back anyway you can avoid sending all data back via
the DSN bottleneck (and just take along a rack of harddisks). If you
look at the earth surface mapping missions (using STS) you will easily
see that the sheer amount of data gathered with some instruments are a
real showstopper otherwise.

There are other ways.
For example, a little bird stuck in earth orbit, talking to martian
orbiters over a fast laser link, with a big dish pointed down at some
earth stations.

Tens of megabytes/second is not hard to achieve, compared to a manned
flyby mission.