I'm convinced that the Chinese
will use the Earth-orbit rendezvous technique to send one or two taikonauts
to circle [not land on] the moon and return between [2008-2010]. It would probably involve sending up a 15,000kg
booster-module [perhaps containing at least 12,000kg of H2(l)/LOX propellant] on their most powerful "Long March" available this decade and parking it at LEO [alt:180km/110 miles] where it would quickly link-up with a two-person 7,000kg lunar-Shenzhou to be prepared for TLI.
The numbers work out.
Delta Vee from LEO to the moon
is roughly 3,200m/s[11,000m/s-7,800m/s].
H2(l)/LOX propellant has an av.exhaust velocity [vac] of roughly 4,000m/s.
[LOGARITHMIC APROXIMATION]
3,200 = 4000LogE2.2
3,200 = 4000[0.8]
Mass ratio of roughly 2.2 would mean 22,000kg/(7,000kg + 3,000kg)

As you know, the Apollo project engineers chose the lunar-orbit rendezvous technique for sending men to the moon both to orbit and land.
But the Soviet space-director, Korolev, chose the earth-orbit rendezvous technique instead in an attempt to send a couple of Soviet cosmonauts onboard a 7000kg Soyuz to loop around the moon [circumnavigating it] before the Americans got there.
A landing on the moon by cosmonauts would have used their own version of the LOR technique.
Alas, Korolev unwisely chose to stick to the Soyuz rockets [7,000 kg to LEO]
to attempt the EOR technique. This would have required the unwieldy, dangerous and unreliable method of launching five Soyuz rockets
within a matter of a few days; four launches to park and "link up" four 7000kg booster-modules at LEO, and
one more Soyuz launch to send up the
manned Soyuz to "link-up" with that
assemblage.
Korolev should have swallowed his pride, and worked with his "rival"
who designed and built the PROTON rocket booster. The PROTON could have parked a 17,000kg BLOCK-D stage at LEO
to link up with a moon-bound Soyuz.

Jacob