The idea of smallarms engagements beyond 300 yards is a rifle range mentality that has nothing to do with the realities of modern infantry combat. As already noted, hit probability at 500 yards drop to nil, and 90% of all small arms fire takes place at 300 yards or less. This has little or nothing to do with marksmanship, but rather the ability to see the target and intervening terrain. Unlike wars in the 19th and early 20th century, modern fighters are taught the use of cover and concealment, the use of camouflage and fire and maneuver. Even the most experience, highly trained shooter cannot hit a target he cannot see.
Volume of fire has value not just for fixing an enemy's position, but also to make return fire less effective even at the cost of ammunition. All those 'wasted' round have led to a reduction in friendly casualties. Better to trade bullets for wounded or dead soldiers, IMO.
BTW, the average range of engagement in the ETO in WWII, where the 30-06 was the standard round, was 75 yards. Marines in Iraq are averaging 50 yards.
Longer ranges are really the realm of intermediate weapon like mortars and other area effect weapons. Point weapon like rifles are to easy to render ineffective by cover and distance.
This thread may be of interest:
http://www.thehighroad.us/showthread.php?t=327113
The 6.8, like every other round developed as an answer to the 5.56, is a compromise limited by what can fit in an M16 and be controllable under full auto fire. The latter is something frequently missed by civilian shooters who've never fired a select fire rifle and don't appreciate how small changes in recoil are amplified when more than one round goes down range per pull of the trigger.
The goal was better lethality at combat ranges. There is a lot of room to argue that adoption of the M855 over the M193 reduced effectiveness of the M16 platform. It seems to make little sense when one considered that the M855 was adopted in part because it offered better long range potential than the M193, while the Army's own studies indicated that small arms aren't effective at longer range and that the effectiveness of small caliber projectiles require high velocities to be effective.
In particular, the 5.56 has been particularly effective due to it tendency to fragment, and this phenomenon requires a velocity of around 2800 fps IIRC. Adopting of the M855 reduced initial velocity of the 5.56 by about 300 fps, reducing the fragmentation range by a good measure in return for longer effective range that wasn't needed.
There is a valid argument for a round optimized for very short barrels which rob the 5.56 of its necessary velocity. But there is an equally valid argument for staying with rifles and preserving carbine length weapon for where they are really necessary.
