I don't find that surprising. The communication link back is presumably by satellite since it is in the middle of the ocean, and probably a directional antenna because of the high bandwidth. You have any good suggestions for having a reliable connection via directional antenna on a flimsy barge that a rocket is landing on in this middle of the choppy north Atlantic?
Tow a cable from the barge to a nearby ship, buoy, or platform on which the antenna sits, that's out of range of the vibrations and faces a different direction?
I believe they are required as part of FAA regulations to ensure that no manned craft are within 15 miles of the landing zone? Something like that, at least, which would complicate a tether-based approach.
I think the suggestion was a separate but tethered unmanned platform, which would presumably have less vibrations. I would imagine it's just not cost-efficient for SpaceX, as they'll be able to recover the footage later regardless.
Who needs a tether? Ubiquiti makes gear capable of slinging 400+ Mbps over 25km in a straight line. For about $1500 they should be able to shove the "last mile" to a ship outside of the exclusion zone and put the sat uplink on the ship.
They have grid antennas with a slightly wider beam, or maybe a sector style with 20-30 degrees could be used. I just think it's easier to engineer a workable ship-to-ship stabilized radio solution rather than deal with the ship-to-sat dish which probably has much tighter tolerances.
On the pictures of the droneship the quite distinctive Inmarsat BGAN (which is in essence 3G bounced of geostationary satellite) antenna pod is plainly visible.
Edit: inside the pod is fairly high-gain directional antenna mounted on motorized positioner, but it is designed to track the satellite from slow (ie. ship) or predictably moving (ie. truck) platforms and certainly cannot cope with vibrations from the landing reliably. Man-portable BGAN terminals usually have fixed antenna and officially require quite lengthly positioning process (on the other hand it will work for some value of "work" when you just throw it into car trunk and park with trunk vaguely pointing to south, but you will be lucky to get reliable phone connection, not to mention video stream, in that case).
How about a camera feed from the tow ship (with a view like the two other boosters landing)? If you can film a rocket 300km in the air, you can surely spot a rocket landing 15 miles away. Unless ofcourse you don't want the world to see it crashing into the ocean live.
