Actually no. The force required to pull the reel up with the wound thread (or hold it in place) is greater than the force required to pull it up (or hold it in place) with the hook. This is assuming all movement speeds are slow, so angular momentum can be disregarded. Using the wound thread, you apply more force (you have to pull harder) over a shorter distance (the reel moves faster than the thread so you only have to pull the thread a shorter distance than the reel moves). Using the hook, you apply less force but you have to pull it a greater distance, equal to the distance the reel moves. The work done on the reel ends up the same either way.
Here you go:
[qimg]http://www.internationalskeptics.com/forums/imagehosting/130124926e74b16479.jpg[/qimg]
Here's the reel, being pulled to the right by the wound thread. This time a parachute in the wind, the one that's fully unfurled, is providing the pull. The wind is blowing left to right.
The reel works the same as in your own tests: it "winds up" the line as it rolls left to right. This means the parachute is being pulled closer to the reel as it moves. By the time it reaches the reel, the next parachute will be unfurling at the right end, to continue to provide pull.
Now imagine that the wind is pushing the parachute at a healthy fraction of the wind speed (say, 50% of wind speed). That means the whole thing has to be pretty efficient, the chutes have to be large (to generate a lot of force from the wind) while the contraption has to be light with minimum drag. That would be a difficult engineering challenge but there's no question it's possible in principle. The parachute cannot actually reach wind speed; moving downwind at 50% of the wind speed it's "feeling" only half the wind speed that it would if it were stationary. So the faster it goes the less wind it feels and the more drag it has to overcome to pull the cart, so it's probably not going to get much better. It -- keep in mind that by "it" we're talking about the parachute -- will certainly not be able to reach wind speed, let alone exceed it.
But, as you know from your experiments with the reel, if the parachute is pulling the line left to right at a certain speed relative to the ground, the reel is rolling faster than that. The particular reel pictured moves about three times as fast as the "thread" pulling it. Relative to the ground, the parachute is moving left to right at half wind speed, so the reel -- and therefore the cart frame -- is moving three times that speed, which is 1.5 times wind speed. In other words, downwind, faster than the wind.
Remember, the parachute is moving right to left at half wind speed, so it's moving left to right (toward the reel) relative to the cart.
Here's an improved model:
[qimg]http://www.internationalskeptics.com/forums/imagehosting/13012493204e1d7668.jpg[/qimg]
Not much different here; again, the unfurled chutes across the top are moving downwind (left to right) at less than wind speed, the cart is moving left to right faster than the parachutes and the wind, so the parachutes are moving right to left relative to the cart; they're being wound toward the top left wheel. This model is very much like the "Under the Ruler Faster Than The Ruler" in the video. The parachutes are equivalent to the top of the large black wheel. The wind is the ruler. The cart moves under the wind, faster than the wind, while the parachutes move slower than the wind allowing the wind to continue to push them.
The Blackbird works differently. Instead of parachutes moving counter to the cart's movement so as to continue to be pushed by the wind when the cart goes faster than the wind, the Blackbird has a propeller. The propeller surfaces' positions, along the axis of the cart's movement, sweep counter to the cart's movement so as to continue to be pushed by the wind when the cart goes faster than the wind.
Respectfully,
Myriad
