When at Caltech they were developing the rockets to be used on the Ercoupe test they noted that they would blow up on occasion. One did so on the Ercoupe. Then they figured out that the ones that blew up had been sitting around for a while. So they would get up early, cast some rocket motors, load them in a station wagon and drive them out to March AFB for the Ercoupe tests.
Eventually they discovered that the ingredient in the rocket propellant that came to be known as the Plasticizer would dry out if it sat too long. When it did, ignition of the rocket motor would cause the now-brittle propellant to fragment, thereby increasing the burning area, resulting in a much faster increase in chamber pressure and ... Boom! They found that the best material available then for the propellant to keep it pliable was asphalt. Later developments led to CTPB and then HTPB propellant binders.
Not to mention all the cruisers/battleships with catapults which were barely twice as long as the aircraft they launched. I wouldn't be surprised if the acceleration of a catapult is harder on the pilot than that of a rocket booster. The catapult has to accelerate a plane above stall speed in a matter of a few meters, while booster rockets like those employed for RATO usually burn for 10 seconds or more (and part of their thrust also serves to counter the weight of the plane till the wings start to do their work)Wasn't some of the WW2 carriers equipped with a catapult on the hanger deck that shot the aircraft out the side ?
Maybe not exact zero ground run , but close to it.
The walrus was a sea plane, in an open harbour it could be launched from a ship, skim down onto the surface and take off again. If the sea was flat it had no terror at all.According to an unimpeachable source, "Air Stories March 1939," (attached) the landing speed of the Supermarine Walrus was 57 mph.