One of the neatest, most elegant guns of all, is the JW&S Automatic patent; each of the 4 or 5 I've handled - all hammerless sidelocks - have the unusual detail of a roller bearing on the forward underbolt.
This is because the barrels take over the cocking of the action in there last few degrees of travel and Woodward being the conscientious people that they were, thought to remove as much friction from the leverage as possible.
Something seems off with that line of thought, being that the barrels do the final cocking. I would much rather the underlever do the cocking of the hammers, until the sears catch. Just after that point is when I would want the sliding lock to clear the lugs allowing the barrels to open. I'd also want the forward lug to put the hammers into an overtravel state. The roller bearing reduces friction since you are overcoming the spring tension from the underlever spring and both main springs.
I guess it comes down to your interpretation of "cocking". I don't consider hammer overtravel to be cocking, since by using the word overtravel you've already established the sears have engaged the hammer/tumbler notch.
How the Scott Spindle won out is beyond my comprehension, it's only slightly more user friendly. With that you reduce the strength of the standing breach and action bar to do the metal removal for the spindle, and in the event it cocks by barrel movement you've got to make room for cocking rods or levers. More machining steps and complexity. Another drawback of barrel cockers is if things are out of time you get firing pin drag. The underlever cocking the hammers will never have that problem. For these reasons I'm in the process of building two(one paid for)back action SLNE hammerless double rifles using the Woodward automatic concept, minus the funky hammer blocks and in return adding true intercepting sears. Coming up with a way to speak to an ejector mechanism would add a little work.
