ABOK#1991 has some interesting characteristics, now that I've spent a bit of time with it.
At first I was a little puzzled by Ashley's drawing of the finished knot, which bore no resemblance to what I ended up with, but then I went back and read carefully the text, which I had flitted through:
" ... by pulling smartly on the end, the knot is jammed and thereafter will not slip."
Well, yes, but ... Ashley was limited on time and space for descriptions, but this was too terse for me to originally understand. In fact, he means that the knot has to be capsized, and by pulling on the working end. Pulling on the standing part will simply reduce the loop size, and actually un-capsize the knot if it's not really jammed firmly.
I'm sure there's an application for the capsized version, but probably not for me. It actually has to be quite loosely tied to pull off this trick (pun intended) and when it's done the loop is less snug than it was, so now you've got a jammed piece of line that's loose around the object you're hitching to. That's two strikes against it in my eyes.
But ... if you go the other way and dress the knot as a completely straight standing part surrounded by two collars, with a "tongue" trapped between them and pressing against the standing part, then you've got a useful slide/grip assembly resistant to movement because of the amount of cordage pressing against the standing part (lots of friction) and conceptually resembling a cable tie probably more than any other arrangement of cord. The "tongue" works roughly like the ratchet of a cable tie but by pressing the standing part against the two collars and kinking it slightly.
For light duty I think it's quite serviceable. I'll try to get a picture when I have more light.
