Hey Robert John - you are about 3 years into this stuff right? A what point did you feel you were doing people more good than harm by showing them stuff on your own?
Heh. No offense to anyone involved (i.e., both Robs) but when I read that last question I had to laugh out loud. I'm sure Rob L. didn't mean it, but I sort of read the last question as being similar to asking "At what point did you feel like you had quit beating your wife and were now helping her?". I.e., the question presumes that Rob John must have been misdirecting his students at some point.
On a more serious note, I would offer the suggestion that there would be 3 serious steps in the process, which I'll take a personal stab at trying to define:
(1.) A person should not be "teaching" others or "showing the way" until they can easily replicate, let's say, the static "ki-tests" that Tohei uses in Ki-Aikido (well, I'm talking about the ones I used to see in books years ago).
(2.) A person should not be teaching moving exercises before they themselves can exhibit reasonable/consistent movement that has ki/kokyu skills throughout it at all times and movement is led by the hara.
(3.) A person should not be teaching Aikido waza, in my personal opinion, until they're fairly consistent at #1 and #2, but of course this is not going to be the case, in reality.
The trick at the moment is to slide as much and as pure ki/kokyu skills into existing organizations as possible. It's going to be very hard to do it adequately. A lot of schools, etc., couldn't care less about putting ki into Aikido, but that's actually a positive thing. Let them go; it's just less to worry about.
One thought I have is that it's best to work with a whole school and not individuals. Piece-mealing things seems to just not work as well as when an active group makes an effort to change over. That's an important point. If you don't have a school working together, someone who is focusing on the skills almost invariably is forced to drop out and work by himself in order to get the engine running (I think Mark Murray and a number of others will vouch for this).
Anyway, those are just some thoughts, FWIW.