I generally only bet when I have a reasonable chance of winning (reasonable >= 90%) [that's why I've never played the lottery]. In this case I have no way of determining with
any degree of certainty how long it will take Steve to complete SR V7. Why? Because he himself doesn't put out a schedule so I have nothing to base my bet on. He could get it done in less than three years or it could be more. Also, is that three years after 6.1 or three years from now (not that it would make a difference to me).
In my early programming days I learned structured programming and was trying to introduce it to my company's programming department. I was given a programming task which I broke down into discreet modules. I knew from experience that each module would take an average of one day to code, test, fix and integrate. When I completed my design I told the staff (and my manager) how long it would take me to complete the program. They were amazed when I came in exactly on time. The benefit of this approach was when I was given an unrealistic schedule, I could tell them what a realistic schedule would be and negotiate from there.
The difference between
@Steve's work and mine (back then) is that he doesn't have to deliver on a set schedule. However, like him, I too strive for zero defects in my finished project (I stopped calling them bugs a long time ago).