I think it is simple.

Here is my language: "Your start is scored for distance and time when
and where you exit the start cylinder the last time"

As Larry points out, the issue of separating pre-start and post-start
traffic is fairly easy to solve. As one minor addition, if you do slip
back in to the cylinder, you can then start again if you stay under
for two minutes, or take your original start, but scored in the old-
fashioned way. The scoring program already knows to look for multiple
starts and take the best one. The scoring program already finds the
"start fix" so the change is not hard to program.

Previous discussions of this option at SRA meetings focused on a
different issue, highlighted in the pro/con of the poll. In no wind
the "start anywhere" option is great, it's like a start line because
all parts of the start circle are equally good. But in significant
wind, the optimal start point is at the upwind edge of the cylinder,
rather than at the wind-triangle upwind point under current rules. If
in addition, it's a crosswind or downwind, the optimal point is 90
degrees away from the courseline. Now, as currently, it's not a huge
big deal to start 30 degrees away from this optimal point, but it is a
bit worse than currently because you start and then make a sharp
course change. Still, the US RC wisely decided not to use a start
line, because in a significant crosswind it funnels all gliders to one
point in space, the upwind edge of the start line. There is some
concern that the "start anywhere" option would have this same effect.

The most important question, I think, for the poll, is how do pilots
feel about this? Are the obvious advantages of "start anywhere" on
days without much wind offset by the potential disadvantage of this
scenario? In your experience, how often is there enough cross or
downwind on the first leg that this would be a problem? The RC is
pretty good at thinking through traffic issues, but this really is a
pilot preference issue, and hearing opinions on the poll will be very
useful.

John Cochrane