We don't have to necessarily instantly jump to 20 MB, but the longer we wait the worse this mess will get.
How do you feel about my counterplan and proposed compromise?
First, we must wait and see how the infrastructure and ecosystem react/adapt to scarcity. Without this crucial empirical data, we cannot make informed future decisions about ideal block sizing. We cannot risk a civil war between BTC1 and BTC2 camps by forcing the issue prematurely (but if I were actually anti-BTC and pro-alt, I'd relish such a Great Schism).
Second, we must use the results of the experiment with 1MB scarcity to decide on a preliminary blocksize increase, to somewhere between 2MB and 10MB.
20MB blocks can then be considered, if and only if the community, pruning/compression mechanisms, and TOR/DSL are ready to support them.
As a compromise, I'll support an increase to 2MB ASAP, on the condition that we then maintain that limit until facts about market reaction to block real estate scarcity are ascertained.
Compromise? Gavin and the team don't have to compromise.
This is a ridiculous counter-plan if there is no way of raising the limit via a soft fork.