Wait. When asked if you would build it in Today's Google, your answer would be no, because Google is clearly uninterested in this project at the moment. :). I mean really, if Google is uninterested, it's Google who effectively said No isn't it?
Er, no. At the risk of getting all Primer on this, the hypothetical situation posed allowed for a history where Reader was never invented but somehow magically I retained the knowledge of how Google would have evolved in the (hypothetical-to-them) event Reader had been launched there.
Given that knowledge, I certainly wouldn't choose to build it in today's Google given I'd be armed with data suggesting it would not be supported. This point, while kind of mindless and diverting, doesn't seem very useful, though.
Actually, there's a bunch of causality minefields here; we should both be concerned that we could inadvertently violate Novikov's self-consistency principle. My conclusion is for us to back away slowly from the thread – I don't want either of us to accidentally become our own grandfathers.
