The network goes live Wednesday morning: Google's gift to its headquarters city is now labeled the largest purely free Wi-Fi network in the U.S. This qualification is necessary because there are larger cities, including some other Bay Area cities, that have advertising-supported free networks (such as MetroFi's) or limited free usage each day. St. Cloud used to be the largest free, non-commercial network, but they have a third the residents. The network was delayed a couple of months from an earlier launch plan to increase density to fill in niches.
Google deployed 380 Tropos nodes and says they spent $1m. The network isn't intended to provide robust indoor coverage, and isn't optimized in my view of their plan for bridges, although Mountain View's general architecture should allow bridged connections pretty easily.
The New York Times focuses on Google not building networks beyond Mountain View and San Francisco, the latter of which EarthLink has said that Google will be a key tenant on a for-fee network that EarthLink will construct. I have said dozens of times since the idea first floated of nationwide Google-Fi that metro-scale Wi-Fi is a low-margin business; applications that run on networks are much higher. Why would Google sink capital into a network that others could build and they could leverage?