I take casual carpool from the east bay into San Francisco on a regular basis (and on the rare occasions I need to drive in, I always pick up carpoolers). What I've observed is that the higher the price of gas, the more riders are available for carpoolers. The win for drivers is clear: a lot of times the morning backup at the Bay Bridge is about 20 minutes, carpoolers can bypass that completely. The other day, I caught a ride with a guy who was going to the same building that I work out of; it was better than getting a cab to work.

Incentivized by time and cost savings, carpooling could be extremely successful. What has failed are the centrally organized carpooling systems operated by governments. And the startups that I've seen that aim to facilitate carpooling have all sucked. I'm certain that carpooling would be more successful here if there were more carpool lanes, like on the hwy 580 and 24 approaches to the Bay Bridge and more carpool lanes in the reverse direction (right now, it's only hwy 80).

While I take mass transit (Muni and BART) on a regular basis, mass transit can't be the only alternative to driving alone. These transit systems are often hamstrung by poor management and drain a lot of tax dollars.

Calling carpooling a crutch and failed urban planning (other comments here) is inaccurate. As a country we pour billions of dollars into bridges, tunnels and highways to keep up with the demand of single occupant cars during commute hours. The lost productivity and the pollution and petroleum dependency growth is an enormous burden on the country. That's a dramatically larger failing than carpooling.