Bike sharing: best and worst
Hangzhou, China's system has 51,500 bikes at 2,050 locations, which include transit stations, parks, etc. (DC's system has many gaps.) See "Here is something China is good at: Hangzhou, a city of nearly 7 million, has the most far-reaching bike-sharing program in the world" from the Global Post.
Melbourne's system, with 600 bikes and 50 stations spread around the city and the inner suburbs, in Victoria State, Australia is failing, with an average of 0.72 rides/bicycle/day. See "Bike share scheme disappointing " from the Sydney Morning Herald.
Bicycle Victoria attributes failure to the limited number of stations and a wet summer, and also compulsory bike helmet laws.
The first story links to a Streetfilms video and the second story includes a video as well.
One of the reasons that the Chinese system is well used is that it is cheap to use. The city has conceptualized the program as part of transportation demand management, and so "profiting" from its deployment is less of an issue. In most situations in North America and Europe, profit has to be made, because the local governments don't have enough money to deploy and operate the system.
Auto-related sectors believe that personal mobility is best satisfied through automobiles for this and the next century (that's 190 years) as covered in "this video" from the Aspen Institute, of a panel at some conference.
According to the National Household Travel Survey, in the U.S., 51% of daily trips are 3 miles or less, and an additional 13% of trips are from 3 to 5 miles in duration. A goodly portion of these trips can be captured by biking and transit, if the right kinds of infrastructure and services are in place, if land use development patterns promote compact development.
I haven't had a chance to read it yet, but Transport for London has produced a report, Analysis of Cycling Potential. According to this entry "Why not make the 'cycling revolution' bigger?" in the Guardian Bike Blog:
The report is fascinating in several ways, not least for finding that over a third of the total number of trips made on an average day in London using "mechanised modes" could in theory be cycled - some 4.3 million, of which 3.5 million would take only 20 minutes by pedal power. It also calculates that nearly two-thirds of these potential trips are currently made by car.
So trip behavior there is comparable to that in the U.S.
I believe that this kind of understanding isn't sufficiently highlighted in transportation planning efforts, and so most people scoff when we discuss the potential for cycling--even though most bicycle, pedestrian, and transportation plans do discuss it.