Check out this nifty service called Mumbai carpooling project. I think this is a brilliant idea. It might have not made use of the buzzing web 2.0 technologies ( like Ajax, Ruby on Rails or such framework) so may not conform to definitions of wikipedia’s web 2.0 definition, but I think it is beyond web 2.0 🙂 [ Btw of late have been pondering if wikipedia’s definition capture the essence of web 2.0]
Carpooling project is a very good example of social coordination machines.
But I think it is not without its problem but they can be ironed out as this evolves and scales. There are currently no mechanism to stop adverse selection happening out here. Incentives to stop misuse has to be added. One way to fix this I think is by adding a real world connection to the users getting this service. I ‘ll detail it .
While registeration phone number should be asked for as mandatory field. An SMS should be sent to the user with a code which should entered back into the website to complete the registration process. The phone number will be only for authentication purpose and should be maintained as sanctum data otherwise users would lose trust. This way bad apples can be checked out of the system as they realize that they can taken to task if they create a rucus.( As their real world can be tracked down easily if need be)
While I saw this I got reminded of an article that I read about an year ago[ finding roommates that match simillar preferences] . That led me that such coordination machines can be built to solve various such coordination problem. Carpooling is definitely one of them and I am very glad it is in implementation. There is one subtle but significant difference between these two, the former is coordination of preference and the later is that of action and there is more value in the later. I can bet that this will turn into a big killerapp because it embraces a fundamental web ( or technology in general) axiom of killer apps. “They are good at connecting people” [ ebay , craiglist anyone]
While I read this I checked out if anyone is coordinating such an effort for bangalore [It desperately needs one] but instead came across this free map [Kiran Joganaladda, a well known blogger in Indian blogosphere] building activity going on. This again is absolutely kewl and I going to track this one actively too 🙂
mtrajan.com 
Hey Rajan, thanks for the comment on my blog. I’m definitely hoping to implement ajax soon (ruby on rails? where’s my google…) For sure there’s a lot to be done on the map and the new mapping technologies are definitely exciting.
I look forward to your detail on the real-world security connection. My idea for a quick solution was to use already established tust networks (ryze, orkut, etc.) so that people can atleast see how they’re connected to prospective ridesharers and either decide to trust/distrust them, or gather more information from the intermediate “common friends”.
I know Bangalore is in the midst of traffic crisis too. Should we start a Bangalore version? Also, the code is open source, so send me a mail if you just want the source code. If you want to contribute any code (AJAX, etc..:-), that would be super cool. The more expert coders we have on the task, the better the app will be, the better things around us can be 🙂
take care
let me join the bandwagon with my new rideshare website…I thought for a long time how to differentiate myself from other rideshare/carpool websites and came up with Google maps integrated search…
I wouldnt be surprised if that has been implemented already too…