July 23, 2007

Scott Adams Stole My Carpool Idea!

I came up with the idea of an efficient carpool network that really works. While I sat over it for an year or so and did nothing about it, by an amazing stroke of co-incidence Scott Adams comes up with something almost identical.

In Palam Vihar, Gurgaon where I live (an otherwise fine place), one long standing problem has been the woefully inadequate public transport. Most residents somehow manage with their cars, however, since single car families are most prevalent - one car is never sufficient. Thinking about this problem over a year and half ago and mulling over the then recently introduced "Centrex" feature by our phone company (call any other resident for free), I came up with an idea that went beyond the needs of Palam Vihar residents.

A massive carpool network that works on a system of credits. Lets say person X doesn't have a car today and wants to go from point A to point B. So he places a call to a toll-free number and enters the locations. Now person Y has a car, is going the same location but doesn't know person X. He calls up the automated voice network, learns about person X, looks up his profile and finding him OK offers him a lift. Person Y just earned a credit sold by X.

To really understand the implications of this, you need to follow a link I'll add later in the post. But at that time I mulled over it for weeks together and developed it in quite detail. I thought Google would be the ideal company to launch something like this. Around the same time, I entered the following brief notes about the concept in my "ideas" spreadsheet as idea #4:
    Description:
    Call to book available rides to your destination from your location in the next 15 minutes. Only registered license holders allowed anonymously.

    Notes:
    Buyer-seller market much like eBay. Credibility feedback plays important role. Needs scale to work. VOIP hosted application to take calls. Algorithms determine who gets matched with whom.

So last month when I come across a post in which Dilbert creator Scott Adams describes an almost identical idea, I couldn't believe my eyes. It was like he scanned my brain while I was asleep! The similarities were uncanny.

More seriously though, this is actually a pretty awesome idea. As Scott says, the biggest problem with carpooling is inconvenience. An all-pervasive carpool network solves this problem. There are only two large differences between Scott's idea and mine. One, I think GPS will make it more effective but I don't think it's absolutely necessary. Second, I conceived it around eBay like feedback by buyers and sellers. I think that would make the system very credible.

P.S: How does the network provider makes money from this? Scott hasn't thought of that. I did. Since I thought Google should provide this service, they make money the same way they always have - from advertising. Since they know where people are going, they can serve pre-registered ads from retailers and companies along the route. If you don't like to get ads / text messages then pay Google a small sum to get an ad-free ride.

2 Comments so far      

Anonymous Anonymous:

how do you propose to keep a balance of sellers and buyers i.e. people giving a ride and people taking a ride. i think this can be a pretty complex system.

My Reasons:

Firstly, What benefit do poeple get out of giving rides to other poeple? credits? what would they do with it.

Secondly, only people who travel daily on a similar route, can offer a ride or their route can be known in advance.

if google is charging money on the advertisement they are displaying, how would they charge the advertisers? based on number of poeple in the car i suppose. What if the car goes just with the driver. google cant charge money to the advertiser but its a cost to them.

What if i've 5 friends coming over.
we all plan to take a trip. but choose to go through the calling network and in a way fix up together. we all decide we are ok with the advertisements from google. Will our ride be free? then, would you also try to identify, if the people travelling together are related to each other or not?

and above all, Personally,
i can live with the guilt of emitting co2 in the atmosphere by travelling alone rather than giving a lift to a stranger.

who would maintain the rating of each traveller. What if we have a new serial killer on the loose. After killing the last person from whom he took the ride, how much time do you think the system will take to update it.

I can think of trillions of reasons why this idea would be a disaster.

I've got a better idea. Good, efficient public transport. look at the transport in small cities like agra. No one takes 1 autorikshaw. Its shared by many people and they pay a fraction of the amount they would pay otherwise to use the rickshaw exclusively.

Or, government can identify common drop points and run small vehicles which can accomodate good number of people, which run after every minute.

this reminds me of another story. remember how US scientists spent millions of dollars to create an ink pen that could work in space. Russians used a pencil instead.

There are a lot of simpler ideas around, which are more feasible.

24 July 2007 at 05:05:00 GMT+5:30 link  
Blogger Manu Sharma:

I'm sorry, I should have elaborated the idea in more detail. I didn't want to make it a really long entry, I assumed that people would read Scott Adam's post that is quite descriptive.

Anyway, my mistake. I'd really like someone to turn this into a real business so I'm going to have to do the effort to explain it thoroughly.

Writing a new post now that will answer all the questions raised above. Will add a link to that post here when it's done.

24 July 2007 at 13:07:00 GMT+5:30 link  

Post a Comment

<< Blog Home