Open anbasile opened 8 years ago
Great idea, especially to test new routes to see empirically what are the most frequent meeting points.
Ha ha! As far as "lean business" models go, this is the leanest (though surprisingly effective) you can go! I have a friend who built his entire company around such a model. He started with an idea for creating software to automate obtaining tax help and advice, but he didn't have the coding muscle to implement such a system, so he hired some tax consultants to stand by and feed advice back to the system. To the end user, the experience was seamless. They had issues and the system offered directed help. It was so successful, that he was able to raise the funds to eventually implement the necessary algorithms to build the system and he was eventually bought out by Intuit for a very large sum of money. He is now director of product development at Intuit, one of the largest software companies in the tax and finance sector...
If you have the time to observe the system and oversee the kind of request that come in, this would be a very smart system! ;-)
On Sun, Jan 10, 2016 at 5:01 PM, Angelo notifications@github.com wrote:
Imagine a scenario where there is a human talking to pickmeup users in real time: 1 passenger writes (or sends a voice message)to bot asking for a ride 2 bot forwards the message to the human 3 driver writes (or sends a voice message)to bot offering a ride 4 bot forwards the message to the human 5 Human responds to both of them
This would allow us to see what is actually happening: are there patterns? what is that users need? Are there any problems in the implementation of Pickmeup? Or there are problems in the interaction model?
The interface would be the coolest one: plain natural language No buttons, nothing to set Eventually an interface will be built for the human so to start annotating things and see what can be automated
Facebook is working on something of this kind: https://wwwfacebookcom/Davemarcus/posts/10156070660595195 We are still small enough to afford doing something similar: I volunteer for it
Please disuss
— Reply to this email directly or view it on GitHub https://github.com/kercos/PickMeUp/issues/22.
Imagine a scenario where there is a human talking to pickmeup users in real time:
This would allow us to see what is actually happening: are there patterns? what is that users need? Are there any problems in the implementation of Pickmeup? Or there are problems in the interaction model?
The interface would be the coolest one: plain natural language. No buttons, nothing to set. Eventually an interface will be built for the human so to start annotating things and see what can be automated.
Facebook is working on something of this kind: https://www.facebook.com/Davemarcus/posts/10156070660595195. We are still small enough to afford doing something similar: I volunteer for it.
Please disuss.