Ideas

How would you ever monetize this? Here's a back of the envelope calculation.

Assuming you upload a small picture (20k) every second, that'd be roughly 1 Meg a minute. Multiply that out by just 1,000 bars and thats a 1 Gig of data every minute. Run this only in the evenings for bars on the East coast of the US (for example) from the hours of 8pm - 12am and that's 240 minutes x 1Gig = 240 Gig a night. Even if you only took one picture a minute as opposed to one a second that's still a lot of data which would cost an absolute fortune in bandwidth bills.

I think the idea would be lovely, but until bandwidth is much cheaper still, it's going to be very tough to do.

Hm.. if the goal is to find out if it is crowded or not, why not just make a huge list of all the bars and a number behind it that stands for the number of people that are in the bar. I don't know if they keep track of that somehow.

Oh, just had a great thought; this is in the book; The wisdom of crowds; let me get it.. (me gets book, looks at index, finds it, page 84) chapter 5 'Shall we dance?: coordination in a complex world. That's exactly about your problem. If I remember correctly their was not really a possibility for the perfect system, because the choices everyone makes affect everyone. So everyone is guessing what everyone is doing in order to use that information to their advantage..

Ok, I'm not really going anywhere with this, but I just thought it was funny I had the book here and you mentioned it.

RE: Adian It would have high bandwidth but your calculations are a bit off because they don't take into account the benefits of compression with a modern video codec. Also it could be done cleverly so that the video is only streamed "on demand". Eg, if there are 0 users looking at a barcam, there would be 0 bandwidth consumed because the server would throttle the sending end. As far as monetization, if the service could save the person from driving around looking for a bar, that would save on wasted gas and time, and so the site could probably charge a one dollar micropayment for a 30 minute session where the potential patron can check out as many bars they want. If you can convince the bars it will increase their business, or can come up with some sort of coupon/ticketing system, the site could charge both ends.

RE: Julius. Well, until we all have RFID implants and the bars have RFID scanners, its not practical to go for a headcount-based approach. Plus, you want to know more than just numbers, you want to know what kind of people are in the bar/club, whether they are dancing, and so on. By the way, the site would have to have a disclaimer for not being liable for fraudulent bar cams that depict a much more enticing scene than actually exists in the bar. Perhaps a community feedback system to alert other members of bars that are sending fake feeds. PS: very interesting about the similarity from the book snippet .. cool insight.

this idea was already posted under the title:

barsounds
on Jul 10, 2006 at 4:07 pm

the only difference is a microphone and not a camara