1. The platform yields savings of 90% not 10%. A billboard costs about $2500 dollars a month and 10 smart tv covering the same area will run you about $250 per month.
I really think you're underestimating costs, and at the same time is not really fair comparing a billboard with smart tv, not in an area covered and not in price if you make it the same size, no lighting but electricity for the tv. But, without a bill with detailed costs let's say it's possible.
However a very lucky customer will win the equivalent of a block reward which would probably be about $15,000.
Well, that makes a lot more sense economically but is not really how it was portrayed in the first place. Luck, not guarantee.
3. The million free views are part of an ad faucet that halves every week. Next week the bounty will be 500K free views. The number of views offered depend on the open space created by unaudited screens. Meaning the screens have been activated but not reviewed yet. This creates millions of hours in free advertising every year.
That would make again more sense than initially but I wonder how are you going to count views on smart tv, you would have to start converting a lot of things from views to seconds of advertising to price per locations, a lot, a lot of work which is going to cost a lot in manpower.
4. Now, if an ad can't get 51% of the approval of randomly chosen individuals then not showing the ad is a net win for the community. We want to make a system were communities vote to see the ads they want to see. Those ads get better conversion, they get good feedback and they connect with the community.
This is the most interesting part as it actually covers one of the works I did in university, the effectiveness, or better say lack of them in repeatedly concentrical targeted campaigns directed at the same group of customers, that's why Google is so good at it, it doesn't show just you what you like but what you are interested in buying, if left alone people will select ads from the products they already like and bought mostly. But that's of no concern to the protocol, it's up to those that make the campaigns.
As for the 51% attack, I still believe is the weak point of this, if people can band together to bring a rating down like it happened with RobinHood they will do this over a decentralized app also, don't underestimate the power of trolls and 4chan. The same way they have bought down yelp reviews to a minimum for locations they have never once eaten at the same they could do with those ads or target publishers.
But, rather than the problem itself, for which I will not change my opinion that is real, I'm actually more interested in someone coming with a decentralized defense mechanism that is not simply relying on the good nature of human beings.
For too long we have been seeign the ads the corporations and governments wanted us to see but now we have a say in the matter.
I have a remote and adblock for that
1. My analysis listed 10 smart tvs hived together to form a billboard. You are cutting about 10 middlemen: agencies, designers (if you purchase a premade template or go for a classified ad), printing, instalation crew, maintenance crew, government fees, etc. Our cost structure is closer to adwords than to the status quo for billboards.
3. The smart tv have a projection software that counts the number of displays. Ads are displayed in 30-second increments. This software is essentially a java-based wallet with an html browser redirecting to a new ad every 30 seconds. These metrics are circulated across all the screens of an area. So the system is decentralized on a city-wide basis.
4. Customers have a different type of wallet than projectors. We call this wallet a explorer app.A customer can update his or her "explore list" to include or block ads that are relevant/irrelevant to his/her interests. In this fashion, we bring the contextual relevance of Google ads to offline ads. Since ad space is limited and multiple customers could have different taste, the projection software would show the ad most relevant to the "group" and for this it needs to ad up interests. This is described in the paper.People who make campaigns can retro fit likes and the manifested interest matrix of the community to correlate them to products people might want to purchase. Please notice advertisers will get this info for free, with the full consent of the customer and with no knowledge of the customer identity.
Bitfari does not relay in the good nature of human beings to publish and ad as much as Bitcoin relays in the good nature of miners to maintain the network. If a group of auditors decides to collude and block an ad, the agency can deploy the same ad to another location (and at no cost to them since no bounty was paid). The entiry city would have to block the ad for it not to be shown.
But you don't have an adblock for real-world ads. Bitfari is that adblock.