From what I see on day 1, our project, currently named “Wireless Dead Drops” (
I hope we find a better name, though “Dead Drops” sounds kind of cool, we could have a cool logo with a skull and crossbones in a giant raindrop EDIT: as Hubert showed me, the name and a logo very similar to what I had in mind are already used by another (very cool, as it appears) project, so we’ll definitely have to find something else) can be grossly resumed in this way:
Suppose that you want to communicate with other people, possibly far away from me, but using only short-range radios. Hopefully the environment is crowded enough so that you will meet people regularily. Make everyone carry a radio! Then every time you meet somebody, let your radios exchange all the messages they carry, until your message reaches its final addressee. You can also put ‘dead drops’ (hence the name), radios that are not carried by anyone but just lying in strategic positions where people are known to pass on a regular basis.
This raises a lot of problems: everybody might be carrying your message, but only one person must be able to see it, therefore you have to encrypt it. Once a message gets delivered, the other carriers will not know, and therefore you have to spread a new message to tell radios to stop spreading it. To do that, messages will have to be identified. How do you affect them a single hash that will not conflict with other messages? How do radios communicate. Do we have to write a protocol (that could look like a simplified version of IP for instance)? When two radios meet, we don’t want them to be on all the time, so how do we know they will indeed synchronize their content? Sure, we could do it manually, but it loses a lot of interest. Since they are going to be using timestamps to date the messages, they have to synchronize their internal clocks, but how can we know which radio is closer to the real time? All this has to be solved while constantly watching energy costs, which must be down to almost zero if we want radios (especially the dead drops) to last for a decent enough period of time; and while keeping in mind that cheap is always best, since this will only work if there is a large amount of radios in circulation.
I can see a simple solution for some of those problems, while I have not the slightest idea of how to approach some others. In a few weeks I will be returning to this post with more hindsight and wille probably smile at what I wrote.
It’s gonna be exciting.