Hacker News Reader

I've taken it upon myself to shit on anything that hasn't been already implemented.

http://en.wikipedia.org/wiki/OLSR

That's my current proposal for routing protocol. How it works at scale, I have no fucking idea, but I'll manually write the redistribution if I have to at the aggregation nodes.

View on HN

I would say don't even try, http://en.wikipedia.org/wiki/B.A.T.M.A.N. is a better bet. OSLR will never work at the kind of scale that this project is aiming at. It's meant for networks on the order of hundreds of nodes.

Also, you still need an addressing system. I was mentioning one of a hundred different problems that I've yet to see one person in the organization qualified to solve. If Van Jacobsen announced tomorrow he was building a worldwide mesh network I'd drop everything else and start contributing, but until then I see a lot of people flailing around and going nowhere.

Reply on news.ycombinator.com