2 ms·
there is lots of buzz around this not just to come up with the next use-case for a unicorn but also to provide break out of the costs associated with securing a
by DyslexicAtheist 10y ago
there is lots of buzz around this not just to come up with the next use-case for a unicorn but also to provide break out of the costs associated with securing and maintaining the infrastructure.
server less computing would in applications where this makes sense reduce the latency down to what is dictated by the last mile. on the other hand there is a huge push towards going mobile in many industries (thanks to LTE, LTE-A), but these wireless protocols do horribly poor routing P2P traffic. (this is a problem that operators are just about getting to grips with in their fixed networks).
When it comes to use-cases I think we have some realistic opportunities if an industry is big or desperate enough to also buy data from individuals (rather than a broker). I ran recruitment firm the last few years and a recruiter would always do anything to get their hands on a larger pool of data. If you give them a tiny new cloud service to search for "niche skill candidates" they will pay for it provided they can add additional people to their database. And a distressed candidates would sure rather sell their data directly to the company (earning a few cents along the way) rather than paying linkedin to host their profile and suck up all data.
the bigger challenge (than creating a p2p network) is how to structure services on top, the article mentions btc miners as the profit model for bitcoin. it is for me perfectly logical to have a company with a centralized service/buisness model that builds data applications on top of the decentralized p2p paradigm (most trackers are not really decentralized and the total decentralized bootstrapping of the p2p network itself is also a problem which isn't totally solved afaik).