4 ms·
Edge computing. Basically shifting off of cloud onto separate peer-to-peer connections. Faster, more secure, more distributed, and no middleman. Think 1990s/mi
by dabockster 9y ago
Edge computing.
Basically shifting off of cloud onto separate peer-to-peer connections. Faster, more secure, more distributed, and no middleman. Think 1990s/mid 2000s but no servers, just client to client.
Soon everyone will want a home server.
- amelius 9y agoHow would we implement a search-engine using P2P?
- zeeveener 9y agoMY_PC: "Who is hosting a webpage with the following information?" My 1000 connected peers: "Here are 500 more peers who say they do" Proceed to witness extreme bandwidth usage every query.
- amelius 9y agoThat only works if the computer you're connecting to has an index of the entire web. It will probably be more like this: - So you're looking for "kitten videos"? Well, you can ask computer X for words starting with "k", then ask computer Y for words starting with "v", and then I'll perform a logical AND operation to get your final results. But how we make this as fast as Google is beyond me. And also don't forget that we have to factor in the problem of security (some nodes may be sending in wrong results, e.g. spam).
- dabockster 9y agoObviously, edge computing won't work for every use case. But it could get us off this obsession with tying everything into a centralized server that you have no control over. /rant
- Toboe 9y agoFor example there is the YaCy project http://www.yacy.net/en/ http://www.yacy.net/en/