5 ms·

I was at a classmate of yours at Brown, although a few years younger. I like to play the prediction game too! Maybe it comes from all those hours in the Sun Lab

I was at a classmate of yours at Brown, although a few years younger. I like to play the prediction game too! Maybe it comes from all those hours in the Sun Lab. I like most of your predictions, but I think you're at least half-wrong on the cryptocurrency side; the blockchain is real innovation. Do you guys know at Joyent that your data hasn't been modified, deleted, lost or added to and when? I bet you don't. The amount of data risk any modern company takes on is almost criminal, just like you refer to in your security predictions. The blockchain technologies let you solve that problem, among many others. I'm not predicting a brilliant year for Bitcoin; I think there's a lot of building to be done on the currency side especially, but I just don't see the case for sniping at it too much, especially while transaction rates stay well in excess of $10mmUSD / day in the middle of falling prices. I'm going to predict some notable traction in smaller non-libertarian markets for Bitcoin by the end of 2015 in return. :)

bcantrill 12y ago
I don't dispute that there is real innovation with the blockchain -- just that the level of innovation cannot alone justify the value of or interest in cryptocurrency. To answer your question: linked timestamping[1][2] does, in fact, solve this problem -- and solved this problem long before the blockchain. Again, not disputing that the blockchain doesn't allow one to prove tamper-proof transaction systems or that this isn't an important problem -- just that it's not the first technology to do it, isn't the only way to do it, and doesn't alone pay the freight for the level of interest in cryptocurrency. [1] http://en.wikipedia.org/wiki/Linked_timestamping http://en.wikipedia.org/wiki/Linked_timestamping [2] http://dspace.utlib.ee/dspace/bitstream/handle/10062/15188/Truu_Ahto.pdf?sequence=1 http://dspace.utlib.ee/dspace/bitstream/handle/10062/15188/T...
- drcode 12y ago
  It seems odd that you would claim that a system requiring a centralized "time stamping authority" is equivalent in functionality to a decentralized mechanism like a blockchain. As for saying that "cryptocurrencies are a laughingstock"... can you quantify that claim in concrete terms using a date and a price range prediction for that date?

bcantrill 12y ago

I don't dispute that there is real innovation with the blockchain -- just that the level of innovation cannot alone justify the value of or interest in cryptocurrency. To answer your question: linked timestamping[1][2] does, in fact, solve this problem -- and solved this problem long before the blockchain. Again, not disputing that the blockchain doesn't allow one to prove tamper-proof transaction systems or that this isn't an important problem -- just that it's not the first technology to do it, isn't the only way to do it, and doesn't alone pay the freight for the level of interest in cryptocurrency. [1] http://en.wikipedia.org/wiki/Linked_timestamping http://en.wikipedia.org/wiki/Linked_timestamping [2] http://dspace.utlib.ee/dspace/bitstream/handle/10062/15188/Truu_Ahto.pdf?sequence=1 http://dspace.utlib.ee/dspace/bitstream/handle/10062/15188/T...

drcode 12y ago
It seems odd that you would claim that a system requiring a centralized "time stamping authority" is equivalent in functionality to a decentralized mechanism like a blockchain. As for saying that "cryptocurrencies are a laughingstock"... can you quantify that claim in concrete terms using a date and a price range prediction for that date?