You are viewing a single comment's thread from:
RE: EOS - The next Dan Larimer Thing
Exactly, you'd have to have a very restricted language for a secure smart contracts or at least start with one and build up slowly.
I think that smart contracts are a dead end, in terms of their robustness and resilience. It simply is not smart to be running so much potentially malicious software in a distributed way like this
As for Dan being smarter then Vitalik well... maybe some of the time.. ;)
yep, I rate both guys as about on par genius level...
As regards to the execution platform, in fact Bitcoin has one of these in it, it was how they added multisig and a few other things. But the scripting language is not turing complete. The reason why a turing complete scripting language is a bad idea, is that you can't write AI without a turing complete language, and do we really want an AI running around on the blockchain? Hm?
Oh that's interesting. I did some programming but I skipped a lot of the more interesting CS stuff since I was on a math major...
I think I would not want to AI to be in the blockchain but outside for analysis... machine learning is quite amazing at what it can do!
I think that as GPUs accelerating databases and AI algorithms continues to develop, that there will be a place for this kind of thing, but I don't think it will be directly tied to our financial ledgers, like you say, 'outside the chain'. It's just a continuing evolution of the trading robot, and trading robots have been a boon to finance, by increasing liquidity without inflation.
I did receive marketing letters from Harry Dent & Co. There was an interesting project: a program which sifted through social media feeds looking for stock market predictions and it would check who would be right and rank them according to prediction accuracy. Such a bot could be pretty neat here on SteemIt as well.
If you look at the #fake-news which is mostly mainstream media, in my informed opinion, track record is a very interesting thing and I wouldn't mind making a business out of it.
I've heard of that program, pretty cool concept. Voting on steem is actually a type of prediction market - how much profit you can make on curation depends on you correctly determining beforehand that a post is going to get upvoted hard (and that's also why it was so important to get rid of the log curve or whatever exact formula it was that compounded the vote power of whales...
A system like you describe would be very interesting for predicting all kinds of things. Humans are more precient than most people realise, that's precisely what it means to be an entrepreneur - to use prescience to recognise an opportunity for profit based on patchy evidence. Prescience is basically a matter of a person having taken in enough data that their subconscious brain does the sherlock on the puzzle and sketches out a possible description of what is not yet visible (the future). I visualise it like a tree graph that grows endlessly new branches into the future, and we can guess at what branches are possible, and picking the correct one, this is the essence of what evolution works towards - changing before the conditions are untenable without the change.