5 ms·
I understand your premise of stored value not being realized until the future. However, it isn't some sort of universal rule. There is value in a lot of things,
by invalidbasis 12y ago
I understand your premise of stored value not being realized until the future. However, it isn't some sort of universal rule. There is value in a lot of things, even in failure, and loss; you are only considering the monetary value in your statement. The possibilities of immeasurable value, that which aren't strictly monetary, is all lost in saying, "stored value is bizarre."
If stored value is truly a bizarre concept, then society should completely reevaluate schools, and learning in general. Even this discussion; because what is the point, or value? Is storing any nuance of thoughts pointless? You have a highly cyclical argument, favoring Singularity, and not allowing for any other perspective.
Your attempt to laud your position and derail Austrian Economics with the "Hauling Gold to Mars" example is insulting and disservices your own ability to make a valid point. Already in place, society has a concept that dismisses your absurdity, I.e., "Contracts;" which fully eliminate the need to haul physical resources or commodity around and merely show them on paper. Not too dissimilar to the blockchain.
From a statistical standpoint, statistics are irrelevant when they are put into practice. For they either ARE or ARE NOT and DO or DO NOT come into effect. They are a summation of the past, not its predictor.
In your very own argument, "You can't predict the future." But then go on to argue people would "be idiots" for investing in "consumption" instead of "productive" things or businesses. In a single paragraph you've argued against your own theory. People can't predict the future, but they can somehow tell what is productive versus not? Hah! I think the last business cycle is evidence enough.
Your idea isn't completely without merit in its attempts to reward persons for making good predictions. But again we already have similar systems for that purpose already in place in the stock markets.
- skilesare 12y agoWe will have to see. It is a theory and I'm hoping to build some real world and computer model experiments. A lot of my ideas of what would be good to do are based on Chris Alexander's idea of wholeness in which we actually have a better than 50/50 chance of telling what a 'good' transformation is. If it is less than 50/50 then very little will work. I don't mean to insult. I need to do more study of what the austrian's really put forward. It is my least favorite of the schools and I haven't spent as much time learning about it as I should.
- invalidbasis 12y agoFair enough. Peace be with you. And good luck in your work.