Quote from: Richy_T on November 03, 2015, 09:03:12 PM

Quote from: JorgeStolfi on November 03, 2015, 05:15:37 PM

Well, I did learn *a lot* of interesting and useful things from the bitcoin project, much of it from this thread. I don't know whether it was worth the investment; but one never knows, even afterwards, what will pay out and what will turn out to have been wasted...

Given that you are a professor in the computing field, I think this falls within your remit.

Yeah, I could say that if someone wanted me to account for the time I spend on it.

Quote

Though you are a bit more hands-on than one might expect

Actually, most of what I learned by watching bitcoin was not about computing, but about other things. Two years ago I did not know about order books, spreads, arbitrage, TA, ETF. I had only a vague idea of what libertarianism was about, and never heard the terms "ancap", "fiat", "gold bug", "day trader". I learned about the money velocity equation and why a good currency must be inflationary. And much more.

One important thing I learned is the rule "past prices do not affect future prices". Actually I first had to read tons of posts that claimed 1000% growth per year guaranteed, backed by straight red lines on exponential price charts. (Remember those? I haven't seen many of them in the last year...). Only to learn, by reading and by experiment, that they were basically all bogus. Trends in ordinary stocks may be meaningful, if they are tied to slow-varying physical phenomena liek consumer demand and overall economic activity. For purely speculative assets, like gold and bitcoin, trends are at best useful only for describing the past, but mostly useless for predicting the future. The only model that seems to work is the "log-Brownian" model, that says that the price at any date in the future is the price today, plus or minus a random (unpredictable) relative change.