4 ms·

> 2050, that super futuristic year, is only 29 years out, so it is exactly the same as predicting what the world would look like today back in 1992 The problem

by typest 5y ago

> 2050, that super futuristic year, is only 29 years out, so it is exactly the same as predicting what the world would look like today back in 1992 The problem with this analysis is that the rate of change of the world is accelerating; the world is changing faster now than it was in 1992. Fundamentally, this is driven by the fact that all the changes from the past cause further changes, in an exponential process. Thus a straightforward linear estimate of the future will underestimate how much things will change in the future.

CuriouslyC 5y ago
Things have been accelerating in a sigmoidal manner, which looks exponential in the early curve, but then plateaus as it approaches a limit. It's hard to say how far along we in the curve now, but given that we're bumping up against the limits of silicon semiconductors now, I'd say perhaps farther than you might expect.
kibwen 5y ago
> The problem with this analysis is that the rate of change of the world is accelerating; the world is changing faster now than it was in 1992. Fundamentally, this is driven by the fact that all the changes from the past cause further changes, in an exponential process Any accelerating trend can't be sustained indefinitely. We know that the accelerating rate of progress is because of computers (which reduce the feedback loops on various processes which then help make better computers, and so on). When we reach a point where computers stop getting better, the rate of progress will stabilize. We've already observed stagnant clock speeds for over a decade. There are still worthy investments to be made, like in miniaturization, affordability, and heterogeneity, but without a breakthrough as monumental as the transistor we have no reason to assume that acceleration will continue.

4 ms·

> 2050, that super futuristic year, is only 29 years out, so it is exactly the same as predicting what the world would look like today back in 1992 The problem

by typest 5y ago

CuriouslyC 5y ago
Things have been accelerating in a sigmoidal manner, which looks exponential in the early curve, but then plateaus as it approaches a limit. It's hard to say how far along we in the curve now, but given that we're bumping up against the limits of silicon semiconductors now, I'd say perhaps farther than you might expect.
kibwen 5y ago
> The problem with this analysis is that the rate of change of the world is accelerating; the world is changing faster now than it was in 1992. Fundamentally, this is driven by the fact that all the changes from the past cause further changes, in an exponential process Any accelerating trend can't be sustained indefinitely. We know that the accelerating rate of progress is because of computers (which reduce the feedback loops on various processes which then help make better computers, and so on). When we reach a point where computers stop getting better, the rate of progress will stabilize. We've already observed stagnant clock speeds for over a decade. There are still worthy investments to be made, like in miniaturization, affordability, and heterogeneity, but without a breakthrough as monumental as the transistor we have no reason to assume that acceleration will continue.