The original note, as written at the time
In an alternate future this era will retain its position as the era of the most visible rapid transformations in tech. The growth exponential will continue as per the graph but the change steps will be mushed out into longer, major, distinguished steps.
Ongoing development
In an alternate future this era retains its position as the era of the most visible rapid transformations in technology. The growth continues as per the graph, but the change steps are mushed out into longer, major, distinguished steps.
The distinction is between the shape of the curve and the shape of the experience, and the two are not the same.
An exponential measured in capability can be experienced as a series of plateaus with sudden jumps, because the jumps are what get noticed. But the same growth can also be experienced as longer, slower, larger transitions — one big change every twenty years rather than a new thing every eighteen months — and the total amount of change could be identical.
That is your point, and it is a better forecast than either the accelerationists or the sceptics are making. Both sides argue about the curve. Almost nobody argues about the granularity.
Granularity has consequences. A society that changes in small frequent steps develops institutions for absorbing small changes; one that changes in large rare steps develops institutions for surviving shocks. Those are different cultures, and which one you get is decided by how the technology is deployed rather than by how fast it improves.
The prediction also implies something about this collection. Ideas written down and left for years look eccentric at the time and obvious later, and the interval is a function of granularity, not of quality.
The curve is not the experience. The steps are.