No posts for ... how long? This may not be the start of a flood, but I can't resist marking today's silly Horizon episode, Human v2.0.
It started badly. The benefits of Moore's law, we heard, were felt first by computer gamers. while Moore himself had expected his law to apply indefinitely. Simply wrong on both counts. Moore first set out his "law" in 1965, rather before games became the driving force of the development of computers. And when he did, he also said,
Certainly over the short term this rate can be expected to continue, if not to increase. Over the longer term, the rate of increase is a bit more uncertain, although there is no reason to believe it will not remain nearly constant for at least 10 years.
Which hardly sound like "indefinitely".
It didn't get much better thereafter. The usual nonesense. Oh look, we can hook a monkey up so it can control a robot arm by thinking ... and we've recorded the death throes of 10,000 neurons in fifteen years ... so we understand the brain now. Eh? Isn't there maybe, just maybe, a bit more to it than that?
The only dissent they bothered finding was someone who also thinks the singularity isn't far away, but is frightened of it. And a hook into a bizarre subplot about the Unabomber.
What about creativity, which is surely a fundamental part of human intelligence? Why on earth would a computer which matches the raw bit-shifting power of the brain suddenly be intelligent? How accurate are estimates of that bit-shifting power anyway? Isn't the model of the brain as a load of interconnected neurons, the model on which those estimates are based, a bit of an idealisation? Will these intelligences fall in love? HTake drugs? allucinate?
Where was the argument, surely absolutely fundamental to this programme, about whether the brain is Turing equivalent or not?
Most importantly, what screwed up idea of human potential gives the title "Human v2.0" (I'll ignore the crapness of that name) to a machine designed to replace humans rather than to an actual human with tool-amplified capabilities?
As my old man said in a comment to a post of mine a few years ago,
Unfortunately, the energy going into trying to make computers replace the thinking acts of people is diverting from the MUCH more valuable work of supporting them.
This surely applies at all leves of use. A program which "analyses" a problem does so within the boundaries set by the designers and programmers which may not be understood by the user. Unfortunately, the computer produces and "answer" which is frequently believed to be correct. Correct here means truly representative of the behaviour of the thing being modelled.
Which brings us inevitably to Doug Engelbart and the augmentation of human intellect. Engelbart who also, it seems, noticed the trend labelled "Moore's Law" some time before Moore wrote about it.
Comments