I think I see why you're confused. The curse was an effective way to kick off the revolt, yes. Not any way to start an earthquake, though. Reality doesn't work that way.

OK, let's squash some of this nonsense right now.

I never believed the 2010 Haiti Erthquake was caused by a voodoo curse, and I'm astonished that anyone interpreted that post in that way. What I found anthropologically interesting is that something like Robertson's "satanic" invocation seems actually to have taken place. Not actually "satanic", but within Robertson's impoverished terms of reference that's about the only way he could describe an invocation of the loa.

I believe, and have repeatedly said, that the supposed "scientific consensus" on CAGW is not a conspiracy but an error cascade. I think most scientists are honestly trying to do right, but have been overly credulous about data and models that have been (and continue to be) fraudulently manipulated by a tiny minority of them. Those of you who think this makes me some sort of nut are going to have some explaining to do when measured GAT drops out of the bottom of the IPCC's 95% confidence band, which looks set to happen before the end of 2014.

I might reply to some of these other questions at more length, but these two deserved to be dispatched immediately

Yes, Kuhn was full of horse puckey. Not only doesn't his book describe science outside of physics at all well, it doesn't even correctly describe 20th-century physics, its ostensible paradigm (using the word correctly now) case.

Years ago I wrote a more detailed takedown in Brother, can you Paradigm?

The only amplification I'd write today is that the shifts between large theoretical models generally (and contrary to Kuhn's claims) go smoothly in physics because test by correct prediction of experimental results is so difficult to argue with. The soft sciences have more trouble setting up repeatable experiments, so it's easier for people to hold on to broken theoretical models.

Since he quoted me, I have replied to the report on Spadaro's article at Imprimatur me!

That brought back a fond memory: when I was at DGC in the mid-80's, working on the fascinating but doomed "DeviOS" project, I used to bring in my Walkperson for late-night debugging runs. I would find a nice gap in the AM band and let the minicomputer generate my musical accompaniment. Definitely felt plugged in.

What do you mean "very quickly"? By "very quickly", do you mean "Mr. President, we believe this to be a Russian first strike, and you must decide in the next 5 minutes if we are going to retaliate" quick?

Cost of Windows is much less to an OEM. I've heard figures as low as $10 were leaked for the really large OEMs, though that may be after the crapware subsidy rather than before.

Actually, a lot of hospitals use voice recognition systems for dictation. There are usually still medical transcriptionists in the loop, but they have at least a partial text to work from instead of just doing straight-ahead transcription. I have heard some of them claim that working from partial text is slower than just doing what they used to do, because the interfaces provided for editing the text are slower than their blinding typing speeds.