[OT] State of the internet

Timothy Miller gandalf at doctorTimothyMiller.com
Wed Oct 3 13:04:27 EDT 2012

On Oct 3, 2012, at 9:31 AM, Bob Sneidar wrote:

> I'd like to point out something I noticed when I was young. People tend to put a lot of stock in what has happened in the past, and then apply it to predict what will happen in the future. From stock prices to global warming, everyone seems to think that trends will continue in a linear fashion because we have seen a tiny segment of something as it progresses over time, and then we make the mistake of presuming it will continue to do so in the same manner. 
> This is quite odd...

Well, here we go, off topic. Possibly of interest to Bob. He started it. Perhaps of interest to no one else. My bad. 

Maybe the presumption that regression lines go on to infinity represents a universal human cognitive bias. Humans are not rational creatures, though we like to think we are. We are barely capable of rationality. Lots of good research and popular-press books on that topic these days. My favorite is You Are Not So Smart. The author is a journalist, not a scientist, but I think he got his facts right. It's an easy and informative read.

I read an amusing article in Behavior and Brain Sciences. The gist: People are so bad at reasoning that many scientists wonder why they evolved to do it at all. If reasoning is usually wrong, then it would not likely have any reproductive value. One theory is that people don't reason to be right. They reason to win arguments for the sake of increasing their social status. That would explain 99% of political and religious conversations, wouldn't it?

Don't know if the full text is available on line. The abstract is here:


But you say, "We must be able to reason or we couldn't write LC scripts."

That's true. Under certain narrow circumstances people do reason correctly. If you must solve the same type of puzzle repeatedly and you get rapid feedback about whether you got it right or not, then you can learn to reason correctly about that particular kind of puzzle. If the LC script does what you intended, you know you got it right. You find out pretty quick.

This is why emergency room nurses usually get the diagnosis right.

Conversely this is why talking heads on TV and experts commenting on the future state of the internet get it right about as often as a monkey with a dart board.



More information about the Use-livecode mailing list