Tuesday, June 18, 2013

Tuesday Links


  • Above shows European car sales in May for the last ten years.  The graph is not zero-scaled.  Apparently, this last May is a twenty year low.
  • I suppose this is standard operating procedure.  Still, I can't help wondering what this kind of thing does to the character of our leaders.
  • David Brooks argues that the mind is not the brain.  Looks like he didn't make it to GF 2045 or his column would have been yet more interesting.
  • A quarter to a half of all bird species are threatened by climate change.
  • Speculation that the NSA is actually getting the content of all US phone calls, not just the metadata.  I still am far from satisfied that we have the full picture here.  I no longer believe anything the administration says on this subject, and I'm not sure how far to think Snowden is exaggerating for effect.

5 comments:

sunbeam said...

Hmmm...

Has what used to be called "POTS" been digitized yet? I'd think this might be in the minority of communications now as opposed to cell phones.

Plus you still have to get a myriad of local telecoms to play ball. Not that they couldn't be "urged" to in general, but you might occasionally run into an operator with a wild hair.

I've thought that calls going over land lines were safer (though not ones that are long distance) from a privacy standpoint, but it really doesn't have to work that way.

sunbeam said...

Oh yeah, as long as my browser is still on your site:

http://newscenter.berkeley.edu/2011/09/22/brain-movies/

This ran a few years ago. The title is self-explanatory "Scientists use brain imaging to reveal the movies in our mind."

I thought this would have been big news, but apparently it didn't strike anyone as a big deal.

It really didn't look like much.

Here is a direct link of images reconstructed directly from the mind:

http://gizmodo.com/5843117/scientists-reconstruct-video-clips-from-brain-activity

Not sure of their methodology, it sounds like they really have to take data from the individual before it is much use, not like they have found the "primitive" pattern that all minds use (a debate and something to be solved by neuroscience).

Still interesting. And I think I could write a real mindscrew scene for a movie using a machine like this, even one that worked in a realistic manner.

"Don't think of your handler. That's pretty hard isn't it? When I say that, it's not easy to have that face pop into your head.

Tell you what, we are going to introduce a little tranquilizer, just a smidgin, into that IV in your arm, 'kkkk?

Don't feel guilty. Willpower isn't going to keep you from revealing what needs to come out. Actually lets look at a little bit of what's going on in your head. I'm sure you're curious about the rig you are hooked up to.

Let me assure you we aren't going to drill you skull with a power drill like the Iraqis do. And we are going to do horrible prole things to you like Rodney Graner and Lynndie England would.

We just want to know what you know. You'll be none the worse for wear after we are done.

Let's get get started, 'kkkk?"

Then our "interrogator" takes a big swill from his coffee cup, like the supervisor dude in Office Space.

Burk Braun said...

I can't believe you link to David Brooks- how lame. He is still pining away for the soul, apparently, as do all well-behaved conservative columnists.

Greg said...

Stuart - you might like a rest after your conference effort (thanks again), but for a more mainstream look at the future, how about Jorgen Randers' 2052: A Global Forecast for the Next Forty Years?

Randers was one of the authors of Limits to Growth. His new book is an exercise in straight extrapolation from official data. He extrapolates the declining trend in fertility and concludes that global population will peak around 2040 (!). The working-age population will peak a little earlier, of course.

He extrapolates the declining trend in productivity growth; together with population, that means the economy will stop growing about 2060. But because fixing weather damage and replacing formerly free ecosystem services like fresh water supply, fending off would-be immigrants, and looking after the old will take more of total production, household consumption peaks around 2045--earlier, in the former OECD.

He extrapolates the energy efficiency trend, and putting that together with the economy trend concludes that energy consumption will peak around 2035. Extrapolating the trend in renewable energy growth, he concludes renewables will be about 40% of the total supply by 2052; total CO2 emissions will peak in 2030.

Randers barely touches on the "automation threat" - he thinks there will be plenty of work for the soon-to-shrink workforce. It just won't be very well paid work, since it will mostly be in manual services and construction and maintenance.

Randers's data is good, his trends seem like the important ones, and he identifies the processes behind the trends correctly. (If anything, he is a little optimistic in not allowing for the likelihood that something goes badly wrong.) But the conclusion is a bit surprising, and he doesn't spend much time discussing how each of the trends could vary, or what happens to the end result.

I'd be interested in your opinion of the book.

Cyrus said...

That graph is dangerously close to "how to lie with statistics." Thanks for pointing out the non zero scale...