a thoughtful web.
Good ideas and conversation. No ads, no tracking.   Login or Take a Tour!
comment by blackbootz

That's a good point. But what will replace economists of the past? Better economists?





kleinbl00  ·  2924 days ago  ·  link  ·  

The true utility of economics is that it's a systems approach to value. The true failing of economics is you have to select boundary conditions for any system you care to model, and those boundaries are political as fuck. Arguing that mass unemployment is a logical and pure outcome of economic dislocations and efficiencies selects the masses as outside the boundary of the system; arguing that failing industries must be subsidized in order to promote full employment does not precisely model the efficiency of the industry.

If you read Freakinomics and ignore/forgive their ridiculous "everyone is stupid but us" slant, you begin to appreciate that the whole of the book is an exercise in listing externalities. My guess is you're going to have giant data-inclusive models that some people will worship like gods and others will tear down like false idols. I mean, look at this shit. I only know about it because an investment pundit I follow is already front-loading it full of crap:

Negative market informational surface, bitch.

Are we gonna know more? Are we gonna know less? I'd argue that the numbers are going to dance to the tune of whomever policy's in vogue at the moment and, same as it ever was, they're going to have motives.

veen  ·  2924 days ago  ·  link  ·  

I thought that graph style looks familiar so I did some digging. Turns out Quid is essentially a layer over the Gephi open-source package. Which is free unlike the $25,000 a year Quid asks for their, uhm, added value. I mean, I made this degree centrality graph in an afternoon, what exactly does the program do that the open package can't?

(Extra bonus: note all the floating dots? Those are data points that have no relation to the rest of the network. Either he didn't bother or Quid doesn't even let him remove those. )

user-inactivated  ·  2924 days ago  ·  link  ·  

I leave predicting the future to people who do TED talks. I'd like to think something more like designing a distributed system in computing than a science, where you start with the properties you want it to have and then figure out a protocol that has them. Then everyone knows it's all a game and they can make up a new game if the one they're playing turns out not to be very good.

Something something flying cars and robot butlers.