[ExI] Model uncertainty (Was: LIGO)

Anders anders at aleph.se
Thu Nov 10 00:38:19 UTC 2016


On 2016-11-09 19:51, William Flynn Wallace wrote:
>
> I am still trying to get my head around that: how could we have seen 
> two crazy-unlikely events in just a few weeks?  My view of the cosmos 
> must be serious flawed.  COOL!
>
> This expresses my feeling exactly.  I think somewhere along the line I 
> lost contact with the human race and am seriously out of contact with 
> reality.  I am still stunned.
>

In fact, this is an interesting development. Pollsters and information 
markets missed the UK election, Brexit and the US election. The models 
are clearly wrong. Even if one accepted that the 15% chance of Trump on 
Monday evening was true and we saw a 15% probability event, the swathe 
of other recent polling failures demonstrate that something important 
has changed.

Generally, I think there is both an epistemic uncertainty about how to 
poll current people properly, and a more meaty uncertainty about what is 
going on politically. I have recently shifted away from my previous 
model that people had a broken epistemology because of networked media 
to a model that what we are seeing is more a tribal defense of core 
values (Haidt's explanation: 
http://www.the-american-interest.com/2016/07/10/when-and-why-nationalism-beats-globalism/ 
).

Now, realizing that one's model is not correct and trying to fix it is 
an uneasy but exciting place. Especially since it might mean one should 
change strategy about a lot of things.

-- 
Dr Anders Sandberg
Future of Humanity Institute
Oxford Martin School
Oxford University

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20161110/837437da/attachment.html>


More information about the extropy-chat mailing list