<html>
<head>
<meta content="text/html; charset=ISO-8859-1"
http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">On 05/12/2012 22:45, Jeff Davis wrote:<br>
</div>
<blockquote
cite="mid:CAHUTwkP6ye6obt_BWtuWHEZPAvBGfDevdAMW0w5w1Pz8HSQU9w@mail.gmail.com"
type="cite">Global warming due to anthropogenic increase in CO2 is
real, but what does that ***really*** mean? The issue has been
thoroughly politicized. Climate models have long been crap, and I
think, still are. The left says "We're doomed! We're doomed!"
and the right says "What problem?, fill 'er with high test."<br>
</blockquote>
<br>
Hanging around climate scientists (in a broad sense) is interesting.
The local Oxford consensus is something along the lines of: <br>
<blockquote>"Humans are definitely changing the climate in worrying
ways, but the models can we make are fairly crappy due to
foundational reasons that are unlikely to ever go away (but we
still want bigger computers! Because they are cool!) But the *big*
hole in our knowledge is the mapping climate -> weather ->
human impact. We simply do not have any good ways of estimating
that. And then politicians and activists take our dear research
and make it *stupid*. Oh, and geoengineering looks like it could
work... which is *scary*, because it is going to be the mother of
all governance and safety problems - now you have climate change
with some of the stupid people in charge."<br>
</blockquote>
Of course, at least in Europe farmers are largely decoupled from
actual climate: given the current subsidy situation and the apparent
impossibility of dislodging it, you can do well by not producing
anything. <br>
<br>
The problem for engineers is that engineering works when you get to
build a clean system that optimizes certain things. But if you need
to interface with messy existing systems that change, behave
irrationally or even adversarially, then it becomes *much* harder.
It often fails as a discipline because it produces too brittle
solutions in the face of this kind of mess. Which doesn't mean that
neat solutions to particular problems are not transformative and
desirable. It is just that, as soon as you scale them up to a big
system it will start to interact with the mess.<br>
<br>
<br>
<pre class="moz-signature" cols="72">--
Anders Sandberg,
Future of Humanity Institute
Oxford Martin School
Faculty of Philosophy
Oxford University </pre>
</body>
</html>