From: T (firstname.lastname@example.org)
Date: Tue Nov 22 2005 - 18:06:51 MST
Chris Capel wrote:
>On 11/22/05, H C <email@example.com> wrote:
>>Success is limited only by obsession and ambition. Intelligence speeds
>>everything up, but I don't believe that it is a limiting factor in many
>I wouldn't say that either intelligence or ambition--or focus or
>organization or anything else--are limiting factors. For that to be
>so, an increase in one factor wouldn't lead to an overall greater
>success past a point. But I don't think this holds. While a complete
>lack of ambition may make it so that a greater intelligence doesn't
>yield much more success, it's more of a NBNS thing--necessary, but not
>sufficient--than it is a limiting thing. For reasonable levels of
>ambition and intelligence and other factors, I think an increase in
>one would always increase lon-term success.
>Here's another quality that I think is necessary for accomplishment.
>Single-mindedness. One of my own main limiting factors is a curiousity
>that's unable to sit for very long on any one subject. I work very
>hard on a large number of different things. It seems to me that this
>particular affliction is quite common among the intelligent and is at
>least as responsible for lost accomplishment as a lack of ambition.
>I think it probably has something to do with want to be holistic, with
>wanting to have a good grasp on the wider principles of everything.
>And part of it is being smart enough to see how terrible one's
>toolchain is and wanting to work on improving it instead of working on
>using it to make something--spending time on "meta" aspects of one's
Good point. I was quite struck by Eliezer's fear of "reverting to the
mean". He clearly meant this in a way which captured both "staying
ambitious" and also "living up to the potential of his IQ" yet those are
two different things. It appears that most people agree with me that
they are not one and the same, and that for whatever reason, most of the
high IQ people alive today are not ferociously ambitious in that sense.
I wonder, thought, whether restlessness in this area is not common. I do
believe that IQ is something real, and that it is a limiting factor in
understanding some things. I think of IQ as being the average height of
a map of intellectual abilities.
Sometimes, I wonder whether we will not need the assistance of
pre-singularity superintelligences to understand the meaning of
post-singularity ones. In many ways computers today are already beyond
the comprehension of a single person. The workings of a computer rely on
logic, computer programming, but also materials science and engineering,
even social science in the form of the success of the Internet to
function as they do. While their IQ is low, they are already amazing. We
use computers to build computers already.
It is an interesting discussion of personal responsibility, which feeds
into Eliezer's desire not to revert to the mean, to what degree we
should concentrate on the Singularity. (E here stands in for many other
people who see this as a Very Important Thing). I am less ... personal
about my responsibilities. If a superintelligence does not maintain our
safety or pursue our goals, then who are we really to constrain it to
things which are within our limited understanding?
I suspect that the Singularity will be less quick that most here
imagine, undergoing a progression of improvement. I think we will see
functioning robots before a distributed intelligence, for example,
because people will solve little problems in non-general ways and cobble
something together with it.
People always overestimate the impact of technology in the short term,
and underestimate it in the long term. For me, the transition period
will be the most fascinating thing to be alive for, if I am that lucky.
Perhaps I will live another 60 years, and perhaps that will be enough
time. That would really be something to see, even if it's only the first
stages of change. Beyond that, I am I dead, and what does it matter to
me if the student surpasses the master?
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:53 MDT