[ExI] intelligence and generalization

Adrian Tymes atymes at gmail.com
Mon Jan 14 20:03:00 UTC 2019

On Mon, Jan 14, 2019 at 11:35 AM Adrian Tymes <atymes at gmail.com> wrote:
> such as learning basics of science
> and critical thought that could be applied to any scientific field,
> picking up details of biology/geology/physics/et cetera later on as
> relevant to specific investigations, or more handwavium - in my work,
> anyway - practices to study other fields broadly and quickly

To clarify this bit: what are the components of mastery, and how might
they be taught or automated?

"I've seen this before." - database lookup from fragmentary
information.  (One will most often not have the exact title of a work,
so the emphasis is on multiple ways to find the information based on
the limited data one will typically start with.)  Kind of like search
engines, but learning the field includes learning how to use search
engines specific to that field.  A database of scientific concepts is
not the same thing as a database of musical riffs, in the same sense
that a professor of science is not the same thing as an experienced
composer of music (although the two could be aspects of the same human
being).  The database would likely have to generate its own indexes
(and analyze existing entries as new types of indexes are introduced),
as humans who enter examples will most likely not think of every
possible connection when providing an example; this is unlikely to be
perfect, and learning to compensate for these imperfections (again:
learning how to use the field's search engines) would be part of
learning a field.

"If you want to do this thing, here's what you need to worry about."
or "Alright, I'm doing this?  Then to do it quickly, efficiently, and
(a much higher chance of) correctly, I start with X, Y, and Z." -
broadly applicable requirements planning for a field, augmented with
details (which can be looked up on the fly) for a specific project.
(Building a skyscraper is not the same as building a house, but many
of the processes are similar or near-identical, and the differences
could be well-documented if an expert construction manager cared to do

Specific motor functions and patterns of thought can be guided with
appropriate AR software, though one would need to get familiar with
that class of software (including being able to competently select
said software).  The class would likely differ by field (formulating
scientific experiments and composing music differing enough that their
software would feel different).

What other elements would be needed, to be said to truly master a
topic?  (Some would suggest the appearance of confidence, but there
are non-confident people who do quite well in a field, and there are
confident but incompetent types.)

More information about the extropy-chat mailing list