[ExI] Calling all Autodidacts

Adrian Tymes atymes at gmail.com
Sat Aug 27 18:27:50 UTC 2011

Insofar as an autodidact is one who engages in self-directed learning - which,
IMO, is something everyone who might ever read this post should be capable

On Sat, Aug 27, 2011 at 3:10 AM, Emlyn <emlynoregan at gmail.com> wrote:
> 1. Where do you lie on the specific / general continuum?

I tend to lean towards general.  Some things I do better than others, but I
have had to do a wide variety of things over the years.

> If there are
> areas you are more comfortable with, what are they?

Engineering, especially Web software, but that's just because I've been
working in that field for over a decade, and my degrees are in Computer

> How much
> difference do you find between your well known areas (perhaps where
> you have a degree?) and those you don’t know?

Depends on the field.  Here are some of the closer ones:

Web software vs. other software: not that much of a difference.  The
timing and input/output flows are different, but the same general principles
can be applied.

Vs. other forms of engineering: ...well, it's still equations and models, and
the basic principles still apply (design what you intend to do, examine it
for flaws, then try to build it, and finally test it).  These days, the design
can usually be simulated in software.  I apparently did quite well at
nanomechanics design, when I dabbled in it, because I was able to set
up a better modelling environment prior to trying to actually fabricate the
design.  I've been exploring rocket design recently, and the first challenge
is to find what I don't know.  The answer seems to lie more on the finance
end than the actual engineering - which brings me to...

Vs. business & other finance planning: there is a famed synchronicity
between software & finances.  Assuming the data you get is reliable, and
the models you have actually hold true, financial planning boils down to
simple spreadsheets.  The problem is, those assumptions often aren't
true.  You need to find ways to tell which data and which models are more
likely to be correct, and which ones are wishful thinking or - sadly, not too
uncommonly - malicious lies intended to make you fail (usually to enhance
the success of others, and that's how you can pick up on these: check the
motivations of all parties involved).

Vs. fiction writing: the writing process (outline, then fill in the pieces)
translates, though again, I may just be familiar with it from long practice.
I'm only technically a published author, so far, but I've used writing as a
form of entertainment for quite some time.

Vs. martial arts: the ultimate goal of some martial arts is to let one act like
a robot, whose body is programmed to succeed in all combat situations.
And, well, programming takes a lot of metaphors from software.  Granted,
there's also the physical toning so that the body is capable of carrying out
the instructions - these days, I only practice as exercise - but that's getting
away from the intellectual learning part.

Of course, there are also fields which are way different - but even there, my
first instinct is to try to map the fields to what I know.

> 2. What’s your motivation / how do you initiate? I think my motivator
> is questions in service of a project. Is that true for you? If not,
> what’s your thing?

That is often my motivator.  If I am not doing something which requires me
to learn X, then why am I learning X?  "For enjoyment" is not a valid answer:
if I am doing something for enjoyment that involves learning X, then I am
"doing something which requires me to learn X", and that something
shapes the specific part of X that I seek to learn.  For example, I doubt I
will learn painting just for the sake of learning painting - rather, there might
be something I wish to paint, or some specific thing (an event, a job, or
similar) for which knowledge of painting would be useful, and the thing in
question tells me which aspects of painting are most important to learn at
that time (brush on canvas is not the same as painted pottery, for example;
I have done one of the latter and zero of the former at this time).

> 3. When you know your target, what kind of process do you use to get
> there? Are you aware of it, or is it largely intuitive?

I am aware of it, but it is also partly intuitive.  My first step is
to google on
whatever terms I know of the field, often winding up on Wikipedia or a certain
few other wikis.  (For each subject X they are listed high on, they have
earned their high listing by being exactly the kind of resource one goes to
when trying to find general information about X.)

At the same time, I try to formulate whatever problem (or other thing, but it's
usually a specific problem) I am looking into, breaking it into parts and steps.
This helps me come up with more search terms, and helps me get a grip on
the problem.

I iterate on this, trying to come up with a general solution while
having specific
sections of unknown "how do I do this?"es.  I keep aware of dependencies
between the sections, so if I find out that a given section is in fact
(which happens), I can revise said general solution, or discard it and come up
with another.  Alternately, if I uncover enough information that I seem to have
a complete solution, I look for ways it can go wrong, then - if
practical - try it
out.  Based on the results of that experiment, I may discover that some of
what I thought I knew is incorrect, or I may have solved the problem that I
sought to solve in the first place.

> 4. How do you solve the “I don’t know what I don’t know” problem?

Categorize it so I can at least come up with terms that describe areas that
include it, then look for papers or articles (including, again, Wikipedia)
describing a complete process to solve problems in the given area.  If it is a
process that completely solves the problem I wish to solve, then I only need
to figure out each step in that process.  As a second step, that initial search
may point me to experts in the field - so, maybe ask some of them.
(Especially university professors, whose job it is to spread this sort of

Just this week, I had a good example of this: I wished to figure out the cost
to launch something into Low Earth Orbit, because I am trying to see if an
idea I had to launch at low cost is viable.  So, what are all the components
that factor into that?  Turns out, very few people know this - it's apparently
not even known well by the companies that currently do this.  But there is
a paper - apparently not available online - which breaks that down into its
components.  Some (quite a bit) of personal negotiation eventually got me
the paper.  Turns out, there are five components; the actual hardware &
fuel are a small part of one of them.  Of the five, three can be streamlined
and shared between multiple launches per year (if you have multiple
launches per year, which I was assuming), while the other two simply do
not apply to the launch model I am considering.

Once I believe I have a complete list, I also go for reality checks from people
I know who know something about the field.  For example: Spike, the above
seems relevant to your interests (and, again, it looks like few people in the
industry have a complete breakdown of costs).  If you'd like to check what I
found, both for your own benefit and so you can let me know if it seems to be
a complete list based on your own experience, please let me know offlist.
(Sorry for the wordiness of that, but again, this is to illustrate to
others how &
why this gets done.)

> 5. Do you record your progress? What sort of tools do you use?

I take notes, but I haven't found any really good tools for the overall flow.
Some tools are useful for certain specific parts.  For example, Excel rocks
for doing complex arithmetic progressions, such as financial projections.
But the main tool is simply something to write with, so I can externalize my
memory so I don't have to keep all parts of a project in my head at the same
time (especially when I am trying to focus on any small part of the task).  I
most often use Notepad (more complex presentation is simply not useful to
me, since I am the only audience, and features like autocorrect can be of
negative utility since I often use terms no common dictionary has heard of;
ironically, the autocorrect in the application I am using to type this does not
know the term "autocorrect").

> 6. Do you talk to other people much, or confine yourself to written materials?

Hell yes I talk to other people.  Written materials can only get me so far, and
the set of written materials I can find on my own is substantially less.

That said, written materials are my preferred first pass, so I don't waste the
time of experts asking for info that is obvious to anyone in the field.  More
importantly from a purely selfish point of view, such material can be found with
a simple google, getting me my answer much faster.

This twinned benefit is recognized by many - and apparently unknown to many
more.  Thus you have sites like http://lmgtfy.com/ , which uses patronization
to try to drill this point into the heads of people who do not yet
understand it.
(To those who don't click the link, the acronym means "Let Me Google That
For You".  It gives you an URL to an animation of how one can google for a
specific topic, then asks, "Was that so hard?")

Again, though, googling is only useful as a first resort, not a final
resort in all
cases.  If it yields the full result, great.  If not, it at least
helps me phrase the
question (possibly by giving me terms that those in the field use to describe
things), and may suggest who I can ask.  Sometimes my first google will
yield nothing useful except new terms for me to google on - and looking up
those terms, or terms found when looking those up, gets me the full answer.

> 7. Do you use esoteric knowledge sources, like academic journals, or
> is it mostly Google? Books? Blogs? Wikipedia? Anything else?

In my experience, academic journal articles are mostly of use after I have a
question of relevance to the specific article; exceptions are rare.  Google to
Wikipedia (in case the term I think of first is mentioned in an article, but not
an article title itself) is often my first path, but only my first -
and not always
my first.  Googling lets me see if there are far more relevant sources for
general knowledge on a topic, which there sometimes are - rarely, but
enough to be worth checking.

Thus I go to online books, blogs, and other such sources.  Granted, I bias
heavily toward information that is freely available online - because the utility
is just that high.  Googling for a subject that turns out to be general
knowledge, then reading a Wikipedia article, has sometimes solved
problems for me in under a minute flat.  I can not travel from my house to
the local library at anything approaching that speed (safety and legalities
aside, I currently possess no transport capable of the requisite acceleration
and deceleration, nor is it practical to obtain such just for this trip), to say
nothing of time that would be invested in sorting through books there.

Only after an initial online search has failed to satisfactorily answer my
questions, do I consider other sources.  Generally, tracking down physical
media requires that the question have very strong importance, and good
confidence that the information can be found in the specific physical media.
Asking other people is often more convenient - either ones I will be in the
company of anyway, or via email, or even via telephone (though I generally
attempt initial contact via email, as a courtesy to strangers whose schedules
I do not know).

> 8. Do you incorporate structured learning materials? MIT
> OpenCourseWare? Actual enrolment in courses of study? Or do you find
> structured courses and materials intolerable?

Unless there happen to be online tutorials that specifically address my
question (which there sometimes are), the investment to pursue these often
exceeds the value of the answer.  These seem better for preparatory learning,
when one expects - in advance - that one will have a number of questions in a
field, so that one will be better able to answer them when they arise.  (Thus
my enrolling in the AI class recently mentioned on this list.)

BTW: "I will attempt to work in field X" automatically means "I expect to have
a number of questions in field X" for most X of interest these days.  That
appears to be the most common justification for taking these courses,
whether by students new to the adult work force in general, or those entering
a new field during their career (as an extension of their current employment,
or retraining because their old specialty has become obsolete).

> 9. Are there tools you use to help? Mindmapping? Diary/Commonplace
> Book? Notebooks? Webpages? Blogs? … Where does this fail you, what
> would be better?

See question 5.  The questions I usually focus on are small enough, that
none of these tools (except keeping notes) significantly help me organize
my thoughts.  I suspect that software that followed the same process I do
might be of use - but it would require automation of a lot of things no one
has yet come up with software to do in a generalized sense.  For instance,
asking detailed questions on a subject the software does not already
understand (in the sense that software can understand a topic): it is
unlikely this will be accomplished without having software that can pass
the Turing Test, as measured by then-contemporary organizations.

> 10. When do you fail, and why?

In almost all cases, it is because the expected resources that would be
invested to answer the question, exceed the expected value of the answer.
Almost any information can be had if one spends enough time/money/etc.
on it...but some information would require so much of this, that calling it
"impossible" (which equates to "would take infinite resources to discover";
note that no one has or can get infinite resources) is but a rounding error.

In turn, most of the time, this is because the expected value of the answer
(at least to me) is trivial.  For example: most of pop culture.  Do I care about
the latest happenings on a given reality show?  In almost all cases, no.
Even the less-than-a-minute to google the show's title would exceed the
value, to me, of the answer.

The second most common case is that the expected expense is too high.
For example: I know of a theoretical experiment that might be able to create
wormholes.  It would cost at least a billion dollars.  I do not at this time
possess any means of directing that amount of resources to be spent on this
question - which is more a question of influence than personal bank account.
That this could be used to generate a transportation infrastructure, the fees
from which could recover the money in a few years, is not a significant factor.
Of far more relevance is that "might": the odds of success are quite low.
Thus, it is not worth doing at this time.  Most likely, it will remain
such until
knowledge of the physics involved improves enough to greatly increase the
odds of success - assuming they do not disprove the theory entirely first,
which is quite possible.  But if they do not, then said improved knowledge
will likely come hand in hand with better engineering capability in this field,
significantly reducing the expected cost at the same time.

The information not being available to the methods I use is simply a special
case of expense.  I am capable of using other methods, but the costs
increase.  (For example: obtaining and reading a printed book has a
relatively high cost - mainly in time - versus obtaining and reading the same
information online, if it is available online without needing another person to
give me access.  If it is not available online and only available in print, then
obtaining and reading the printed copy is the minimum cost of access to the
information.  Some information is worth this.)

Very rarely is there information that is, in fact, literally
impossible for me to
obtain.  (Information in this class is usually obscured by history.  For
example, a 100% accurate diary of the life of Jesus Christ.  Even that
assumes time travel is actually impossible, and not merely really difficult.)

More information about the extropy-chat mailing list