get stressed about helping their kids to be successful), and in this
book, they talked about why "intrinsic motivation" is the best and
maybe also the main way for people to be successful. (And, arguably,
it works the in a very similar way for rhesus monkeys!)
Now, the book also points out that not every task is hugely
intrinsically enjoyable right off the bat. E.g. a parent hopes that
their kid will learn that cleaning their room is *a good idea*, but
initially they generally have to *make* the kid clean the room.
The way the kid becomes "successful" is a combination of learning to
do things they "have to do" and "doing what they want to do!" --
Intrinsic motivation was broken down along three axes:
1. Autonomy (people like to be able to make their own minds up about
what to do).
2. Competence (people like it when they can know they are doing well).
3. Connections (people like to do activities that enhance their
connections with other people).
I'll (subjectively) rank a couple different environments that I've had
experience with on these axes:
MATH GRAD SCHOOL.
1. Autonomy. Relatively low -- at least when it comes to required
first-year courses. Some *successful* students may have taken a more
autonomous approach to these required courses (e.g. getting into them
as advanced undergrads when they were optional).
2. Competence. Relatively high -- tasks are very structured, and
since it it math, after all, there are "right answers". Furthermore,
there are some *social* ways to measure competance ("Am I doing as
well as my classmates?"), and, indeed, establishing "competance", for
many, may *require* intense social interaction (peers, classes).
3. Connections. Optional -- if you want to work *with* people, you
can establish connections that way. If you don't work with people,
you end up feeling alienated. (If you go that route, you also lose
important measures and contributors to competance, see above.)
HACKING OPEN SOURCE PROJECTS
1. Autonomy. Very high -- you pick the project, you select the
modifications or project goals; you can build on the things others
have produced. You often don't even need to ask permission: you can
just fork. Also, you may not even depend on anyone else for advice
apart from what you get from reading the manual (although advice can
certainly be helpful).
2. Competance. High -- the code either compiles or it doesn't, takes
a million years to run, or it doesn't. There are, again, social
measures of competence. ("Hacking the good hack.)
3. Connections. It depends. Certainly it is possible to develop weak
connections with people in help forums and so on. If your hacking
leads you to become involved in project leadership, your connections
with other project leaders may be strong (but often long distance!).
If you happen to find people to work with locally, connections as
strong as any local hobbyist, artist, or other research/development
connections can form.
HACKING HDM/METACOMMONS STUFF
1. Autonomy. Very high. There are some set outlines of the project,
but many ways to go within it, and no one "right answer" (although
there are plenty of measures for competence, see below).
2. Competence. High. Since HDM and/or the metacommons is such a
general project, one could competently work on it *many* different
ways. (As a programmer, as a linguist, as a mathematician, as an
economist, etc.) It is rather difficult to work to a high level of
competence in *all* of these domains, but it is nevertheless fairly
easy to measure the *degree* of competence that one has in balancing
the different areas (along how many dimensions is the project
progressing?). Also, precisely because the task bears on many
different areas, any individual is likely to be fairly competent in at
least one of them! And finally, there are some pretty clear
ends-oriented assessments (is it benefitting anyone -- say, students
or researchers?).
3. Connections. *Potentially* high, but with some risk. Again,
because it is a general project, you can make connections with people
in many different areas, you can talk about the project with just
about anyone, and, with some skill, find a way to connect it with
things that matter to just about anyone. Perhaps one runs the risk of
"being spread too thin"... but, if so, that is an "occupational
hazard" that one can work to avoid. Furthermore, unless one works
very hard to connect, connections across disciplinary boundaries may
be hard to establish or relatively superficial.
HACKING HDM IN MATH GRAD SCHOOL (REMIX!)
1. Autonomy. High -- but at the cost of being an alien, being
alienated, and not doing well in the program. Because of this
"outlaw" factor, the perceived degree of autonomy may be higher than
it really is.
2. Competence. High -- because one is doing things that no one else
is doing, it seems like one is breaking new ground. Besides, to a
comparatively large degree one sets one own standards for "success".
3. Connections. Medium. One is sacrificing one set of connections
(the school in-group) for another (other remotely-situated hackers).
From this sketch, I think we can see why I've found HDM consistently
motivating over the last 5-6 years of my life, but why I have also
been very frustrated with the work at times.
I think that I am an *extremely* intrinsically motivated person in
general, and over the years I've been especially excited about math,
anthropology, and writing. However, there have been a lot of kinks to
work out in these areas too. -- To what extent am I doing the activity
because I enjoy it, versus doing the activity because it will give me
some sort of extrinsic reward? It worked great when "fun" was a more
explicit part of the picture; I was less able to find the "fun" in
activities that (for whatever reason) that I didn't find sufficiently
"autonomous".
0 comments:
Post a Comment