Glen wrote:
I like the idea of informal data gathering:
http://www.randomiseme.org/
"RCTs are used by scientists to find out what works best. Here, you can
create a trial on anything you wish, and participate in trials set up by
other people."
I like the *idea* as well, however reviewing the active and completed
trials leaves me rather underwhelmed.
I tried to go through the tutorial and see if I could ascertain (or even
intuit) how well their methodology should work. It didn't seem very
promising. In particular, it doesn't look like they offer "laymen"
enough guidance in what makes for a good sample set or "enforcing" it?
The style of questions asked also seem likely to suffer various
selection biases?
For example, among their completed trials the first two (1-how to
remember things; 2-does complementing th ebarista improve my chances of
free coffee) were declared "completed" yet the first one consisted of a
sample set of 8 and the second a sample set of 11 ? The first one
apparently onlyr *required 1 participant?* while the second required 100
(but as declared complete with 11?).
I guess I was only mildly surprised (in a disappointing way) at the
inanity of the questions to date. As I remember the early days of
Kickstarter, most of the projects were pretty inane but as some really
good examples came on, more followed. I find the "Whitehouse.gov"
petitions nearly equally inane, so there is good company here. Maybe it
is just "internet culture"? Perhaps it is too early to evaluate this?
These all seem like good tools *in principle*, I wonder what it takes to
make them good tools *in practice*? I suppose an easy, trite answer
is, *good participation*, and maybe it really is that simple? Kind of
like (presumably) democracy, the free market, and innovation.
hmmmm?
- Steve
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
