Everything You Wanted to Know about
Social Engineering -- But Were Afraid to Ask...
The Scientific Method
Science News magazine often gives me a chuckle.
Its reporters are totally addicted to a story line that runs
something like "Scientists were really, super sure that
thus and such was gospel truth, until this researcher turned
up evidence that says they were wrong, wrong, wrong." For
example, the Jan. 15, 2000 issue carries the story, "Oxidized
plutonium reaches a higher state." It reveals that "For
many years, scientists thought this…was plutonium dioxide…Now
a team of researchers has taken the luster out of this description.
They've found that plutonium dioxide reacts very slowly with
water and oxygen to form higher oxide phases…The additional
compounds 'were there all along, but no one realized it,' says
John M. Haschke, a chemist and consultant from Waco, Texas."
Welcome to reality. Er, make that, welcome to the concept
that we perceive reality "through a glass dimly," as
the Bible puts it and as the modern science of neurology emphasizes.
How do scientists try to come to as good an understanding
of reality as possible? We use what is known as the "scientific
method." The English philosopher Francis Bacon (who lived
from 1561 to 1626) first conceptualized this.
Basically, here is how the scientific method works.
1) Come up with a hypothesis, for example, "no dog can
live to be over 30."
2) Come up with a way to test your hypothesis. This must include
a way that someone could prove your hypothesis false if the right
data were to turn up (i.e., it must be falsifible). An example
of a hypothesis that is not falsifible is "the entire Universe
resides only in my mind." The trouble with that concept
is that no matter what experiment you run, you can always say,
"Of course! I imagined the result, so therefore the Universe
is all in my head." A falsifiable hypothesis is "No
dog can live past age 30." If anyone can prove that a dog
has lived past age 30, then the hypothesis would be proven false.
3) Gather and evaluate data. OK, we're working on the dog question.
We get the age at death of 1000 dogs and they are all less than
30. Does this prove the hypothesis? No. However, if you get enough
dog data, scientists will start calling this a theory instead
of just a hypothesis. What if someone produces a dog that they
say is age 31? Does this disprove the hypothesis? No, because
that dog might not really be age 31. Someone could have made
a mistake on the dogs birth date. Scientists might devise experiments
in which they raise large numbers of dogs from birth and look
for ones that they can be really sure passed age 30. On the other
hand, such a dog might have somehow gotten identities switched
with a younger animal. Welcome to the concept of "experimental
error."
4) Scientist try where feasible to run "double blind"
experiments. Even an honest researcher can be fooled by the subconscious
mind into making mistakes that end up "proving" his
or her hypothesis. If the people taking care of the dogs believe
they can't live past age 30, they may make mistakes in their
care that lead to them dying at a younger age than otherwise
might be possible. It is not unusual for the people running an
experiment to be tricked by the head researcher into thinking
they are testing a totally different hypothesis. That is a "single
blind." If the experimental subjects are humans, the researchers
may create a second blind ("double blind," so both
subjects and people conducting the experiment don't know what
is going on). They could be misled about what the experiment
is supposed to test. Or the researchers and subjects could be
using something fake such as sugar pills instead of a medicine
supposedly being tested. Ooh, is it possible that I am a researcher
and happyhacker.org is a scientific experiment? Oh, no, trust
me. Honest!
5) Publish the results. Other scientists will try to find faults
in your data gathering and they might repeat your experiments.
Yes, sometimes this process uncovers scientific fraud, where
a scientist purposely lies about his or her experiments. Or further
experiments might discover an error in how the first experiment
was conducted, for example dog identities getting swapped.
6) So, let's say that many experiments have been conducted and
everyone believes the theory that no dog can get older than age
30. Does this turn the theory into fact? Uh, uh, maybe somewhere
a Chihuahua is hiding who could prove them all wrong.
If you adhere to the true scientific method, you have to face
a disturbing possibility. According to the scientific method,
you can never prove anything to be true. You can only prove things
to be false, and at best develop theories that seem to agree
with every experiment and every bit of data you can throw at
them.
So next time someone tries to incite you to commit crime against
someone on the basis of news articles, email or IRC chats, remember
to smirk and shake your head. If you absolutely insist on uncritically
believing what you read, I have a story for you. Brian Martin
eats kittens. Honest! Would I lie to you? Now go hack his Attrition.org web site.
It is absolutely urgent that you register your protest against
him this way. Please be especially careful to hack the parts
of his site where he says I smoke crack.
More on social engineering --->
Back to the index of "Everything You
Wanted to Know About Social Engineering -- But Were Afraid to
Ask --->
