Monday, September 2, 2013

Taleb's Principle and the Weirdo Principle

This may look off-topic... and again, maybe it isn't  ;)

Perhaps most of you have heard of the "Peter Principle", which basically states that "Employees tend to rise to their level of incompetence."

Well, I want to state a new principle not too dissimilar from it, which I will call "Taleb's Principle", and stated in one sentence says something like this: "Being too far ahead in a field is the same as being too far behind, no one will listen to you".

I thought of this as applied to research, more specifically, when you have a really good idea that is ahead of its time, no one will listen to it (or to you) because they don't understand it. The other researchers on that field will just think it is silly, or the idea is completely dumb, or they will even be offended by it.
If you're lucky, then maybe they will think the idea is obvious, which means that your idea is just in the "boundary" between being too far ahead and being understandable, and if you invest in it and insist a bit more, it might just start to be understood.

Imagine you would travel back in time to medieval Europe and try to teach the "experts" of that time to make some sort of modern device... you don't need to tell them about smartphones allowing people in different continents to talk and see each other, no... something simple... huhhh, let's say a steam locomotive. You could try to explain to them how a steam locomotive would work, and propose them to build one. If you were lucky, they would think you're a crackpot and ignore you... if not, you'd get burned at the stake as being some sort of warlock or witch.
Imagine now that you would travel forward in time to 200 years from now, it would be the other way around... most likely you would think they were nuts and "weird", and everybody else around you would think you had some sort of "cognitive deficit" (how can you not understand such simple stuff as teleportation and hive-minds?). At best they would ignore whatever you had to say, at worse they you put you in a "special needs school".

When you do research, or some kind of work that involves peer-review, being too far behind "the pack" is bad because most of the other guys working in that field have already had the good ideas, so there isn't much more left for you to do except "pick up the crumbs". The problem is, if you're too far ahead from "the pack" then it is also bad, because the rest of the peers (or experts) in that field will not understand what you're talking about and just ignore you... or even "burn you at the stake".


Like all good ideas, this is not completely new, and it is a re-hash of something someone else has talked about, namely Nassim Taleb complains about this in his book Antifragile.


Some fields of research suffer from this effect more than others.
In an area of research where it is easy to check whether or not a new proposal is valid or good, the effect of the Taleb's Principle is very small. I think that a good example of this is IT in general. If you have a new idea that you think works, just build/code it and show the rest of the world the results.
Unfortunately, there is sub-field in IT where checking if a new idea is valid or not is an extremely difficult task, and that is (yes, you guessed it) Concurrency!
Verifying that a concurrent data structure or algorithm is valid is a very difficult and time consuming task. This creates an incentive for "experts" to not even bother to look at your code because it would take them too long to prove it is wrong, and they already (think they) know that.

I think that the best is either to be the first in a field of research (which eliminates the "peer-review" problem), or be smart enough (but not too smart) to make new things that can be still understood by the experts in the field.




Another somewhat related idea is the "Weirdo Principle" which states that "Anyone that is greatly smarter than you, will look like a weirdo to you".

Humans have difficulty judging intelligence (in its various facets), and can only do so if it is a bit above their own capabilities. For example, if someone gives me a complex problem/code to write up and it takes me a few days to come up with a solution, and then someone else comes up with a similar solution in just a few hours, then I would think that person is really smart. And if they do it on several different occasions for different types of problems, then I would start to realize that that person is much more intelligent than I am (at least when it comes to solving those kind of problems).

What if someone else solves the same kind of problem, also in a few hours but using a solution I don't understand? Then what would I call that person?
What would _you_ call a person that does something you don't understand?
Whatever it is, I bet it's not going to be "smart"... most likely it will be something similar to "weird", or even more pejorative.

A corollary of the "Weirdo Principle" is that if you don't know anyone that you think is "weird", then most likely a lot of people that you know think that _you_ are weird. And if you know a lot of "weird" people, then maybe it means that you're in an environment with really smart people... that's not so bad, is it?


Btw, if you don't like this post, then send flame mail to Miguel and Hugo because they were the ones that gave me the idea to post it up  :)

No comments:

Post a Comment