Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

"A classic is something everybody wants to have read, but no one wants to read." --Mark Twain

It's amazing how much low-hanging fruit there is in classics. When I read Darwin's "Origin of Species", I found that a huge amount of it was devoted to this thing called the "Knight-Darwin Law" which was extremely important to Darwin but which seemingly vanished out of all knowledge around the turn of the 20th century, as if no-one was actually reading the book since then. Then I wrote a couple papers related to that Law, which ended up being my most successful work so far.



For anybody else not familiar, the Knight-Darwin Law is the idea that plants do not self-fertilize exclusively (many plants have both male and female sex organs). Darwin goes on at length in On The Origin of Species about the mechanisms that plants employ to avoid self fertilization.

My own take is that this makes a lot of sense. Male and female sex organs are pretty sophisticated adaptations. Features which are that complicated generally don’t evolve unless they are useful and there are easier ways for an organism to clone itself than self fertilization.


Darwin actually articulates a much-more-precise law (which serves as a cornerstone of his whole book). I've argued in my papers that the KDL is in fact an infinite-graph-theoretical statement (remarkable because infinite graph theory was not a mainstream "thing" until well after Darwin).

Let G be the graph of all organisms (past, present and future), with an edge directed from u to v if and only if u is a biological parent of v. The spirit of the KDL is: "G does not contain any infinite directed path consisting entirely of vertices with only one parent". Or equivalently: "Every infinite directed path in G necessarily contains a vertex with two parents."

For its time, the above statement is astonishingly mathematically sophisticated. If anything, Darwin is still greatly underrated.


Darwin is fantastic. Not only did he bootstrap the field of evolutionary biology despite his lack of access to modern genetics, but he also wrote about it beautifully. On the Origin of Species is still a great read and a lucid explanation of a difficult subject. He's a great person to strive to emulate.


I was particularly impressed by the part about pigeons: familiar examples used to introduce new ideas.


I never thought pigeon breeding could be an interesting subject until I read Darwin.


It reads like you are adding the mathematical sophistication to an intuitive idea. An "an infinite directed path consisting entirely of vertices with only one parent" is simply a chain.


Compare König's lemma [1] ("every infinite tree contains a vertex of infinite degree or an infinite simple path"), often considered the first example of infinite graph theory. It wasn't articulated until 1927, almost half a century after Darwin's death.

[1] https://en.wikipedia.org/wiki/K%C3%B6nig%27s_lemma


It won't impress mathematicians until it's stated in such a way that a layperson would be completely boggled. /s


It's also a fairly interesting read, for that matter.


> Then I wrote a couple papers related to that Law, which ended up being my most successful work so far.

I have a serious hard time marking progress in the modern 'science' field.

There are too many people afraid of Industry, so they get a PhD, and to graduate a PhD you need to prove you are correct.

That means too often finding 'conclusions' in 'data' that you 'studied'. The incentive system in science is warping the scientific method to proving your idea by any means necessary. This is not an opinion, its been documented that many PhD studies could not be replicated. (psychology is particularly bad)

In Industry, this isnt true. If it doesnt work, your customer knows. Industry is my indicator of progress. There is no lying about data or stretching conclusions. When you leave the math of chemistry, these conclusions become more and more opinionated.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: