Hacker News Reader

Comment by frognumber | Hacker News Reader

Mechanisms like this are why I hate evolution as taught in K-12. The basic model (random mutations + selection of the fittest + sexual intermixing) is not just oversimplified, but oversimplified to the point of being wrong on many levels.

(It's not just the way DNA can jump between species; there's are many other mechanisms, such as evidence of rates of mutation changing during major changes like ice ages, dynamics of sexual partner selection, etc.).

Combined, those have deep implications, but outside of the scope of a HN post.

View on HN

I learned relatively recently what survival of the fittest actually means. That the mutation which was capable enough to adapt to the environment survived. Before, I use to think that species deliberately created better offsprings for new environments.

We are result of random mutations. Couldn't find what it was like before random mutations or how life started creating it's clones.

Are they random, though?

What does "random" even mean? How intelligent is evolution? Is it stupid and random? Does it have the same level of direction as a single-celled organism? As an insect? As a mammal? That's still an open question.

There's a spectrum of hypothesis. To give a few:

- They are guided by an intelligent creator

- They are random and stupid

- There is intelligence built into the system which results in mechanisms like changing mutation rates during periods of crisis, sexual selection / reproduction, genetic crosstalk, DNA encoding, etc.

It makes a big difference. We can compare different optimization algorithms we run on computers (various gradient descent, simulated annealing, genetic algorithms, etc), as well as different encodings we run those over. The results aren't the same.

That has profound implication philosophically, theologically, and scientifically.

The oversimplified school version is not helpful in that it potentially leads to the wrong conclusions.

In science we work with the best we can. At this point we have zero proof of inteligent DNA, omniscient creators or directors. This paper is good, but it doesn't add any breakthrough to what we know: virus-like particles taking DNA from one individual to another, ocasionaly jumping between species. At random.

Also, in science, we work from null hypothesis and build on it. You can't work towards an hypothesis like "there is a supra-inteligence". You work from "there's nothing", analyze what you see and propose the simplest and more probable explanation. At the current state proposing a supra-inteligence raises more scientific questions than answers.

All the scientific evidence we have at this moment points toward blind, random, stupid and un-directed evolution as the best explanation, thus we should ethically teach that. If you want to hypothesize otherwise, bring proof and observations that can't be explained with the current model.

Philosophically and specially theologically you can say whatever you want, as long as the framework is internally consistent. For example, the "We live in a simulation" hypothesis is quite consistent, but we can't prove it, for or against. Nobody stops you from researching about inteligent creators, inteligent DNA, etc. But you shouldn't teach your hypothesis, like we don't talk kids about the "simulation hypothesis" <https://en.wikipedia.org/wiki/Simulation_hypothesis> or "zoo hypothesis"<https://en.wikipedia.org/wiki/Zoo_hypothesis>.

> At this point we have zero proof of inteligent DNA, omniscient creators or directors

You're deeply misunderstanding what I wrote, in part by confusing "intelligence" for "sentience." My washing machine has "intelligence," as per the sticker on the front. So does an ant. Intelligence is a complex and often emergent phenomenon, and a colony of ants has /much/ more intelligence than a single ant.

Evolution very obviously has some intelligence (and much more than my toaster). The key question is what type and how much. That has _deep_ implications, including for fields like evolutionary psychology.

And I do teach kids about the simulation hypothesis and similar. They're useful philosophical explorations. Why don't you? Do you think your child will break if exposed to different ideas and perspectives? I haven't found that to be the case.

Ehhh. This part of how teaching works. When a 6-year-old asks about infinity being a number, it's not usually a good idea to bring up one-point compactification of the complex plane.

> When a 6-year-old asks about infinity being a number, it's not usually a good idea to bring up one-point compactification of the complex plane.

Why?

Bruner recommended doing so, and I've always found it to work well, as have others who I know who have done so.

I hedged with "usually". Feel out your audience when you teach.

Plus, honestly, one-point compactification of C is less intuitive than two point of R (where you'd have + and - infinity), so I stand by it being not great to bring up. There are many models here ("infinity's not a number", "there is one, unsigned infinity", "there are two infinities", "there are transfinite numbers", ...) and picking the right one to bring up at the right time, I argue, is part of good teaching. Not just "time to expose this person to the definition that's the "fanciest" I've personally got experience with".

I usually go with "all of the above" rather than "the fanciest." With my six-year-old, I did:

- Countable / uncountable infinity (but I didn't go beyond that)

- Positive / negative infinity, optionally with NaN

- One-point compactification of the real numbers (which is pretty simple -- draw a circle, and map points on the line to the circle)

I don't recall if I ever did the one-point of C, but likely I did.

The major point, actually, is to understand that there are many ways one can extend numbers, not all compatible with each other, to give bigger sets of numbers. The specifics weren't too important.

It's a very good idea. With as much visualisation and ELI5 (sic) explanation included as possible!

Child has no way to communicate that it reached it's understanding capacity do why you would you assume it's capacity for it?

Most of the school education is simplified to a level. The higher the level the more complicated/detailed it gets.

In an early chemistry class, the teacher was happy to observe that atoms bond in certain ways, with a certain number of slots available per atom (depending on element), presented various molecular diagrams, and mention that while some combinations don't seem to occur in nature, some have been created artificially.

Some kids asked about molecules they had just imagined. They were reminded that some molecules don't seem to exist, or told why what they were proposing didn't fit the model.

I asked somethnng like, "If the model's rules allow that atoms would be able bond in ways that are not observed, doesn't that mean that the model is incomplete or incorrect?".

The teacher was not pleased, and became dismissive: "The model is complete, and correct.".

Reply on news.ycombinator.com