[getoffmycase]

If I had a nickel for every terrible biology to computer processes analogy I saw on this website since joining, I’d probably have enough money to buy a beer. Which is what I feel I need after I see something like this.

Biology does not act like a computer. You cannot reduce biology to an operating system

[dang]

Ok, but can you please make your substantive points informatively? Just putting someone else or the community down doesn't help. It just makes the thread shallow and dyspeptic.

If you know more than others do, that's great, but then please share some of what you know, so the rest of us can learn (edit: like you did here! https://news.ycombinator.com/item?id=36330052) If you don't want to do that, that's fine too, but in that case the thing to do is remind oneself that the internet is more or less wrong about everything and move on.

I know it's tempting to leave an empty negative comment to relieve oneself of annoyance, but this is the worst choice, at least on HN. It not only isn't the curious conversation we're looking for, it actively impedes it.

p.s. You're a good HN commenter generally - so thanks for that!

https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...

[TeMPOraL]

> Biology does not act like a computer. You cannot reduce biology to an operating system

Nobody is doing that. Analogies are drawing rough outlines in the thought-space[0], they aren't a definition. As such, they are helpful.

--

[0] - Or latent space, if I want to make an analogy inside the analogy apologia.

[fabian2k]

I don't think the programming analogies are helpful here, they seem to cause a serious amount of confusion. The basic idea isn't so bad, but too many people seem to try and push them much, much further than they can work.

For a developer this analogy also implies a lot of assumptions that they know to be true for code that are simply wrong for biology.

[theGnuMe]

One can view the cell as an information processing entity. If we agree with that view then it can be analyzed as an abstract computational process.

If you are upset about using the word 'computational' then consider it to be a dynamical process. We can then use mathematics to analyze this system.

In any event, genes (programming instructions) encode proteins (applications) that run in the cell (operating system).

Now biology is weird and has multiple feedback steps, some of which we probably do not even know yet, but the basic approach is solid.

[fabian2k]

Those analogies imply things that simply aren't true. Genes and proteins aren't digital, they are real entities with physical and chemical properties that affect everything they do.

Of course you can model various aspects of cells mathematically. But that doesn't require any analogies to software.

[theGnuMe]

There's no requirement that a computer or computational process be 'digital'. Analog computers exist, in fact, the first computers were analog.

At any rate digital (0,1) strings aren't that different than DNA strings (A, T, C, G) and just because we have 4 characters in the alphabet doesn't mean you can't analyze it as an abstract computational process.

You can also discretize the concentration of molecules such that above a threshold switch like behavior occurs (gene turns on or off).

Also people have done experiments where they program DNA to perform computations to solve various problems like the traveling salesman problem. This is a direct application of using biology to solve a "digital problem" https://www.nature.com/articles/news000113-10

So here we have an example of an artificial logical problem encoded into DNA and solved using biology. That means biology can simulate computational algorithms.

[FrustratedMonky]

"model various aspects of cells mathematically"

I think a lot of people here will equate software/programming to mathematically modeling.

Saying you can model/math it, but not use software analogies, is just really trying to split hairs, since models/math is also software.

[getoffmycase]

The fundamentals of biology are not difficult to grasp. Evolution, DNA -> RNA -> Protein, basic cell signalling, etc are all really easy to grasp with just the tiniest bit of effort. There's really no place for bad analogies, especially such a misleading one.

Sorry to bite your head off, but the reason that I'm passionate about this topic is, and I'm not joking, young earth Creationism. An analogy like the grandparent is something simple to grasp by many people, and then the Creationists can quickly turn around and say, "Well you see how biology is like a computer; somebody built a computer; therefore, God created us in six days, 6000 years ago."

[dang]

Swerving into religious flamewar is not a good move either. Please don't do this on HN.

[simpletone]

> The fundamentals of biology are not difficult to grasp.

The fundamentals of biology are extraordinarily difficult to grasp. Not only that, the fundamentals aren't set in stone and are subject to change. The rise of epigentics being a recent example. The only people who claim the fundamentals are not to difficult to grasp are people who have a superficial and incorrect understanding of it.

> Sorry to bite your head off, but the reason that I'm passionate about this topic is, and I'm not joking, young earth Creationism

So you got triggered because you have a political agenda? It's my experience that people who know nothing argue with or feel threatened by creationists. I say this as an atheist.

> "Well you see how biology is like a computer; somebody built a computer; therefore, God created us in six days, 6000 years ago."

The biology-computation analogy has been used since the founding of computer science. Everyone from Turing to the commenter you attacked has used it. Heck, even biologists view biological systems are biological machines.

You are fundamentally no different than the creationists you argue with. I'm almost certain you know nothing about biology or computer science other than pop culture nonsense.

[dang]

Please don't post in the flamewar style to HN, no matter how wrong someone is or you feel they are.

Also, please edit out swipes from your comments.

Both these points are in the site guidelines: https://news.ycombinator.com/newsguidelines.html. If you wouldn't mind reviewing those and sticking to them when posting here, we'd appreciate it.

[Viliam1234]

How exactly is the analogy "DNA: file.c" helpful to answer the question from the article "Where did DNA come from?"?

[jibal]

How does a computer act? I think your understanding of that is too narrow ... consult the Church-Turing thesis; biology computes. And an operating system is just one sort of program ... it makes no sense to talk about reducing to an operating system; that's a category mistake.

[Ralfp]

This is sometimes called „Andrew Grove Fallacy” after Intel Ceo’s interview in Newsweek where he famously commented how drug research should look at CPU engineering for inspiration to improve itself, missing point that in biology we dont have privilege of knowing how each part of the system works, compared to, eg. designing CPUs, and making invalid analogies.

[tivert]

I've also heard this described as "Engineer's Disease": "We think because we're an expert in one area, we're automatically an expert in other areas." (https://news.ycombinator.com/item?id=10812804).

I think the core lazy assumption that enables is the idea that all the other fields are properly understood as just like your field. Sometimes it's so bad that a software engineer will outright dismiss the ideas of actual experts as misguided, and insist on some "disrupting" the fields with some half-ass software-thinking.

[dekhn]

Parts of biology act like analog computers even if the whole of biology exceeds what we see in computing.

But more importantly: the existence of DNA demonstrates that information processing is universal and that there are many common aspects between current approaches to silicon-based information processing and biology-based information processing.

[FrustratedMonky]

Is there an alternative to analogies and metaphors when using language?

What terms/concepts should be used in their place?