Seperis (seperis) wrote,
Seperis
seperis

  • Mood:

eureka - because I'm watching this part now and it bothers me.

So first, I'm going to tell you about one of my personal things: the Balsam Wood Test.

In SGA fandom, there's a fanfic--I cannot remember which--where in a throwaway scene, Rodney and co are testing an unknown Thingie for reactions to common substances. It reacts to nothing at all, great. Then someone throws in a piece of balsam wood, and as it turns out, Thingie reacts badly to it.

To balsam wood.

Anyway, that stuck in my head--I love that kind of thing--and eventually, the idea turned into a concept of how to reliably test reality when all you have is your subjective self to work it out. The Balsam Wood Test.

Now, Eureka's Matrix: I love it. I love it for so many reasons, but all of them are relationship and people based. I love the characters dealing with it. I love the drama around it.

I hate the fact that anyone, anywhere, would think the Matrix could, even by accident, forward the study of science as it pertains to anything but the study of artificial reality and maybe the limits of computer programming. That's not just insane, it's--I need a word here, just go with 'are you high and have been since the Enlightenment?'

It can't be done, full stop. Even if it was run by an AI, it couldn't; if the AI actually could do that, you wouldn't need a Matrix because you wouldn't need people to discover anything; the AI could do it all. A computer could not, ever, reliably reproduce science as we know it--much less Eureka-level science--well enough to fool actual scientists for more than five seconds and maybe not even then.

(I'm not entirely sure it's really possible to create a Matrix reality indistinguishable from reality-reality, but that's another story.)

You see, there's no such thing as random numbers in programming. But I'm getting ahead of myself.

We have no idea what objective reality is; it's all subjective to varying degrees. Physics tries very hard, but even things we designate laws are very much 'well, nothing has contravened this yet so here we are'; anything lighter is 'current best explanation we have but we're open to suggestions' at best. Entire swathes of science exist based on math with the understanding we also haven't discovered all the math. To be generous, we're not even aware of about 99.999999999999999999% of physics. Of the part we're aware of, we maybe are sure of none and best guess a very tiny amount of that. And i won't even begin to describe where we are in pure math; maybe a little better? But probably not.

Like, this is true of all science, but I'm focusing on physics because the Matrix kind of requires it; that's the baseline on pretty much everything.

The Balsam Wood Test: in the Matrix, if Rodney had thrown a piece of balsam wood into that machine, there would have been no reaction because even best guess anywhere in history would not have prepared anyone for the idea that combining Thingie and balsam wood would go bad. That's not even a maybe; no sane programmer would throw that into a probability table because it wouldn't occur to them. So if that had happened in Matrix Eureka and they used the machine in the real world and someone was wearing a balsam wood necklace and it fell in the Thingie, boom: so much would have gone wrong it's ridiculous.

It's absurd; it's ridiculous; it's insane. Balsam wood: who would have called that as the nemesis of Thingie? Who would have called injecting the pus out of a smallpox blister into someone as an early form of inoculation would actually (kind of) work? Inoculation wasn't even a thing that existed when someone tried that.

(No, seriously.)

Balsam Wood Test: reality is absurd.

Science is the discovery of all the ways its absurd and try to work out why (sometimes, it doesn't fail completely). You cannot create something new within a structure where nothing is new or can ever be. Unlike computers, reality has no constants, just variables. Some of those variables are persistent as fuck, but as I said: nothing has change them yet. Binary is yes or no; there's no such thing as maybe. The only questions in the Matrix already have answers; you cannot answer a new question and you cannot change the answer of an existing question. And that is the opposite of science.

Exception: the study of programming. Then fuck yeah, you can find out all kinds of new things...as they relate to code. Probably a fuckload on engineering virtual machines for gaming, modeling, maybe--no promises--some advances in pure math and definitely some revolutions in graph theory, but not the fundamentals of the universe and reality as we know it. And nothing in math that would radically change our understanding of math either; that's because there's no such thing as random numbers when it comes to computers.

I'll come back to that, promise.

And even all this assumes it is possible to program a reality for greater than one person to believe, which is a huge maybe in itself. Perceived reality is subjective, and jacking directly into people's brains would actually make it much, much harder. We're all of us constrained to a certain extent by the physical limitations of our bodies and how they interact with the brain and much like physics, science is well below 1% at best when it comes to pretty much most shit including biolog. I cannot even imagine how to programmically recreate the body of someone with an autoimmune disorder or insomnia or hell, chronic fatigue syndrome well enough for them not to feel something is off above and beyond, much less individualized experience with such. And that leaves off psychological conditions and I am seriously stumped how on earth no one seems to consider the problem of the brain's ability to randomly override pretty much any function for the fuck of it but sometimes for also legit survival related reasons.

In other words, if I get chased by a bear in the matrix, if the brain thinks my body has been sleeping in my bioprison, it probably is going to hit me with enough adrenaline to knock me out of the matrix and/or cause heart failure because BEAR DEATH WHY ARE YOU SLEEPING YOU IDIOT. And unlike the Matrix, the real world's rulesets are persistent variables and you cannot program my real life brain not to do the unexpected. The brain does crazy shit for fun and wtf; hook someone into the Matrix, there's no way to know how the brain would react to that. It may not even let a person accept that as reality even on the off-chance it was perfect. The brain regularly rejects reality as reality for fuck's sake.

In other words, biology is fucked in the Matrix; for fuck's sake, how do you simulate unknown mutations, much less frequency to match something even passing for real and useful in the real world? With random numbers? Heh. I'll get to that.

And every bit of this assumes programmers won't make mistakes and as a professional QC analyst: oh God, that's funny. It also assumes that mistakes are the reason programs sometimes don't do what you want and sometimes do something you didn't expect: that's even funnier. Computers be crazy; they're subject to reality, where there are no constants, only variables, and only a very few persistent. The more complicated the program is, the more chance even perfect programming will interact in unexpected ways; not because anyone did anything wrong, but because that's the nature of complex systems. You cannot predict the unpredictable.

Now, my biggest and seemingly minor problem except it's a major one: random numbers.

Well known but not appreciated fact: there's no such thing as random number generation in a program.

It look random, and we're developing very sophisticated ways to simulate the random number, but--it's not and can never be truly random because the basis is and will always be a formula. It may take a very, very long time to work it out, it may require a massive amount of data before you can see it, it may be incredibly difficult and very improbable you will work out the pattern, but there is a pattern, all starting with a function (or program) who's only job is to produce seemingly random numbers. Which means that every single thing inside the Matrix would not ever be random, ever and reality is--well, really really random.

Like I said, the formula can be very sophisticated: it could be 'use my gps coordinates right now, add six, and divide by the age of the president of the US who was born closest to this date at his time of death'. It could be that 'plus the number of cats in this pound in Chicago on this day five months ago, then translate the number to binary, and divide by the date of the nearest holiday to this date'. Add in 'Let's base twelve this entire thing now' to round it off.

That's still a pattern.

Maybe not one a person could work out on their own, but. A computer could find the pattern. They're actually pretty good at that, provided you know what you're doing and sometimes when you don't. And if you have a computer sophisticated enough to build reality and you are the type who really believes--insanely--that you're doing this to Forward All Sciences, then short of hobbling your Matrix-reality computers to not work--and truthfully, that's so meta my brain hurts--all you'd need to break the Matrix is someone to track random storms, random tornadoes, random hurricanes, random anything and given enough data, a pattern will emerge eventually. A normal scientist, maybe not: but building a Matrix for Science means you want the best minds in the world, so yeah, they'll find it. Which means a.) broken immersion or b.) illegitimate science because in the real world, random number generation patterns do not predict when tornadoes happen. We don't know anything but conditions that could make them happen; to predict in the Matrix, all you need is to know the formula and once you know there's a pattern, finding the formula is just a matter of time.

(Not to mention the sheer amount of processing power needed just to create seeming randomness. The more power, the closer you can get to random, but--seriously, you'd need entire machines dedicated to nothing but creating those 'random' numbers. Now my head hurts.)

This little problem with random numbers will also cause problems in pretty much any higher math and all of physics--random chance and chaos are actually really really really really important to the very fundamentals of science--as well as really fuck up any legit programming people in the Matrix try to do, and that's just the shit I understand well enough to write here (no promises on if I understand more than the problem exists); there are entire branches of math and computer science that simply won't work in a programmed reality at all.

On any other show, I'd go with it, but Eureka--which is literally About So Much Insane Unknown Science--I just cannot deal with a Consortium who seem to at least know what science is (though maybe not) thinking 'this is a really brilliant idea' like--ever.

I needed that rant so badly. I feel better now.

Okay one more thing: for fuck's sake, your insane matrix made a dragon in like the first week. A. Dragon. The programming created a dragon. Dragon.

DRAGON.

You think your Matrix can be a haven for real, legitimate scientific discovery when it randomly makes fucking dragons? The Matrix can't even manage to reproduce known reality but you think unknown reality won't be a bit of a problem? WHAT THE HELL IS WRONG WITH YOU?

Okay, really done before I lose my shit over rendering errors and how it didnt' seem to occur to anyone that when the brain is jacked directly into the matrix, your physical body isn't actually involved, especially say, the optical nerve or like, the physical eye. Rendering errors occur exclusively in a visual medium with a physical body looking at something.

The Matrix is not a visual medium; what they see is what is programmed into it. This isn't happening on a computer screen or hologram for them; they are not physically inside some kind of super sophisticated chamber of Matrix doign shit with their own bodies; this is happening inside their minds. The only way a rendering error should even exist is if their physical bodies are involved in a simulated environment. THIS IS IN THEIR BRAIN. THERE WILL NOT BE RENDERING ERRORS.

There will be weird shit like birds in rocks, yes. There will be even weirder shit that can happen. But the only way they will see a rendering error is if you specifically program it in to happen under certain conditions, and why would you do that?

Dragons and rendering errors and random numbers and balsam wood causing destruction of everything. Okay, really done.

Really done. Promise. Mostly. Posted at Dreamwidth: https://seperis.dreamwidth.org/1081526.html. | You can reply here or there. | comment count unavailable comments
Tags: eureka, jenn's life
Subscribe
  • Post a new comment

    Error

    Anonymous comments are disabled in this journal

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

  • 0 comments