Hey, you, who's reading this in the future: you can find out who I am today at https://rarf.zone/about/. 💙


E-mail me at:
info@rarf.zone
Mastodon/Fediverse
yiff.life/@katja

Of all the silly Not Magic™ that Katja does, this is probably the most… wacky kind? Everything else just breaks physics in a manner that's totally compliant with normal causation and the flow of time, but this isn't. So I figure it might be the most fun to write about.


For what definition of "cannot"?

Ontologically incapable.

For what definition of "death"?

In this case, it's the information-theoretic one. The statement that "Katja cannot die or kill" is false when "death" is taken to mean medical death. So, basically, they can't die or kill in any way that matters. They can't die in the sense of getting kicked into the afterlife or ceasing to be; they can't kill in that sense, either.

What actually does the time looping and stuff?

Something to do with their hard light. Their physical body is just a holographic (but entirely physical and substantial) reflection of their hard light, which is both literally their soul and also the means through which, somehow, they can engage in limited violations of causation and the normal flow of time.

How do they do this time fuckery and seeing the future, though?

Well, they use their Weird™ eyes for that! ("it" here refers to each pair of eyes.)

  • Cassandra: cyan; sees outcomes. Basically, it displays what will happen as a result of whatever Katja is currently (not) doing… …to the extent their knowledge allows predicting it. So, it works very well when it's just a matter of inanimate, macroscopic objects. Chemical reactions? Works pretty OK. Biological processes? Also decently accurate. Predicting actions of friends or people whose mental states are obvious? Fairly useful. Computers that Katja has kernel or deeper access to? Easy. Strangers or unfamiliar computers? It's useless. (So, effectively, Cassandra makes it so Katja can't not see the reasonably foreseeable consequences of their actions.)
  • Edifice: magenta; sees risks. Or, rather, the whole set of possible outcomes in some given space, across… …usually, a long enough period of time that Katja can't actually use all the information they're getting through it. Could be as little as a few seconds, in a crowded space. It's sort of like seeing a bunch of overlapping vignettes of possible futures, with the more likely ones being brighter due to more overlap. The apparent prominence of each is somewhat a function of likelihood × scale of consequences, drawing focus to particularly salient ones. (This helps Katja see results of the actions of themself and others that aren't necessarily within the realm of even contemplating via normal means.)
  • Madness: yellow; sees possibilities — for 🌌reality hacks🌌, specifically. Usually doesn't correspond much to ordinary reality, as it sees, essentially, exploits that can be leveraged to edit reality. Often, its visual field "looks" like colourful blocks, planes, and lines in space. Objects in its vision are coded by colour, opacity, pattern, apparent prominence, and other variables to indicate what can be done with any given thing it perceives. (This just enables Katja to do their reality hacks.)

So, basically, by combining input from Cassandra, Edifice, and Madness, they can see the future, in a fairly limited fashion.

Not killing, generally

Basically, this applies to any sentient being. So Katja can't even fish, let alone hunt.

How it usually works is that, at the moment Katja starts intending to do anything that, regardless of how indirectly, will inevitably result in a death, a "restore point" is set. If that intent goes away, it goes away. However, if that intent ends up translating into intent strong enough for them to act on it, that that causal chain is initiated, then: ✨things✨ happen. Via some mechanism or another, a loop is set in the flow of time, and the moment that causal chain either results in information-theoretic death or exits the range of Katja's foresight, leaving that question vague — say, when someone can no longer be resuscitated — time jumps back to that restore point… … …but them, whoever/whatever died, and any observers will keep the exact same mental state they had at the moment the loop jumped back.

So, for example, if they end up accidentally doing something that, in a ridiculous but fundamentally deterministic Rube Goldberg way, results in the inevitable death of some person, they'll remember what they did that caused that chain of events, the other person will remember what preceded their (now reversed) death, and hopefully, that should prevent a repeat. If it doesn't, that's a new restore point, new loop, so on, until things don't end badly.

Hypothetically, if they tried to kill? It'd be the same thing. No matter how excessive the force they used was, time would just snap back to before they decided to do that action. (Practically, though, they wouldn't.)

Not dying, generally

Similar behaviour here, usually. Due to not being able to retroactively work back from the moment of their death to set a restore point at the moment someone else developed the intent to do something that would result in their death, they instead passively (as far as they're concerned) maintain restore points at various specific points in time, updating these dynamically. Where Cassandra and Edifice don't manage to give them the information they need to avoid dying, and that death ends up either happening or becoming inevitable, there's two possible outcomes:

  • If they had any warning as to their impending death, they will activate that loop in time, jumping back to the nearest safe restore point. Everyone involved, once again, retains all their memories.
  • If they had no warning, their body will just fade into nothingness, them using the most recent safe restore point to reconstruct their body, once again out of thin air.

Are there any special cases that explain all this "usually" and "generally" stuff?

You're in luck, because: just wait for my next post.


You must log in to comment.