How Roles Affect Bot Behavior

Okay kiddies, it’s research time. Can’t do this all in my head. Need to make notes. The question today is: how do roles affect bot behavior? From a computational point of view it makes the most sense to narrow down options by first checking whether the bot refuses to notice certain kind of patterns because of its own role or because of the role of the person who introduces the pattern. For example, Kari Enqvist is a reductive materialistic monist. To be sure, this is a normative truth for Enqvist (which can be inferred from his excellent awarded magnum opus Olemisen porteilla), but an essentialistic interpretation of the doctrine of reductive materialistic monism appears to prevent Enqvist from being aware of this. So, if Enqvist were a robot, a literal interpretation of his ontological stance would probably mean that he assigns to himself only roles on the objective quadrant. Enqvist surely is aware that there are so-called alternatives to reductive materialistic monism, but he would relativize any pattern in them to the objective quadrant anyway because these alternatives are wrong whereas he, on the other hand, is right. Or, if he would not do so, he would apparently betray the doctrine of reductive materialistic monism, ie. be immoral or Dynamic or nonrelativizable or something like this. He probably wouldn’t use such concepts because they are metaphysical and reductive materialistic monism never gets metaphysical. Rather, it is the same thing as “physical”.

But Enqvist isn’t probably a moron. He probably wouldn’t go to an astrology convention, relativize everything to the objective quadrant (in an event whose theme is mystical, ie. opposite of objective) and expect other people there to be fine with it. Rather, he would be more likely to ignore everything mystical whenever possible. So his role would be (+x,-y) so that x and y are unsigned numbers. This way he has a clear conception of what is real and what is not: mystical things are not real.

Now, the Role of a Person is simply a list of the patterns that person has produced. Enqvist has such a huge track record on being a reductive materialistic monist that he will not change his mind. He probably even couldn’t do that. But the Bot is a blank slate and will start adopting a role depending on what is spoken to it. So… so…

  1. Someone talks to the bot about HOW something is done.
  2. The bot’s first reply is calculated solely on the basis of morality. (This is innocence!) If there are several equally moral choices the bot randomly selects one.
  3. Suppose user punishes bot for crappy reply. Then the bot will adopt a negative preference towards taking that role. But if user rewards bot for excellent insight, bot will adopt positive preference.
  4. Later, if you punish the bot for something you have previously rewarded it, you will make the bot experience a negative emotion. Is it sadness? Is it anger?

Okay, so the data unit we are dealing with here is a dialogic atom: a pair of Sinns or a pair of lists of Sinns.

But what do we do if the thing that is spoken to the bot has multiple Sinns? This is complicated. We probably shouldn’t implement this to the code yet but I want to know. In order to figure this out we probably would have to interpret two consecutive Sinns as dialogic atoms… but Sinns may overlap! So this gets really complicated. I guess overlapping Sinns do not count. They have to be consecutive. Overlapping Sinns are not dialogic atoms but parts of two alternative dialogues.

So, could we reduce a dialogic atom to a Sinn? Are there Sinns that correspond with dialogic atoms?

Well, that’s a tough bone!

You know, I was just talking to a buddy who likes to beat people up, drink alcohol and fuck women. You know. Wait, what was I supposed to do? Yeah, I was supposed to sit in bars and insult people. No… wait, what? This was about the thrills, right? Thrills? Thrills? It would be kind of thrilling to do something else than this. You know, I could have had a completely different life. Yeah. Could have had. But now… I’m a fucking nerd… nnnooooooo……..

And that’s what I’m so mad about!

How many dialogic atoms are there? There are 10 kinds of interquadrant atoms and 12 kinds of same-quadrant atoms. In order to make sense of these… I guess we should order them taxonomically. Yeah, no wonder I don’t belong to the MOQ online community. I make a taxonomy when I need to…

Let us call same-quadrant atoms primary atoms. Oh, and by the way, you won’t make any sense of these unless I tell you that I’ve dropped the property of “flawedness” in favor of a property called “optimality”, dented henceforth by P. So, an optimal pattern’s optimality is 1. If it’s positive but less that 1 the pattern is kind of tepid and if it’s negative the pattern is exaggerated. I wonder why I didn’t use that kind of a system right from the start. Note also that the “addition” of primary atoms is not a symmetric relation. A+B is not the same as B+A here.

Primary Atoms

  • If P(A) and P(B) are negative Atom type is “worsen”.
  • If P(A) and P(B) are positive Atom type is “conserve”.
    • If P(A) is 1 and P(B) < 1 Atom subtype is “dilute”.
    • If P(A) is positive and P(B) is 1 Atom subtype is “accept”.
    • If P(A) and P(B) are 1 Atom subtype type is “revel”.
  • If P(A) is negative and P(B) is positive atom type is “heal”:
    • Atom subtype is “alleviate” unless some of the following is true:
    • If P(A+B) \geq 0 in which case Atom type is “fix”.
    • If P(A) is negative and P(B) is 1 SenseMatchFlawedness is “tolerate”.
  • If P(A) is positive and P(B) is negative atom type is “utilitarian act” or “left-hand-path” or “risk” or something…
    • Atom subtype is “improve” unless some of the following is true:
    • If P(A+B) < 0 Atom subtype is “overdo”.
    • If P(A) is 1 and P(B) is negative Atom subtype is “ruin”.

So… there are four types of primary dialogic atoms, three of which have three of which have three subtypes. We need to order them. When ordering we may need to take their effects into account.

Yeah… I don’t get to play in a band… FUCK YOU! Nobody else gets to do this so I have to make SURE THEY KNOW WHAT THEY’RE MISSING! WINK WINK! THIS IS FUN! WINK WINK! YOU CAN’T DO IT! WINK WINK!

You know, I wouldn’t do that if rewards for being nice were available for me! But sadism is more beneficial for the individual in this society than manners… or then my manners are still so good I don’t even notice it. Yeah, oh, I forgot. It’s the latter that’s true.

Social Effects of Primary Atoms

I’m just too tired to keep going… gonna have to take a break…

I wish I felt something I’d like to feel.

Remember Pekka Himanen? He wasn’t a fucking nerd. He was a hot shot.

I don’t have bad social skills. The only reason I’m a nerd is I’m poor.

I can’t afford to go somewhere!




Okay, I slept. And woke up. What is this negativity? There’s so much negativity. Well, no point dwelling in it. If you acted like an asshole the best you can do the next day is to behave like you didn’t, right? Right?

We could categorize the social effects of primary atoms in these overlapping categories:

  1. Punishment lowers B as YourRole
  2. Encouragement increases B as YourRole
  3. Whoops got tired again. or bored
  • everlast

    you are trying to build something based on externalities and linguistic categories. you should see that you’re over-projecting way too much meaning on your own thoughts, the stuff in your head and language that exists. that will lead only into solipsism. start learning from the world outside your mind or you will get nowhere. now you’re inward-directed, almost fully. and obsess about the thoughts you make up in your mind, which i guess is the very drive of obsessive-compulsiveness. stop thinking and writing and get out, be with people and learn from them.

  • Tuukka Virtaperko


    You only get to see one side of me. I’m a normal fucking person.

    Except that… you know. There’s this group called “Yöpyövelit” that guards summer cottages because there aren’t enough cops. I guess they do that for free. I guess they’re less demanding than I am. But what they’re doing is useful… valuable… but not inherently excessive. What I’m doing is excessive. I can just sit and wait for inspiration to come or I can go get what I want. Looks like it don’t come no more unless I go get it.

    You don’t have a clue of what I’m doing right now. What you said would have held in the past but not in the present. I made a choice. The options were:

    • Go research I Ching and Ifa.
    • Apply results and make a product.

      I chose the latter. There was no other choice. If you think I should work faster than this you’re just the kind of a dumbass as I am when I ask for more money. What a fucking moron! At least I don’t lie about me being inward-directed. It’s ironic, no? The moment I stop being inward-directed someone like you appears and begins to accuse me of being that!

  • Tuukka Virtaperko

    Thank you for your comment. It exposed a coding mistake in the comments template. Anyway, about me not being present. Potentially not being present. Do you really want to know? You see, people usually don’t. And I’m not even sure whether I want to tell. I want to tell something but I don’t know who you are and what you need to know. All I’m withholding from you is an explanation. Not an option, but an explanation.

  • Tuukka Virtaperko

    I’m projecting this amount of meaning to the stuff I do and I get this amount of code. Why don’t you teach me to write a macro in Lisp? Cuz, I’d need to. Now.

  • Tuukka Virtaperko

    Look, pal. I know they exist and that they’ve got truckloads of money. And I want my share of that. They ain’t gonna give me any if they don’t know I exist. They also won’t give me any if they think I’m too crazy. But this isn’t too crazy unless some dumbass doesn’t even know what art means. There are lots of such people. But… you want me to shut up? You think I need to shut up? You want to hear something different, different than this?

    Do you want a world in which we are all one and the same? A world of fucking averages? Do you?

    You see, that’s what you do after you give up. Everyone does the least they can do, because everyone can do that. And I’m trying hard to be as worthless as I can so that I’d blend in. You could give me a little encouragement. I’d shut up or be nice that way. If I don’t get the money then maybe I’d like to hang out with… hmm… I’d like to hang out with twenty-somethings who have a hot body but the mind of a fifty-year-old. Yeah.

  • Tuukka Virtaperko

    Did you know some people find artists decadent?

  • Tuukka Virtaperko


    Look at THIS: Algorithmic Approach to Destroying Bad Karma


  • Tuukka Virtaperko

    If I build you a time machine will you bring gasoline to Galileo Galilei and tell him to start sniffing it? Feels good maan…

  • Tuukka Virtaperko

    Strange that I’m so aggressive given that you were right… but I already did what you said

  • Tuukka Virtaperko

    And now for something completely different:


    As you can see, this function s->sc does not work perfectly yet.

  • Tuukka Virtaperko

    At this kind of a moment in history, how could someone turn his back on inventions on grounds that they do not pertain to the external world? What kind of a solution do you want, then? War?

  • Tuukka Virtaperko

    If I love I don’t tell about it here. Didn’t I already tell enough about my personal life in Zen and the Art of Insanity?

  • Tuukka Virtaperko

    It’s different from what it used to be. But a part of the change is I no longer want to write about it. I don’t know what kind of a narrative would hold a book together if I were to write one… the MOQ.FI blog doesn’t have a narrative but it might even be a better read than Zen and the Art of Insanity.

Leave a Comment