I guess the software would end up supporting that way of using it. A friend set would be a presence that trusts your friends at 0.9 then is queried with a min_similarity of 0.9. (meaning, anyone they trust at strength less than 1 [if they’re good bayesians, they wont trust anyone at 1!] will not appear in the results).
Another thing you could do, that might lead the recommendation network to grow a lot faster than traditional friend networks: You could have a general good taste web, post all of your content to your presence in that web, and use it to curate all of the tags you were interested in. I’m inclined to say that would be lazy, suboptimal, and not encouraged, but you could do it, it would work admirably well as a recommender system.
It would have an advantage over the traditional you only see the specific channels you follow thing in that friends wouldn’t always have to hear about or spend time appraising their friends’ new channels, in order for them to get seen and supported as soon as they’re created. If you’re interested in politics, then if a friend makes a politics alt, you’d see it automatically. The disadvantage of that, of course, is that taste is only general to a certain extent, a person with taste in epistemology might have terrible taste in politics. Maybe you like politics, but not how your friend does it. You could block their politics channel. If that isn’t enough; have a politics presence that directly trusts the politics channels you definitely like, then have it only weakly trust your friend web, so that the automatic friends’ politics stuff only comes up once you run out of the other stuff.
We should probably give that culture of usage a name. Maybe “overloaded web”. A lot of the time it would be a pathology. You would be able to distinguish their content, using tags, but you wouldn’t be able to distinguish the different kinds of endorsements they make, there would only be the one. That sucks. It might help to, for instance, make it really easy to submit any post tagged politics to both your general good taste presence and your politics-specific presence automatically, to make it easier to maintain them both.
I burned out fairly recently too. I worked on a game for two years before realizing I couldn’t motivate myself to finish it. I then made a tiny game in a week, that helped a lot. It was more art than entertainment. It reminded me that I could make games that I really considered to be interesting, that showed me things I didn’t already know (it was a little evolution sim.)
I’m now making two small but potent games that I think the world might really need and I am a lot more motivated… I don’t think it’s about their objective importance. It’s probably really about being able to convey passion for the design to friends, who then visibly reciprocate it. That’s probably what’s doing it.
I imagine that might have been difficult for Arbital. It might be difficult for the above proposal too, it’s all very abstract. It might be a good idea for me to sit down and write a short document that narrows in on all of the ways I expect Tastenet to reduce catastrophic existential risk by enabling the spread of useful information and decreasing political conflict. “It Might Kill Twitter” probably isn’t enough to keep me going. We don’t need a better faster twitter, we need something qualitatively different.
Have you played The Witness? I’ve put your name on the beta list for crycog. I am looking for puzzle designers, but I should probably try bugging the friends I’ve already conscribed a few times before giving up on them. I have so far not tried bugging or in any way cajoling or hyping them so I should just see if it works at some point.
I’m wondering if there’s much point in showing other people that game right now though. I’m not sure anyone believes there’s going to be depth here but me and so there is no guarantee they’ll be able to find it. This seems to be the case for most successful games. I hear the concept of something new and interesting and like, there will usually not be an intuition that says “ah yes I can see why that would work”.
My intuition hates The Mind, for instance. I guess if someone asked me to contribute to a game like that though… maybe I’d say “okay this is for people with fewer tacit communication insights than me, I can do this condescendingly.” That said, to be clear, I believe that I am right. I do earnestly believe I wouldn’t be able to enjoy The Mind. You’d have to fight me to get me to try it.
The other game I’m developing is a tabletop game where basically… players have utility functions that act on an environment that they coexist in, violence is possible, they have to sort things out in such a way that it maximizes their own utility function. There is an emphasis on negotiation as a theory-heavy skill, as most games allow only one winner, every time the opponent gains you lose, there aren’t very many games like this, and that’s kind of shocking considering how normal in life this kind of situation is.
What sorts of stuff have you worked on? What sorts of things do you still want to create? (what are your absurd and unlikely design ambitions)
I guess the software would end up supporting that way of using it. A friend set would be a presence that trusts your friends at 0.9 then is queried with a min_similarity of 0.9. (meaning, anyone they trust at strength less than 1 [if they’re good bayesians, they wont trust anyone at 1!] will not appear in the results).
Another thing you could do, that might lead the recommendation network to grow a lot faster than traditional friend networks: You could have a general good taste web, post all of your content to your presence in that web, and use it to curate all of the tags you were interested in. I’m inclined to say that would be lazy, suboptimal, and not encouraged, but you could do it, it would work admirably well as a recommender system.
It would have an advantage over the traditional you only see the specific channels you follow thing in that friends wouldn’t always have to hear about or spend time appraising their friends’ new channels, in order for them to get seen and supported as soon as they’re created. If you’re interested in politics, then if a friend makes a politics alt, you’d see it automatically. The disadvantage of that, of course, is that taste is only general to a certain extent, a person with taste in epistemology might have terrible taste in politics. Maybe you like politics, but not how your friend does it. You could block their politics channel. If that isn’t enough; have a politics presence that directly trusts the politics channels you definitely like, then have it only weakly trust your friend web, so that the automatic friends’ politics stuff only comes up once you run out of the other stuff.
We should probably give that culture of usage a name. Maybe “overloaded web”. A lot of the time it would be a pathology. You would be able to distinguish their content, using tags, but you wouldn’t be able to distinguish the different kinds of endorsements they make, there would only be the one. That sucks. It might help to, for instance, make it really easy to submit any post tagged politics to both your general good taste presence and your politics-specific presence automatically, to make it easier to maintain them both.
I burned out fairly recently too. I worked on a game for two years before realizing I couldn’t motivate myself to finish it. I then made a tiny game in a week, that helped a lot. It was more art than entertainment. It reminded me that I could make games that I really considered to be interesting, that showed me things I didn’t already know (it was a little evolution sim.)
I’m now making two small but potent games that I think the world might really need and I am a lot more motivated… I don’t think it’s about their objective importance. It’s probably really about being able to convey passion for the design to friends, who then visibly reciprocate it. That’s probably what’s doing it.
I imagine that might have been difficult for Arbital. It might be difficult for the above proposal too, it’s all very abstract. It might be a good idea for me to sit down and write a short document that narrows in on all of the ways I expect Tastenet to reduce catastrophic existential risk by enabling the spread of useful information and decreasing political conflict. “It Might Kill Twitter” probably isn’t enough to keep me going. We don’t need a better faster twitter, we need something qualitatively different.
That’s funny I used to make games professionally and as a hobby too. If you want to send me your games I’d be happy to take a look.
Have you played The Witness? I’ve put your name on the beta list for crycog. I am looking for puzzle designers, but I should probably try bugging the friends I’ve already conscribed a few times before giving up on them. I have so far not tried bugging or in any way cajoling or hyping them so I should just see if it works at some point.
I’m wondering if there’s much point in showing other people that game right now though. I’m not sure anyone believes there’s going to be depth here but me and so there is no guarantee they’ll be able to find it. This seems to be the case for most successful games. I hear the concept of something new and interesting and like, there will usually not be an intuition that says “ah yes I can see why that would work”.
My intuition hates The Mind, for instance. I guess if someone asked me to contribute to a game like that though… maybe I’d say “okay this is for people with fewer tacit communication insights than me, I can do this condescendingly.” That said, to be clear, I believe that I am right. I do earnestly believe I wouldn’t be able to enjoy The Mind. You’d have to fight me to get me to try it.
The other game I’m developing is a tabletop game where basically… players have utility functions that act on an environment that they coexist in, violence is possible, they have to sort things out in such a way that it maximizes their own utility function. There is an emphasis on negotiation as a theory-heavy skill, as most games allow only one winner, every time the opponent gains you lose, there aren’t very many games like this, and that’s kind of shocking considering how normal in life this kind of situation is.
What sorts of stuff have you worked on? What sorts of things do you still want to create? (what are your absurd and unlikely design ambitions)