DON'T BE LATE -- you will be judged (p2p ratings systems in the future)

I don't like when people are late. lol.

But: It's not that I don't understand that shit happens and that occasionally there's some quirk that causes you to be late. And who cares. No worries.

It's that your intention should never be to be late. If you say 5:15, you should feel a sense of fierce commitment to be there at 5:15. Because otherwise you're being disrespectful and abusive towards another person's time.

You're making them wait when they could have been doing something else.

I'm pretty sure some people, probably not on a conscious level, will deliberately prefer to be late so that they aren't the person waiting, lol.

And then cultures even evolve around this, where saying "8 o'clock" means everyone assumes that it's really some approximate time later than that. So there's just so much more confusion and overall time wasted when people behave like this.

In my world, saying "8 o'clock" means my intention is to be there when the clock says 8:00, lol. If I meant something different, I'd say the different time that I intend to be there.

ratings

I think a lot about peer-to-peer ratings models and how it will all work in the future. And how it will encourage polite and cooperative behavior.

For a ratings algorithm to work correctly, there needs to be some ability to "rate the rater". To rate the accuracy and how good of a judge that person is. So that, simple example, if someone accuses people of theft who aren't actually thieving, the system needs to learn to dampen or eliminate this voter's impact.

So then the problem always comes back to how to get the ball rolling. (That's the part I can never work out in my mind.. and always just assume there's some brilliant algorithmic thinker who will solve it somehow.)

If there was a way to know Joe, Tom, and Mary over there are reasonable people with good intentions, then we can start there -- they rate other people and rate how good those people are at rating, and the eggshell starts to peel tethered from that.

But of course in practice you can't actually do that. At least, it's a huge leap of faith. Who gets to decide that these people are actually reasonable and the best source of where it should begin? It doesn't make sense that the ideal backbone of what we'll use to govern our relationships with each other would begin like that, with these few special people.

So I never can get to how it technically works, even tho I'm sure we need something like this and this is what the future looks like.

And then it hit me...

It isn't one model that we all have tiny little ripples in. It's we all have our own model.

I rate people as I see fit, and you can look at my model if you want to and if I want to share it with you. People who believe in my outlook on things can look at my model and factor it in as they see fit.

And my model itself is the tiny ripple in the big ocean of various models, but among people who I know and am more intimately connected with, it's a bigger deal. Within your circle of influence (among the people who matter to you) your model is meaningful.

layers

There are layers to it. I'll have my own backbone ratings. But then if I trust that you're a good rater, I could choose to base my model on, say, 75% me and 25% you. (More realistically, it could be 5% me, and 95% a collection of other people who I trust, all to varying degrees.)

And then "my" model is more robust and goes beyond just the people I know and the roughness of small sample sizes and my own mistakes etc. I use my judgment of who I think is good at rating other people to form my model.

And so there would be two different things: there would be my own backbone ratings (which will be rough, but right to the heart of what I think of people) and then there will be my overall model which incorporates other peoples' backbones or overall models.

You probably don't share the backbone with that many people, but you share the overall model more loosely, maybe completely publicly, since it's less intimate to your own thoughts and feelings.

And so we don't have one official rating in the overarching system. We have a different rating in each person's prism.

If you're seen as a good judge, you'll tend to have more influence. People will be more likely to look to your prism, and the people who include you in their prism will generally have a better prism because of it.

And there could emerge a few prisms that are de facto recognized as really good ones, similar to how we have Moody's and Better Business Bureau and Equifax and all that stuff today (it'll just be far more meaningful and dynamic).

But technically you don't have one official score; you have different unique scores in each unique prism. And it's constantly evolving how much anyone cares about any particular prism.

seems intuitive

I mean, it's special and introverted. But that's just how it works, imo. I feel like we all have our own world and it flows from there, into a mutual and collective, rather than vice versa.

Besides actually working (like, not having the "where does the eggshell begin" problem), it also seems more viscerally satisfying. If someone wrongs you, it probably wouldn't feel that satisfying to send a ding into the ocean of all their other feedback.

And then, what, you still need to worry about a retaliatory ding and be ready to defend why they're lying and yours was correct, and drag it out into this whole process just to hopefully be able to have a tiny little impact?

Rather, you can lower them in a big way in your own view of things, and the people you regard as worthwhile will see it and be able to know it, and there's nothing to fuss about.

one more technicality

I think you'd also want to be able to build in a "delay".

Suppose you have some negative feedback about someone. You maybe don't want them to be able to know that it was you who gave that feedback (if you had lunch with them earlier in the day, they'd be able to put 2 and 2 together).

If you think someone is vein and self-centered, that's intimate to you and it doesn't need to be their knowledge that you think this for it to still be legitimate and accurate and helpful to other people.

Plus you don't want to have to worry they just ding you back because they're mad.

And even tho your backbone ratings are private, if other people are tapping into you, then the person you just met with would be able to see public prisms change to be able to deduce what feedback you seem to have entered.

So you could choose a time delay.

Customized or random. And maybe gradually bleeding the new info rather than all at once would be helpful too (like if you were entering some characteristic about them, or changing their score, they could get a tenth of a point per week rather than the full point all at once, or something like that).

And of course you'd choose to skip the delay for things like "that guy assaulted someone!" (In that case you're perfectly happy to be known as the person who caught it. It's not nuanced or intimate. And it's important that people have this info right away.)

So it's situational, but there are plenty of judgments that are best kept private and don't need to be publicly known that you feel that way.

---

Take notes, Black Mirror S3E1 😃 You were cute but you only scratched the surface.