3 ms·

Bayesian statistics at work. If a mechanic or a car salesman speaks to the man he'll be correct 99% of the time. Same for a pram salesman talking to the woman.

Bayesian statistics at work. If a mechanic or a car salesman speaks to the man he'll be correct 99% of the time. Same for a pram salesman talking to the woman. I blame these monkeys: https://www.livescience.com/22677-girls-dolls-boys-toy-trucks.html https://www.livescience.com/22677-girls-dolls-boys-toy-truck...

rich_sasha 6y ago
I hate stereotyping, but also sympathise when people use Bayesian reasoning. These two are hard to reconcile, but in an imperfect information world, we have to make assumptions.
- TuringTest 6y ago
  Bayesian reasoning also means that treating a female client like shit is likely to encourage her to make a fuss, but somehow "bayesian reasoners" never make that inference.
  - rich_sasha 6y ago
    Yeah, it’s not a one-size-fits-all. If an old granny asks about computers, will you contemplate that she may well be a Linux engineer and know lots, or will your first pitch assume she doesn’t?
    - TuringTest 6y ago
      If she's buying at a Linux shop, I'd give a try to the hypothesis that she's an ancient hacker with l337 sk1lls, and make sure before assuming she's just entered the wrong store and really was looking for the grocery next door. For the same reason as the car example above or the Pascal's Wager - the cost of a prediction failure is so high that it changes the probabilities of your wrong-case calculations, no matter how strong the stereotype.
      - rich_sasha 6y ago
        It's not an either-or. You can treat people with respect, and assume they are knowledgable, and yet make Bayesian reasoning-type arguments. If I'm driving through what looks like a dodgy neighbourhood and need to stop and ask for directions, I'll be extra vigilant - based on essentially impressions. But I won't assume anyone I speak to is a thug, much less treat them that way. In other words, Bayesian reasoning is not an excuse for being an arsehole. Really, no cognitive process can be...
JoeAltmaier 6y ago
Probably. But its pointless - the cost of a mistake here is zero. The value is in not pissing off the customer. So its a lazy, biased approach and it pays to be self-aware enough to cure oneself of it.

rich_sasha 6y ago

I hate stereotyping, but also sympathise when people use Bayesian reasoning. These two are hard to reconcile, but in an imperfect information world, we have to make assumptions.

TuringTest 6y ago
Bayesian reasoning also means that treating a female client like shit is likely to encourage her to make a fuss, but somehow "bayesian reasoners" never make that inference.
- rich_sasha 6y ago
  Yeah, it’s not a one-size-fits-all. If an old granny asks about computers, will you contemplate that she may well be a Linux engineer and know lots, or will your first pitch assume she doesn’t?
  - TuringTest 6y ago
    If she's buying at a Linux shop, I'd give a try to the hypothesis that she's an ancient hacker with l337 sk1lls, and make sure before assuming she's just entered the wrong store and really was looking for the grocery next door. For the same reason as the car example above or the Pascal's Wager - the cost of a prediction failure is so high that it changes the probabilities of your wrong-case calculations, no matter how strong the stereotype.
    - rich_sasha 6y ago
      It's not an either-or. You can treat people with respect, and assume they are knowledgable, and yet make Bayesian reasoning-type arguments. If I'm driving through what looks like a dodgy neighbourhood and need to stop and ask for directions, I'll be extra vigilant - based on essentially impressions. But I won't assume anyone I speak to is a thug, much less treat them that way. In other words, Bayesian reasoning is not an excuse for being an arsehole. Really, no cognitive process can be...

JoeAltmaier 6y ago

Probably. But its pointless - the cost of a mistake here is zero. The value is in not pissing off the customer. So its a lazy, biased approach and it pays to be self-aware enough to cure oneself of it.