= 27d567bc2d48d9f40e9ccc20ea370b15)
Perplexed comments on Narrow your answer space - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (111)
Please explain what you mean. How would you measure how balanced a particular value was in how it affects men and women? Would you weigh it by the number of men and women present on LessWrong?
A key problem is that there are two different, conflicting definitions of "discrimination", and we're asked to satisfy both. One says that "discrimination" means "following a policy that favors the natural inclinations of men more than the natural inclinations of women". The other says "discrimination" means "supposing that men and women have different natural inclinations".
If I can manipulate a value to be more balanced in how it affects men and women, that presumes that men and women do in fact have different natural inclinations. That would mean that instead of choosing a "balanced" value, with which to treat everyone, I could optimize better by treating women one way, and men a different way.
You could optimize better yet if you treated everyone as an individual.
Only if you have sufficient data on the preferences of everyone individually.
For the people you interact with regularly, you will quickly accumulate the data. For the people you don't often interact with, console yourself with the fact that they don't matter all that much.
Or you could operate using priors for individuals that match whatever little you happen to know about them. For instance their gender. For some purposes that is a lot of information.
If you find it best to treat males one way (until they individually complain) and to treat females a different way (until they individually complain), then go ahead.
But (speaking to Phil, now) don't cry about the unfairness of the existence of people who don't fit your stereotypes. Don't bitch that you are receiving mixed messages, when you receive them from mixed people. And definitely do not appoint one woman as spokesperson for womankind by asking what she wants "as a woman".
There is no way to get through life without sometimes offending people. Live with it. But still, do the best you can. Rules change. Keep paying attention, and you will keep learning.
I made my reply because your quip made sense only as a response to a straw man. All the worse because it so smoothly presupposes a position that is trivially and obviously silly.
Your presentation here is a more significant misrepresentation of Phil and this time it is one that is overtly poor form. In particular it is condescending and riddled with negative labeling while at the same time demonstrating that you completely failed to comprehend his words.
For that matter the re-framing of my statement as treating (males or) females a different way until they individually complain wasn't much better. "Act like a Bayesian" isn't exactly an outlandish personal habit for someone to have and complaints (hopefully) don't come into it. If taking into account everything you do know while you are still making first impressions doesn't reduce complaints then you are quite possibly doing it wrong.