[dang]

You just need to move your game theory to a different level. The expected value of lying about such things is super negative and the expected value of telling the truth is super positive.

I'm sure there's a model in which lying some of the time but not too often has marginally higher expected value, but it's also going to have significantly higher risk and that's not worth it, plus you have to be disciplined enough to actually apply such a strategy. One slip and you're dead! I'm too lazy for that.

[paulcole]

My point is obviously that you’d tell a lie that looks like the truth people want to believe. Which when you get away with it is the biggest EV of all.

[dang]

Not once you control for risk, though. "When you get away with it" is a big assumption.

[paulcole]

You could 100% edit any post you want here and replace it with a link to a funny YouTube video and call it a fat-finger typo and get away with it. There’s no risk for you (as long as you don’t try it too often).

[dang]

That would be reckless. We must estimate risk differently.

Btw the more accurate reason why we don't do things like that is that we don't want to because it would feel bad. It's not who we want to be. However, the actual reason doesn't always have much persuasive power and when I'm sensing that's the case, I use the cynical argument ("not in our interests", basically), because it's also true. But as the cynical argument isn't persuading you, maybe I should switch back!

[paulcole]

I mean I’m confident we have 100% different views for “who we want to be.”

This site is a marketing campaign for a VC company. If you see it the same way, I’d be shocked.

[dang]

YC's a business and operates HN in the end for business reasons*. I don't have a problem with calling that marketing, but I'm puzzled why you bring that up in this context.

* https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...