Re: Crocker's Rules (WAS: Women, fire...)

Eliezer S. Yudkowsky (
Tue, 28 Sep 1999 18:41:36 -0500 wrote:
> The problem is that these rules introduce tremendous friction when
> operating with someone who does not follow them, which is almost the
> entire world. It means that you are constantly giving offense to
> people, which leads to wasted time on your part as you try to correct
> the misunderstanding.

Once again: Crocker's Rules is not about what *you* say to *other* people; it is only and solely about what *other* people can say to *you*. It's about what you read, not what you write. It's about incoming communications, not outgoing communications. It doesn't say anything whatsoever about outgoing communications.

> It's also the case that most people will be unable to follow these
> rules successfully.

Exactly, which is why I don't quite understand the *universal* assumption, despite my explicit disclaimer, that Crocker's Rules involves assuming that other people are using Crocker's Rules. Crocker's Rules is a complex mental discipline requiring deep self-awareness, emotional maturity, and specific mental content. Assuming it would be silly.

Let me put it this way: Although getting the *benefit* of Crocker's Rules - other people being able to send you information in more efficient form - requires that you declare Crocker's Rules, it's equally possible to use Crocker's Rules with no positive external sign. It's just that if somebody calls you an arrogant bastard at some point, you don't get angry.

> That is, although they may try not to take offense,
> there will still be areas where they can be emotionally hurt. In a world
> where communicators care nothing for the sensibilities of others, such
> harm will be relatively common.

Again: It's not about the *outgoing* data, it's about the *incoming* data. In a world where communicators do not have sensibilities that can be hurt, emotional hurt will be uncommon. It's about listening, not speaking. Look at the phrasing.

> They of course the recipients try not
> to show it, or they even try to lie to themselves and deny their pain.
> This can lead to hidden resentments or even emotional problems.

Yeah, I wouldn't want people adopting Crocker's Rules as the result of peer pressure; if they weren't ready, they would probably permanently damage their chances of developing the discipline. It's a good point.

> In my opinion, the best policy to follow is the maxim from computer
> protocol design: be conservative in what you send and be liberal in
> what you accept. That is, try not to take offense, but try not to give
> offense either.

That's pretty much Crocker's Law in a nutshell. It's like a little sticker that says "You can't crash this protocol no matter what you send it" - not a declaration saying "This protocol may emit random line noise."

> Politeness exists for a reason, and I believe that denying it is going
> to impose costs which could have been avoided.

Politeness exists for a reason, but it has costs. If we can understand and counter the reason, we can optimize for efficiency instead of politeness, and recover the costs.

           Eliezer S. Yudkowsky
Running on BeOS           Typing in Dvorak          Programming with Patterns
Voting for Libertarians   Heading for Singularity   There Is A Better Way