Nick Boalch wrote:
I think this is probably something cyclical:
- IAR starts short and simple;
- IAR gets expanded by people trying to be helpful;
- IAR gets further expanded by people adding exceptions and corollories to the previous additions;
- IAR becomes a hideous monstrosity...
I've never paid much attention to IAR's convolutions or its talk page, so this may have been tried already, but I wonder if it would work if part of the capsule description were along the lines of, "Ignore any rule you can get away with." That is, rather than trying to prescriptively limn which rules can't be broken versus which ones can, and how (an approach which really does seem to demolish both the spirit and the letter of IAR, and especially that letter A), the notion would be that you can ignore any rule you want to when it's reasonable to do so and when it helps the project. How do you know whether it's truly reasonable and helps the project? If people thank you afterwards. What feedback mechanism is there to assist aspiring rule-ignorers who don't yet have decent intuition about what is and isn't reasonable, and are therefore liable to screw up and do something grossly unreasonable without realizing it? Why, the recriminations and flamewars that spring up whenever anyone does that, of course. Fearing those recriminations, a reasonable and conscientious editor will naturally and automatically shy away from dangerous rule-ignoring actions that they can't be sure would meet consensus.
Now, someone's going to say, "That's fine, but it only works for editors who *care* about not pissing people off, and who are interested in learning from their mistakes. It doesn't do anything to stop anarchic renegades." And that's true. But it's only the conscientious editors who would have read and tried to follow any detailed prescriptive rules and guidelines anyway -- the renegades would just ignore those, too.
(I realize I haven't said anything here that wasn't implicit -- or maybe even explicit -- in the IAR policy since day 1. I guess the point is to keep reminding ourselves, and the policy's critics, that yes, you really can ignore any rule when doing so helps the project, and yes, this is potentially a horribly dangerous policy which a determined rule-ignorer could really abuse, but no, that doesn't mean we're going to try to castrate the policy in order to prevent abuse; what we're going to rely on to prevent abuse is people's being reasonable, and the knowledge that most things can be undone. And it's true, people won't always be reasonable, but we'll deal with that as it happens, because trying to absolutely prevent it would cause more problems than it would solve.)