Wikipedia:Ignore all rules (abbreviated to IAR) is perhaps one of the most controversial guidelines we have on Wikipedia. It is controversial because it goes against all the other policies and guidelines which we have. This is my essay on what I think is appropriate and inappropriate use of IAR.
I am not utterly opposed to IAR edit
Even though I have classified myself as having a Lawful Neutral alignment (follow the rules, for better or for worse), I do think that the idea behind IAR is correct. I am not someone who believes that we should follow the rules and policies, just for the sake of following the rules and policies, up to the point where it is ridiculous.
Example: Three revert rule exception edit
What?! I am going straight for, not a guideline, but an official policy? Yes that is precisely what I'm doing. The official policy does come with two official exceptions, people are allowed to revert their own userpage as much as they like, and they are allowed to revert "simple vandalism" more than three times. These situations were considered to be so common that a built-in exception clause was required.
Now for the IAR... There is no exception provided for administrators who routinely clear the WP:AIV page after blocking a vandal. Yet it would be utterly ridiculous if we started blocking administrators for consistently "reverting" back to the "LIST EMPTY" position. The reason we have the 3RR is to prevent edit warring, and the routine maintenance of AIV does not constitute that. Does it mean we should add this as an exception clause to the 3RR policy? I would say: Only if some administrator (with square eyes) actually starts blocking admins for it. Otherwise it will just be m:instruction creep. As it is, AIV works well and nobody has ever had any trouble with ignoring the 3RR in this case.
Note: Since this essay was written, the policy has indeed been updated. Now listed as an exception is "Reverting for maintenance". 06:14, 27 June 2006 (UTC)
Example 2:Speedy conclusions edit
Wikipedia:Articles for deletion/L. Paul Bremer was closed as a "speedy keep". Technically, it probably wasn't a valid speedy keep. The nomination was a good faith one, albeit an utterly misguided one, based on the article containing errors. It was also quite obvious from the debate that this was definitely going to be kept with overwhelming majority.
Nonetheless, it was technically an out of process deletion, and Rossami, an administrator who I greatly admire and respect, said that he would have reopened it had he seen it. (See the discussion). He has some valid points, but I think that invoking IAR on such a clear, overwhelming "keep" debate is no big deal. Following process would have led to the exact same result.
Note: Since the deletion debate above (Jan. 2005), the snowball clause has been introduced (Dec. 2005).
Example 3:Manual of style edit
The most frequent place where invoking IAR is justifiable is when writing articles. Wikipedia:Manual of Style is an awfully long document, and I think that nobody can be expected to learn it all in one go. If you are unsure of what the absolutely right style is, don't worry too much about it. When you write an article, just make sure it is verifiable (try to cite sources as well), contains enough information to make it useful, and do what "looks right" and be bold.
If you were right about the way you formatted your article, then that's good. If you were wrong, somebody will probably come along and correct it. Your contribution was still valuable. If you were wrong and nobody comes along to correct it, well then it was probably not a big deal anyway.
Responsible and irresponsible use of IAR edit
IAR can be a great danger, however, if it is abused. Those who choose to ignore a policy or guideline must be prepared to accept accountability for it. In particular:
Admins should be especially cautious edit
The administrators' deletion, protection and blocking tools are things which can only be reversed by another administrator. Usage of them is against the usual open-content philosophy of Wikipedia. It is the reason that we don't give them to everybody. When we entrust users with those tools, we trust them to use their new powers in a responsible manner.
Also in regular editing, admins (and incidentally other experienced editors) need to be cautious. Admins need to be good role models, new contributors often look to the veterans for advice. If we want the newcomers to follow the rules, it is best to lead by example. "Do as I say, not as I do" just won't work.
Process exists for a reason edit
The various processes we have, Articles for deletion may be the most familiar one, all exist for a reason. In the case of AFD, we have that process in order to give users a fair warning that an article might be deleted and to give the community a chance to have a say.
IAR should not be used as an excuse to subvert those processes. If users, and administrators are the ones with the tools to do so, regularly set about ignoring the outcome of those processes because they don't agree with them, we will have Wild West conditions.
Philosophically, let's assume that IAR meant that: "You may disregard the outcome of a process if you don't agree with it". To avoid hypocrisy, we would need to give that right to everyone. Well, several people have disagreed with the outcome of processes. I know I have. I voted to "delete" Wikitruth, and would like to see the article deleted since the chief claim to notability is being critical of Wikipedia and getting passing mentions in an interview with the website's owner. But the process showed a consensus to keep. Five times. If I were free to disagree with consensus and do exactly what I wanted, I would have had that article deleted. But if everyone did whatever they pleased with blatant disregard to what the outcome was, what was the point of the process? Therefore, "You may disregard the outcome of a process if you don't agree with it" is untenable because it means that the process has no meaning because anyone may disregard it. We might as well scrap all our processes, give admin tools to everyone and say "just delete and undelete whatever you like".
Breaking the rules can upset people edit
It has often been said that Wikipedia is an encyclopedia, and that it is not supposed to be a community. Fair enough, but the fact is that it is the community that has built the encyclopedia, and that the community consists of people. By breaking the guidelines which people have been advised to follow, you might find that those hurt by the breach feel that you have acted unfairly. People who want a page to be kept will be upset when you delete it. They will feel upset even if the page was deleted in process, but deleting it out of process will make them angry and make them feel worth less because it seems that their opinions don't count. Telling them to shut up, and not care about process will just make you seem outright arrogant.
On the upsetting people issue, there is no surer way to upset people than to block them, so make sure all the blocks are in process. Never use IAR as a justification to block someone!
When to ignore IAR edit
If someone is upset at your breach of rules, they will probably let you know. If a rule is brought to your attention, then that is a time to ignore IAR.
Conclusion edit
IAR is a way to cut through red tape and to avoid everything turning into a ridiculously inflexible grind. Use common sense. Wikipedians are people, not robots who need to be programmed precisely. But don't use IAR as a justification for breaking the spirit of the rules with complete disregard for process and people.