Wikipedia talk:Policies and guidelines/Archive 6

Provfing wikipedia as a reliable source:
What policies should I include to prove to teachers that wikipedia is useful and accurate? In a project last year, wikipedia was allowed to be a source, but it didn't count as one of your five, which annoyed me as wikipedia was one of the only places to find info on fireworks - three out of my five were from here. 72.10.100.230 17:01, 31 August 2007 (UTC)
 * You should just not use Wikipedia as a source. Instead, use Wikipedia's sources as your sources. —Kenyon (t·c) 04:38, 2 September 2007 (UTC)

Rejected / historical
I'm not sure if I agree with Kevin's recent edit about rejected and historical pages. Generally, when a proposal gets little feedback after a month or so, either I flag it as historical, or he flags it as rejected (depending on which of us sees it first, I suppose). However, the "rejection" gets disputed by many editors whereas the pointing-out-inactivity does not, and many such proposals are reactivated at some point, which "rejection" appears to prohibit.  &gt; R a d i a n t &lt;  08:03, 3 September 2007 (UTC)


 * There's dormant too... I like the idea of historical for processes, dormant for inactive proposals, and rejected for opposed-and-not-improving proposals. The wording Kevin added to "rejected" proposals is not necessarily wrong, but it is rather cumbersome.--Father Goose 08:56, 3 September 2007 (UTC)


 * I agree with Radiant that the Historic tag seems to be a euphamism which is more palatible, and that "rejected" causes sensitivities. I also agree with FG that the wording is now clumsy.  Perhaps we need a better word to replace "rejected" while keeping the categories more clear.  I see "historic" as being an archive of something that was once in use.  --Kevin Murray 16:59, 3 September 2007 (UTC)

Reform. Policy, guideline, essay, proposed, rejected, historical, dormant, failed, draft
In general, the policy structure is confused. I think a serious restructure should be considered. --SmokeyJoe 07:56, 6 September 2007 (UTC)
 * There are core policies, with respect to other policies, but this page makes no mention of that fact. I would create core policy as its own class.
 * There is too great a variety and number of guidelines; In general that name is a misnomer, in that they do not offer “guidance” but instead define a rule, a rule that is sometimes an ideal or is applied non-uniformly. Guideline is not well defined.  Some are close to policy – the borderline is not well defined, and others are in the dirty and grubby overlap area with the ill defined Essays.  I would like to see core policy created, policy retained, rule-type guidelines shifted to policy, and guideline reserved for recommendations, not rules.
 * Essays are effectively undefined. There are some very good essays, carrying tradition and respect with the community.  Others are sensible statements of truth, enforceable by the merit of their logic.  Yet others are contested opinion.  Many more are low quality stuff of little purpose.  I like the idea of subclassifying Essays, with the userfication (userEssay of weak essays as the most obvious first step.
 * Proposed policy/guidelines are too frequent, too easy to create. The mingling of Proposed policy and Proposed guideline reinforces the lack of distinction between a policy and a guideline, and these should be separated for this reason.  Maybe, before proposed there should be draft idea, and it should not be allowed to be proposed until it is serious and stable.  Proposals would then be taken more seriously.
 * Rejected is worded overly strongly given how easy it is too apply. I am uneasy with Kevin Murray’s enthusiasm to tag rejected.  The wording implies to me that the rejection was firm and final and supported by consensus.  Such rejection should be reserved for serious rejects, and most proposals should be tagged with a new, softer failed.  I prefer failed to dormant.
 * Historical should be reserved for things that once were, but are no longer. Historical should not be applied to failures due to lack of interest, or because the issue was too difficult.


 * There's a number of good ideas here. I'll write down a bunch of responses.
 * I don't feel comfortable making the distinction between "policy" and "core policy"; in fact this distinction is pretty much irrelevant, not defined anywhere, and contradicted on other pages. Policy is policy, plain and simple.
 * Guidelines are meant to be recommendations, not rules. I believe that most of them are. If you know of any that are not, please point them out so that we can discuss them.
 * Essays are tricky, as discussed above. It would seem there are (1) opinion pieces, (2) guidelines masquerading as essays, (3) rejected proposals masquerading as essays, and (4) miscellaneous pieces that people put an essay tag on because they feel it needs some tag. I am of the opinion that (2) and (3) should simply be re-tagged, although there tends to be strong objection to that (the general codification fallacy). I am of the opinion that (1) and (4) don't really need a tag. Perhaps we don't need this tag period, especially as it is the most ill-defined of the bunch.
 * Proposals are almost without exception proposed guidelines. Perhaps the template should reflect that better. People who say they propose a policy generally are confused about the difference. Note that there have been virtually no new policies since 2004. I'm not sure if I see a meaningful distinction between "proposal" and "draft".
 * I don't see a distinction between "rejected" and "failed" either, although the latter sounds like a better term to use. I'd suggest renaming rejected to failed.
 * Likewise I fail to see a distinction between "historical" and "dormant". The main difference between "rejected" and "historical", imho, is that in the latter case you are welcome to try again, and in the former case you would be wasting your time.
 *  &gt; R a d i a n t &lt;  09:10, 6 September 2007 (UTC)
 * I could follow Smokey Joe almost right down the line on this, especially regarding the blurred distinction between policy and guidelines. I support "failed" replacing "rejected". --Kevin Murray 11:43, 6 September 2007 (UTC)
 * Replies to Radiant’s points.
 * #3 The notability guidelines read and function like rules.
 * #5 The intended distinction between “proposal” and “draft” is that a proposal would need to be serious and is worth a major announcement. “Drafts” would be useful as discussion pieces, for ideas needing refinement, and certainly not yet suitable to be tagged as a guideline.
 * #6 I thought that reject could be kept and reserved for the few serious but thoroughly thumped proposals (rejected by consensus), but, on the other hand, the talk page of the failed proposal should adequately reflect the strength of its failure/rejection.  --SmokeyJoe 12:44, 6 September 2007 (UTC)
 * (3) hm, that would indicate they need work. What do you mean precisely, are they not allowed to have commonsensical exceptions?
 * I think there is room of plenty of more work with that can of worms known as notability. Exceptions are not the problem.  The problem is that it’s a huge issue, and that notability is not treated as a guideline, but more as “comply or go to AfD”.  Failing the NPOV key policy is less likely to get your favourite article sent to AfD.  I suggest that this discussed should be continued at WT:N.  --SmokeyJoe 08:03, 8 September 2007 (UTC)
 * (5) yes, but ongoing proposals as well as accepted guidelines are still open to refinement. I don't think this is a good distinction, it encourages people to create one final version that might then not be modified any more.
 * (6) that is reasonable.  &gt; R a d i a n t &lt;  14:10, 6 September 2007 (UTC)


 * --Kevin Murray 14:21, 6 September 2007 (UTC) replies to above
 * #3 Agree that the notability guidelines that read and function like rules, and should be called rules or policies.
 * #5 I see the draft status as a safe-haven for producing the concept and attracting attention, without attacting opposition to adoption. However, along with this I would prefer to see the proposal process more clearly defined including a cooling-off period where drafts are required to have a minimum of X days of exposure to review with the proposal tag beofre gaining policy or guideline status.
 * #6 I don't strongly object to keeping reject as a stronger alternative to the failed tag, but would prefer fewer categories. --Kevin Murray 14:21, 6 September 2007 (UTC)


 * Ah, I think we've been here before. Proposal, at present, do not go through some kind of official procedure. Some people have expressed a preference that they do. These people are welcome to design a system that makes that possible; however, until such a system is consensually accepted, the fact remains that at present there is no procedure, just common sense.  &gt; R a d i a n t &lt;  14:30, 6 September 2007 (UTC)
 * I don't agree. The actual practice at WP is for guidelines to demonstrate consensus though the proposal process, not by design but by evolution.  That all guidelines have not followed this procedure should not be used to validate improper actions.  Common sense is subjective and personal.  --Kevin Murray 14:35, 6 September 2007 (UTC)
 * That's what you would like to happen. But, as you yourself admit, all guidelines have not followed this procedure that you assume exists, and it is not improper to arrive at consensus at a different way.  &gt; R a d i a n t &lt;  14:41, 6 September 2007 (UTC)
 * But that's all beside the point, because stating that an essay that gains consensus is a guideline does not make a judgment either way about how the essay is supposed to gain consensus. Note, by the way, that to my knowledge there has never been policy based upon an essay.  &gt; R a d i a n t &lt;  14:44, 6 September 2007 (UTC)


 * I think that the discussion of what becomes a guideline is better discussed at the guideline paragraph. The essay paragraph should be clear that essays have no authority (or call it not actionable etc) -- these are personal opinions of editors or a group of like-minded editors.  The recent versions seem to elevate the status in a confusing manner.  --Kevin Murray 14:53, 6 September 2007 (UTC)
 * Unfortunately, the present version is false. There are several essays that have authority, precisely because they have consensual support.  &gt; R a d i a n t &lt;  14:55, 6 September 2007 (UTC)
 * Can you please cite the examples? --Kevin Murray 15:00, 6 September 2007 (UTC)
 * Off the top of my head - WP:WL, WP:CREEP, WP:AADD. There are probably a handful of others.  &gt; R a d i a n t &lt;  15:16, 6 September 2007 (UTC)
 * Actually, I see these often referred to, but hardly ever enforced. They're just run-of-the-mill arguments around the wiki. Nobody gets sanctioned for wikilawering per se, but rather for other violations such as WP:POINT, WP:EQ and other relevant guidelines. Nobody gets sanctioned for WP:CREEP or stopped from doing it on this basis, but rather because a new "instruction" fails to achieve consensus in whatever project page or article talk page the instructions are proposed; it's just advice, another argument that's easy to refer to. Same with WP:ATA/WP:AADD. It's often referred to but is inconsistent with the actual guideline, which is WP:DGFA. And, people end up voting anyway, in keeping with longstanding deletion discussion convention. ... Kenosis 20:47, 6 September 2007 (UTC)
 * This is Radiant's creative use of essay/guideline creep here. Some editors may view essays like WP:AADD, formerly known as WP:ATA until it became notorious, useful or even choose to follow it, but that in no way should and could be construed as having "authority" or are in any way binding, despite Radiant's best efforts to imply otherwise. The every fact that they do not enjoy consensus enough to advance to the next level like guideline, is evidence that they are not only not widely accepted, but often widely contested. Odd nature 23:17, 6 September 2007 (UTC)

WP:WL is a good essay. It presents a relevant argument, with just some incidental advice given. The amount of advice is not enough to justify a guideline tag.

WP:CREEP is a fair, very brief, essay, followed by some good, helpful recommendations. The essay effectively incorporates an untagged guideline. I think that instruction creep is a bigger and much more complicated issue than the amount of detail WP:CREEP would suggest. There is scope on this subject, with research, to write a more expansive essay (also a better main space article Instruction creep), and to write a proper guideline that provides advice.

WP:AADD is a focused information repository. An essay or a guideline could readily be written on the subject, but WP:AADD is neither. In a formal document, I would put WP:AADD in an appendix.

I think we should here sort out what we think an essay and a guideline should be. As for draft versus proposed, I don’t think any firm rules should be applied. Wikipedia doesn’t need rules about rules about rules. WP:POL could simply suggest that early stage guideline ideas be tagged with draft. A proposed guideline should exist to attract attention, and after a week or so be accepted, rejected/failed or relegated to draft if editors consider it to have merit but not yet ready. Unlike Radiant, I don’t have a problem with a new guideline recommending new behaviours. None of this, however, should be considered rule. An experienced wikipedian with a good idea may well skip the draft step. If the idea is already in undocumented practice, then it could be appropriate to skip proposed. --SmokeyJoe 08:27, 7 September 2007 (UTC)
 * I have allowed myself to get confused between tagged pages and categorized pages. proposed and reject seem to be automatically categorized, but not historical.  Is this right?  There is also no-consensus, which I completely forgot about.  historical, no-consensus and dormant seem to be the same thing.  I think they should be merged and sub-categorized just like proposed and rejected pages.  In trying to make sense of the system of policies and guidelines and other things, the categories are very helpful.  --SmokeyJoe 07:52, 8 September 2007 (UTC)


 * Somehow I failed to notice dormant redirects to historical. I think 'historical' should be limited to "things we used to do but don't do anymore": WP:ESPERANZA, everything at Archive, etc.  I'd prefer scrapping rejected in favor of no-consensus for all lapsed and/or rejected proposals.  It's non-judgemental but still not a euphemism.


 * Draft vs. proposal provides no useful distinction. Mature proposals may still undergo extensive revision, and even guidelines and policies are constantly in flux.  If "draft" is meant to mean "we're still working on it, don't reject it yet", it will supplant proposal altogether as "proposal with rejection immunity".--Father Goose 17:58, 8 September 2007 (UTC)
 * I see the "Draft" status as a safe harbor to prevent undo criticism of the viability during the initial development. I see proposals as subject to change, but up to debate over the need for the policy as well as the content.  --Kevin Murray 18:20, 8 September 2007 (UTC)
 * I agree with your assessment of Historical. --Kevin Murray 18:20, 8 September 2007 (UTC)
 * I prefer SJ's "Failed" to "No-consensus", as a replacement of "Rejected", which sounds too harsh. --Kevin Murray 18:20, 8 September 2007 (UTC)
 * I don't like "Dormant", since I think to will become the new euphemism for failure and there will be constant hair-splitting. I want to see some closure to these proposals lest they just accumulate in perpetuity.  --Kevin Murray 18:20, 8 September 2007 (UTC)


 * At what point does "draft" become "proposal"? Criticism of a proposal's viability is appropriate at all stages.  As a putative proposal-maker (well, an actual one), I'm going to stick with "draft" until I'm ready for "guideline", since upgrading to "proposal" has nothing but drawbacks.  I'd switch to "proposal" only as a formality right before switching to "guideline".


 * Both "failed" and "rejected" don't take into account that consensus can change. "No consensus" is more neutral and I think less likely to be abused by all sides.


 * I agree that "dormant" is unnecessary, especially if no-consensus is available. Although I agree that the Wikipedia namespace is cluttered with dead proposals, that'll be true whether we use "failed" or any other tag.  If it really hasn't been worked on in months, let's corral everything into Inactive proposals.--Father Goose 00:03, 9 September 2007 (UTC)


 * In my mind, draft would be synonymous with Brainstorming. In fact, I see nothing wrong with "Brainstorming" for any new, unformed idea, including a policy proposal.  Did someone just create draft, or was it there before?  A brainstorm could become a proposal whenever something thought it appropriate.  I see no reason for why a brainstorm would be impervious to fatal criticism and being rejected/failed summarily.  --SmokeyJoe 09:21, 9 September 2007 (UTC)

Let's clear up some misunderstandings. As the five pillars say, wikipedia has no fixed rules. Everything is actually a guideline as per the dictionary definition. This fact is reinforced by Ignore all rules being defined as policy. IAR sounds twisted; it *is* twisted: it's a consensus-style workaround around people who want to make up rules and try to force them on everyone else ;-P

The project namespace was conceived as a place for people to write down their experiences, and document best practices, not to make up new rules. Hard rules are effectively that which is enforced by the Mediawiki engine ("only admins may delete images", "a bureaucrat may assign admin rights", "you need either a checkuser flag on the relevant wiki or an ssh key to the relevant mysql server be able to view IPs people have edited from, unless they edited 'anonymously'"). So interestingly enough, it is a little known fact that the actual rules are set by the developers (who may or may not be asked and/or paid by the wikimedia foundation).

While I have my own list of what each category means, I'd just like to point out that the current category system is very game-able, and has in fact been heavily gamed.

One of the worst offenders is "proposed". We checked, the proposal method simply Does Not Work in over 90% of the cases.

Some people still cling to "there must be a proposal system or wikipedia will die!"[1]. It turns out that practically all currently used guidelines (whether they're actually marked as policy, essay, guideline, or whatever) were written as documentation of actual on-wiki behavior, they were never "proposed".

The snowball clause is a famous example of a common wikipedia practice that gets applied in ... actual practice. Some people seem to think that changing the tag on the top will somehow change people's behavior (perhaps they were thinking of Sapir-Whorf, somehow?). After much tag-warring, I think we can empirically establish that that hasn't happened.

If you do want to change people's behavior the best method I've encountered so far is to be BOLD and set a better example, convince people that you are right, and come to a (new) consensus, and then finally you can document what everyone now agrees on.

--Kim Bruning 05:19, 9 September 2007 (UTC) ''[1] Did you know that Esperanza and the Association of Members Advocates said their processes were unmissable for wikipedia? At the time many people were convinced it was true. ''


 * I don't mind if there's some kind of change to the "proposal system" (actually, I'm not convinced we have one, template or no). Editors will still make proposals, and they damn well should, even if 90% of them go nowhere.  Separately, changing the tag on top can make a very big difference: it's hard to get people to discuss something marked "dead".--Father Goose 09:18, 9 September 2007 (UTC)
 * With due respect, that's pure blind faith in something that hmm, well frankly doesn't deserve it. Editors should be editing the encyclopedia, and maybe occasionally documenting their best practices. This is what works (and it would also save me a HECK of a lot of my time if they just stuck to that :-p ). --Kim Bruning 03:16, 11 September 2007 (UTC)


 * Entirely consistent with Kim’s point, I don’t see the restructure of WP:POL as aiming to change behaviour by imposing rules, but to clarify classifications. Accordingly, the categories are more important than tags.  It would be nice if WP:POL gave helpful guidance for what a policy/guideline, etc, ideally should be, to help with comprehensibility, but not so much to constrain behaviour.  --SmokeyJoe 09:28, 9 September 2007 (UTC)
 * I would agree. We should not be inventing new prescriptive rule-sets, but we should be seeking continuity which reflects the demonstrated consensus. I would only differ in having an expectation of constraining behavior where it is a major departure from expectations.  --Kevin Murray 15:09, 9 September 2007 (UTC)
 * "having an expectation of constraining behavior where it is a major departure from expectations". That's a fairly abstract concept, can you explain what you mean? --Kim Bruning 03:16, 11 September 2007 (UTC)
 * Kim, I know that you are a smart guy; when you read that phrase again in the context of the paragraph and the discussion, the meaning will be evident. Otherwise, please contact me on my talk page, and I'll paraphrase the concept for you.  --Kevin Murray 03:29, 11 September 2007 (UTC)
 * If the meaning were evident, I wouldn't have to ask.:-) There's 2 or 3 different ways to interpret it, and I don't want to read you wrong by accident. --Kim Bruning 03:55, 11 September 2007 (UTC)

Historical WARNING
The historical tag was originally intended (and probably should still be used for) pages that we have stopped using for whatever reason. We would like to document past mistakes. Short of vandalism and such, pages in the project namespace should practically never be deleted, but instead be kept as a kind of archive of past ideas, discussions and decisions.

WARNING ABOUT DELETION

--Kim Bruning 05:25, 9 September 2007 (UTC)
 * I agree with the theory here, but in practice I think that we build-up a lot of obsolete pages, and a lot of energy gets expended periodically when someone new decides to resurrect a failed concept. Might we be better off having a summary page of rejected proposals and delete the pages from the general view. My understanding is that deleted pages can still be accessed by Admins on request.  However, I do believe that Kim's position does represent the present consensus.  --Kevin Murray 15:24, 9 September 2007 (UTC)


 * Tons of energy is expended (wasted) constantly on all aspects of Wikipedia. Failed concepts should be documented, not erased.  People will resurrect them anyway, unaware of the prior attempts.  Wikipedia thrives on its openness -- "closure" is an illusion here.  Oh, and read that "Warning about deletion" bit above -- deletion is deletion.--Father Goose 17:17, 9 September 2007 (UTC)
 * Don't worry! I don't feel that strongly about it and consensus is not likely to change.  I nominated a failed proposal for deletion about 6 months ago and the consensus at a broadley discussed MfD was to keep.  YES, "closure is an illusion" at WP and I am a realist (I hope).  Cheers to all and lets' enjoy the rest of a beutiful weekend.  --Kevin Murray 18:40, 9 September 2007 (UTC)

So to recap
The actual issues appear to be that (1) people want more tags for what they perceive as accurate categorization as pages; and (2) people who disagree with some common practice believe they can stop this practice by calling its description an essay.

Based on (1), new tags are made like "draft", "no consensus" and so forth, that turn out in practice to go unused. Based on (2), people really really want to define essays as "not being enforced and having no authority", because there are authoritative, enforced pages that these people want to deprecate, and they believe they can do so by putting a tag on top. Needless to say that doesn't actually work.

I don't think either debate is getting anywhere.  &gt; R a d i a n t &lt;  12:22, 10 September 2007 (UTC)

Failed is equivalent to No-consensus;

Dormant is redirected to Historical;

Draft (my intention) is equivalent to Brainstorming;

Draft as already exists is something different;

Proposed contains, and tends to contain, too many ideas for too long. I suppose the thing to do is to get in there and stir them up.

In the end, I think we should leave those tags alone, but can we add automatic categorization to No-consensus, Historical and Brainstorming? I’d like to review such tagged pages.

Essays are still a problem, with diamonds lost in the dust. While hesitant to suggest more tags, I think I have to. I would like to see UserEssay and GoodEssay (or something like that), with both subclasses automatically categorized. UserEssays are opinions of the user, which they wish to advertise to the community through the subcategory. GoodEssays should be consensus supported, high quality, literary efforts. This may be pretty complicated in practice.

I’m not sure what I think about the guidelines. Guidelines are helpful things, until there are so many that you get tangled up in them. I think guidelines should be recommendations and not enforceable. Some guidelines are enforceable (eg WP:N, enforceable by article deletion). These enforceable guidelines are not really guidelines, but they are not policy either. --SmokeyJoe 14:20, 10 September 2007 (UTC)
 * What, in your opinion, is the difference between "GoodEssays" and guidelines? I don't believe that guidelines are really "enforceable" the way you state, because it is quite acceptable to nominate a page for deletion without using guidelines (e.g. based on an essay, or your opinion, or whatever).  &gt; R a d i a n t &lt;  15:17, 10 September 2007 (UTC)

Not-good essays would be things that are in poor shape, are disputed, or are non-consensus opinions.

Two quite different GoodEssays, which are not good guidelines, in my opinion, are What adminship is not and User:Uncle G/On notability. Both are clearly relevant to wikipedia, and so should be prominently catalogued. The first is statements of fact/information/opinion/logic, but is not a guideline because it is not attempting to guide any action, in an immediate sense. The second is a well structured, logical argument, and it strongly recommends particular actions or inactions, but is not a suitable guideline because it is not to-the-point. As a guideline, it gets too bogged down in rationale. As an essay, it is a good reference for a guideline.

A large number of essays are not really essays, but are explicit recommendations, or guidelines, according to the real-world usage of the terms. One example is Don't stuff beans up your nose. A whole lot of these essays could be simply made guidelines right now, but I guess that we don’t do that out of fear of further confusing those guidelines that are pseudo-rules.

Your argument that rule-type essays, such as WP:N, are not enforceable is disputing the converse. I don’t say that enforcement requires a guideline. I say that deviation from a notability guideline can lead to an article being both nominated, and deleted, where in the absence of the guideline, the article would have been overlooked. This is enforcement.

An example of a rule-type guideline and a helpful-type guideline covering the same thing is Notability (fiction) and Manual of Style (writing about fiction). The first is enforceable via article deletion. The second advocates a different style without implied threat of enforcement via deletion. As the first is effectively an enforceable rule, rather than a recommendation, I believe it should be held to a higher standard, and not be allowed to masquerade as a “guideline”.

--SmokeyJoe 01:32, 11 September 2007 (UTC)
 * Heh. In essence you're saying that they aren't guidelines because they aren't actionable. Which, in my opinion, is true, but people object to that definition. With respect to "A whole lot of these essays could be simply made guidelines right now", I suggest that we do just that, to alleviate further discussion.  &gt; R a d i a n t &lt;  08:55, 11 September 2007 (UTC)
 * Did I say that? UncleG/On Notability is actionable (your usage, I think) and yet is a good essay and would not be a good guideline.  I think that your usage of actionable is: recommends an action, whereas a common usage is: contains a rule that when breached can be enforced.  There are arguably too many guidelines, but if a particular essay is really a guideline, then it should be accepted as a guideline, and appropriately tagged&categorized .  I agree.  --SmokeyJoe 14:25, 11 September 2007 (UTC)

Essay "has no authority"
That's an odd thing to say, seeing as that some essay pages *do* have some level of clear consensus. Since "Essays have no authority" is clearly untrue in the general case at this point in time, I've removed that phrase. There's other issues too, certainly, but those are less easy to defend :-P

And no, I'm not trying to "change the status of essays" or anything like that. Policy pages are descriptive, not prescriptive, so you need to describe current practice, not what you would personally like to change it to.

If you want to re-add the phrase, please provide evidence that no current essays have any authority. (Do be aware of cases where people would like to see certain pages marked as essay that currently are not...) (see also: rationalization proposals above)

--Kim Bruning 07:19, 11 September 2007 (UTC)
 * Indeed. Clearly several essays do have authority. One might argue that this defies the point of "tags", as the main reason for using the tags is to indicate which pages do (and do not) have authority.  &gt; R a d i a n t &lt;  08:53, 11 September 2007 (UTC)
 * If these pages have authority then they should be tagged as either guidelines or policy. I thought that we were trying to become more specific in the definition of "essay", but the recent edits seem to lump everything into essay as a default of not meeting another of the criteria.  And then we further confuse the issue by implying or stating, depending on the version, that an essay could be equivilent to a guideline or policy.  Why bother saying anything at all? --Kevin Murray 14:27, 11 September 2007 (UTC)
 * Again: toss essay and go back to notpolicy.--Father Goose 17:53, 11 September 2007 (UTC)
 * The problem is this: whether we use the "older" wording or a "newer" more specific one, there are at present several pages marked "essay" that do not meet the definition of "essay". The solution is probably to fix those pages, though.  &gt; R a d i a n t &lt;  14:45, 12 September 2007 (UTC)

I think there are three types of authority in Wikipedia: So I don't think it's exactly true that essays "have no authority". I'd say that they may have persuasive authority, but are not necessarily binding. CO GDEN  18:13, 11 September 2007 (UTC)
 * 1) Codified authority. Policies and guidelines have this type of authority. Based on consensus and widespread practice, rules and guidelines are codified that are intended to be binding.
 * 2) Precedential authority. These are practices, based on the way things are usually done in Wikipedia that might have the effect of rules, and might be binding in some cases, but aren't necessarily codified in a policy or guideline.
 * 3) Persuasive authority. This is the category in which essays and proposals reside. They are not binding rules (yet), but they may have authority based on their persuasiveness. Essays and proposals may be cited as authority based on arguments contained therein.
 * RE the statement that "rules and guidelines are codified that are intended to be binding." : No. "Policies" are intended to be binding. "Guidelines" are intended to be just that, guidelines that have a weight of authoritcy but are not necessarily binding, and which can be trumped by a local consensus for good cause. "Essays" have no inherent authority, although they can serve as a good summary of complex or recurring arguments. If users wish to assert that part or all of an essay should be termed a guideline, it should be templated as a proposed guideline and subject to discussion to determine whether its status will be raised to give it that level of advisory authority. Otherwise they have no weight of authority, but rather are an opinion or argument in support of a particular position that some users may wish to take. ... Kenosis 20:04, 11 September 2007 (UTC)
 * The page currently says that guidelines are "actionable", which means that they are binding. I think that reflects current practice, because I've seen lots of times where articles have been changed and/or deleted under the presumed binding authority of such guidelines as WP:RS and WP:Notability. I think the difference is that for guidelines, a certain degree of judgment and skill is required in the execution. They both reflect Wikipedia policy and consensus, however. CO GDEN  20:41, 11 September 2007 (UTC)
 * Sometimes they represent consensus; guidelines do not represent policy, they are rough guides (sometimes very rough guides) on how to implement policy. Septentrionalis PMAnderson 18:23, 17 September 2007 (UTC)

Essays have no authority. If they become generally accepted by the community at large, it is time to discuss making them a guideline - which has virtually no authority, as a guideline is for guidance, as the name implies. Policy has authority. One puppy's opinion. KillerChihuahua?!? 20:18, 11 September 2007 (UTC)
 * No "binding" authority, but I think you could say they have "authority" in the sense that they might make good arguments, and they might be cited. I've seen lots of essays cited as reasons for making changes in articles. If it's cited, it must have some degree of persuasive authority, just not binding authority. Maybe that's just splitting hairs, but I wouldn't want to discourage editors from citing essays. If an essay is cited enough as a persuasive authority, it might eventually become a binding authority because it reflects consensus. CO GDEN  20:41, 11 September 2007 (UTC)
 * It seems that when the essay has authority it should be labeled a guideline if consensus is demostrable. However, if the essay is cited frequently but upheld rarely that is not evidence of consensus.  --Kevin Murray 22:21, 11 September 2007 (UTC)
 * Citing an essay and winning the argument on that basis is like getting a boilerplate demand letter with a fancy letterhead and having the recipient think "well, it sure sounds authoritative to me". That doesn't make an essay authoritative. Note also the language of the guideline templates.  Guidelines have a measure of authority, but it is purely advisory authority that has been agreed to be called a WP guideline. ... Kenosis 02:14, 12 September 2007 (UTC)
 * Chihuahua and Kevin are both correct that such authoritative "essays" should be re-tagged, ifo only to clear up misconceptions. Examples of such pages include WP:WL, WP:CREEP, and WP:AADD, all of which are consistently upheld (with the occasional exception, as always).  &gt; R a d i a n t &lt;  14:45, 12 September 2007 (UTC)


 * Essay: A discussion of traffic lights, their history and their value/importance etc, based on one person's opinion.
 * Guideline: You really shouldn't go through yellow lights, but you won't get a ticket if you do.
 * Policy: Do not go through red lights. You will get a citation and might kill or be killed in the process of violating this policy.
 * Get it?  &#0149;Jim 62 sch&#0149;  22:22, 11 September 2007 (UTC)
 * Looks clear to me! --Kevin Murray 22:28, 11 September 2007 (UTC)
 * Perfect. Yes, that is it in a nutshell. KillerChihuahua?!? 22:46, 11 September 2007 (UTC)
 * I'd modify the first example to a: "a personal suggestion that the standard colors of lights be changed. or a discussion .... " That fits in more with the sort of problems in how strongly to consider essays that was discussed above. DGG (talk) 16:09, 13 September 2007 (UTC)
 * Except on wikipedia, "Don't go through a red light or you're dead meat" is almost always marked "essay". --Kim Bruning 23:31, 14 September 2007 (UTC)

Note that nothing is binding, because consensus can change. Just to make your day and drive the point home, we also have a policy (ironically enough) called Ignore all rules, and yes the Arbcom can, has and will let you do that with impunity. --Kim Bruning 23:31, 14 September 2007 (UTC)
 * I think a reasonable solution is (1) deciding on a workable definition of "essay", and then (2) doing something about the pages marked as essay that do not fit that definition. The only real alternative is admitting on this page that "essay" isn't a well-defined term. I strongly object to nothing that "essays have no authority" as long as people use essay on pages that they would like to have no authority, but actually do.  &gt; R a d i a n t &lt;  13:49, 17 September 2007 (UTC)
 * I agree. An idea for how we might describe good, project space essays:
 * A project space essay must be relevant to wikipedia, whether it is reflective of wikipedia’s history or current trends, or persuasive on how things might better be done. It should be well written and coherent, even if it is still a work in progress.  It should be more than mere assertion, ideally combining facts and logic.  A good essay will not be readily disputed.  If an essay reflects a minority or non-consensus opinion, then it should be worded to so as to indicate this.  (eg “Some wikipedians believe …”)
 * I have little problem with an essay carrying “authority” due to the force of its own reasoning. --SmokeyJoe 14:57, 17 September 2007 (UTC)
 * I think I am not alone in wanting to see every page in the Wikipedia namespace have a tag simply to distinguish policy from notpolicy. Having to tag every page is ugly, but, I feel, necessary, because people will say "per page in Wikipediaspace" to support any argument, even when the page in question might be total bollocks.  (I exclude Wikiprojects from tagging, because their status as notpolicy is apparent in their name.)  essay became the de facto replacement for notpolicy, so there are plenty of essays that aren't essays.  Is it really useful to distinguish essay from notessay when they are notpolicy in both cases?
 * I understand Radiant's concern: people either try to to "demote" guidelines that they don't like to essays, or treat rejected proposals as essays. Again, why can't we just bring back notpolicy?  essay is a dinkelhammer, and we're trying to treat it like a flitzendauber.  Why is it necessary or useful to say anything more than "this is not a policy or guideline, full stop"?--Father Goose 20:58, 17 September 2007 (UTC)
 * There is a lot of accumulated wikipedia experience, even wisdom, and it will increase (in quantity, if not density). Essays are a good way of recording this stuff.  I’d like them to stay, but to have some kind of an expectation of quality.  What is dinkelhammer?  What is flitzendauber?  --SmokeyJoe 00:45, 18 September 2007 (UTC)
 * Two nonsense words... I used "dinkelhammer" earlier on this page, so I was extending the joke. I'm not saying we should get rid of the essays at all, just the need to call them "essays" -- "essay" is as much a nonsense distinction as the nonsense words I used.  Essays are simply non-policy pages.--Father Goose 03:50, 18 September 2007 (UTC)
 * The problem is, FG, that if we replace "essay" with "notpolicy", there will be people who try to "demote" guidelines that they don't like to notpolicy.  &gt; R a d i a n t &lt;  09:21, 18 September 2007 (UTC)
 * The recent attempts by Radiant to override this discussion based on a much earlier language that went largely unnoticed neglects that this particular page never discussed the language in depth. The discussion that began in July 2007 attempted to refine the language in such a way that it would be clearer that essays need to go through a process and be posted as "guidelines" so as to invite the scrutiny of the broader community before being cited as having a weight of authority beyond simply the argumentative position put forward in the essay. I've reverted it back to what came out of this year's participatory discussions here. ... Kenosis 10:43, 18 September 2007 (UTC)
 * You're missing the point. There are several essays that do have authority, therefore stating on this page that essays do not is incorrect. That is not "what came out of this year's participatory discussions", that is simple falsehood. Neither does the "process" you allude to actually exist at the moment. What actually came out of these discussions is that the term "actionable" is confusing, and that people would like a solid definition of what "essay" means.  &gt; R a d i a n t &lt;  11:17, 18 September 2007 (UTC)
 * Then put them up as proposed guidelines. ... Kenosis 19:46, 18 September 2007 (UTC)
 * OK, this comment has two parts, and you may agree with one and not the other, so I'll mark them separately.
 * PART I:
 * To quote you from earlier, Radiant,  'Chihuahua and Kevin are both correct that such authoritative "essays" should be re-tagged, if only to clear up misconceptions' . That being the case, can't we agree on a wording saying 'essays are not authoritative', then propose that any that we feel should carry authority as guidelines or policy?  This seems like a better solution than saying that we cannot adopt a definition of a category until all pages in the category fit it (as there is no reason to move pages out of that category unless they do not fit it). TSP 14:47, 18 September 2007 (UTC)


 * PART II:
 * It seems to me like we may be working with multiple definitions of 'authority'. If a railway worker tells me not to walk on railway lines, he has, by my definition, authority; I am expected to follow his instructions, whether I agree with them or not.  (I can dispute his use of that authority, but I should follow it until I can get it changed.)  If a passer-by tells me not to walk on railway lines, or not to stick my fingers in a live electrical socket, he has no authority - he has not been granted any right to instruct me.  The passer-by is right, most people would agree he is right, and I would be foolish not to follow his advice, but that does not mean he has authority.
 * In Wikipedia, consensus carries authority; so in a particular situation, a consensus decision (whether it refers to policies, essays or neither) has authority. Also - as a handy shortcut - policy and guidelines carry authority of their own right; once something is adopted as a policy, it should be assumed to represent consensus until the converse is established.  This is to avoid having to establish consensus for, for example, the principle that facts shouldn't be added without a source, every time the issue comes up.  This authority may be challenged at any time, but while something stands as policy, that gives it authority of its own right.
 * My understanding of our policy structure is that, if something has received consensus to be adopted as a guideline or policy, this grants it authority; we should (in the absence of other imperatives) try to follow it, and we are right if we query others for their failure to follow it. If something hasn't been through this process, then we may agree with it and follow it, and we may refer people to it as a good argument which applies to a particular situation, but no-one is expected to take a particular actions simply because the essay says so.  Similarly, our article on begging the question, for example, is accurate, would probably be generally agreed to be, and it might sometimes be appropriate to refer to it in a debate; but it's not policy.  It may have consensus as being accurate, but it doesn't have authority as an instruction.
 * If there are essays containing instructions, which we expect people to follow because the essay says so, rather than simply because the essay makes some good arguments (perhaps referring to policies and guidelines) then they should be made into policies or guidelines.
 * I can't agree with a situation where users are expected to follow - in addition to local consensus - policy, and guidelines, and some essays, but with no way to find out which. This seems like a hopeless situation to leave new users in.  "Why aren't you following that essay?  Didn't you know that that is an essay with authority?"  If we expect people to follow things, they should be tagged in a way that indicates this.  Until they have gone through whatever consensus-building process we go through to tag them as such, I don't consider them to have authority per se, even if they have consensus as being accurate.
 * Am I making any sense? TSP 14:47, 18 September 2007 (UTC)
 * But that is what the situation has always been. The only "authority" we recognize is consensus; policies and guidelines are only places to conserve past consensus, so we don't have to reinvent the wheel every time. (ArbCom and the Foundation are partial exceptions to this, but they attempt, usually, to listen to consensus; when Jimbo doesn't, the result is usually a lasting unpleasantness, like the Great Userbox War. (It may be useful to add to the policy, if we are going to stick with the present definitions, that tagging some page guideline doesn't make it one, if there is no consensus to support it. Septentrionalis PMAnderson 16:45, 18 September 2007 (UTC)
 * I agree. I also agree that this is how it has always been.  What I am trying to address is the assertion that some essays have authority.  I don't believe this to be the case - they may represent a consensus, and that consensus has authority; but the only documents that have authority of themselves (for as long as consensus to uphold them exists - i.e. until consensus is to remove their status) are policies and guidelines.  If there are essays which authors believe should have authority of themselves, they should be proposed as policies or guidelines.  TSP 19:05, 18 September 2007 (UTC)
 * An interesting irony here might occur if WP:CREEP were put up as a proposed guideline. But, I don't mean to complicate the disucssion, only to lighten it up a bit. ... Kenosis 19:51, 18 September 2007 (UTC)
 * To follow up more directly: An essay may, on certain occasions, be said to have consensus among the participants who write the page, and/or who comment about it, and perhaps even among those who view it, though the last of these is more difficult to clearly discern. Raul's laws, for example, may be said to have consensus. But the process of making an essay a proposed guideline will invite broader participation and scrutiny from more of the broader community, and turning it into a guideline and putting it on the list of WP guidelines will invite yet more scrutiny and perhaps further participation. Same with a proposed policy. Recalling the WP:A proposed policy, it had a fairly strong consensus among its participants, was raised to the status of policy, and upon being scrutinized by the broader community was fairly promptly removed from policy status. An essay, if its proponents wish to give it the greater weight of advisory authority of a guideline, should propose it as a guideline and allow it to be subjected to a greater level of scrutiny before citing it as authoritative. ... Kenosis 20:08, 18 September 2007 (UTC)


 * Radiant: while I understand many of your concerns here, ultimately I feel you're just not taking the right approach. People who want to "demote" guidelines will do it regardless of how "essay" is defined (they were doing it back when essays were defined as non-actionable).  I don't really see any danger from this behavior: if a guideline has consensus, it will retain its guideline status.  If it doesn't, it will get "demoted" one way or another, and I really don't care if they do it by calling it "notpolicy", "essay", "rejected" or anything else.  "Notpolicy", however, is a vastly better name for how essay has been used Wikipedia-wide.  The more we try to define "essay", the more we end up chasing our tails.  All people need to know is "this is not a policy".  Any generalizations we make beyond that are wrong and unhelpful.--Father Goose 04:46, 19 September 2007 (UTC)
 * Father Goose, I cannot agree that essays should be nothing special beyond notpolicy. Essays are a good way to explore, express and record wikipedia wisdom.  Knowledge that gets repeated over and over again belongs in an essay, and the existence of CAT:E is important in making essays accessible.  Above, I’ve given a suggestion for a summary for “Essays”.  Note that it is not a “definition”, but a goal that I would suggest essay writers should aim for.  I agree that essays will be effectively impossible to define tightly, and trying to do so is not a good idea.  As for dinkelhammer, I was pretty sure it was a nonsense word, but flitzendauber almost translates from German to something meaningful.  --SmokeyJoe 13:47, 19 September 2007 (UTC)
 * It is true that "not-policy" can be properly applied to "essay". But it's also true that "not-polilcy" can be applied to "guideline". That is why there have been these three basic categories instead of just "policy" and "not-policy". Anyone can write an essay, as we all know-- some are more known than others, and some are more respected than others. If one wishes to begin quoting an essay as having the weight of advisory authority attached to the word "guideline" (see the templates used for guidelines), it should be templated as a "proposed guideline" and undergo the additional scrutiny. Then, if it is upgraded to "guideline" and placed on the list of guidelines (see the list of guidelines on the WP:PG project page), it will invite even broader scrutiny as it gets significantly more noticed and begins to be much more applied in practice. If the broader consensus disagrees that it should retain the status of guideline, the consensus will tend to show itself as disagreeing with the smaller group of participants that led to its being raised to "guideline". In the future, the word "guideline" quite possibly may be expected to need a more involved process to try to avoid the glitches and arguments that may occur upon its being raised to the quasi=authoritative status of "guideline" and placement on the list of guidelines. Or in the future it may become agreed to need a stated "trial period" -- I have no way of knowing what approach will be taken to try to avoid the occasional messes that occur when a project page is raised in status, then rejected and lowered again. But for now, an "essay" is still expected to go through a process, not merely be cited by anyone who cares to as having consensus and being a de-facto guideline. The most significant reasons for this are that the phrase "de-facto guideline", or similar expression (1) dilutes the meaning of the word "guideline", and (2) inflates the actual amount of consensus the essay in fact has achieved. ... Kenosis 15:22, 19 September 2007 (UTC)
 * SmokeyJoe: I'm not advocating to get rid of essays, or even the category: I just think we should drop the word (and concept) "essay". It's a word someone chose about a year ago to substitute for notpolicy, only we're using essay on plenty of non-essay pages to signify that they're not policy (or guidelines).  Yesterday I was bold and removed the word "essay" from the text of essay, and it works just as well, if not better.  I think we could rename the template notpolicy (or, let's say notarule) and start extracting ourselves from the contortions that this word "essay" created.
 * Kenosis: When I say "notpolicy" I mean "notarule", and refer to both policies and guidelines. Wikipedia has policies in the English sense (guidelines) which are not Policies in the Wikipedia sense (policies), and I apologize for the confusion that results from that.  Again, I'm not advocating to get rid of anything, just to rename the whole essay morass to notarule so that we don't have to struggle with definitions (or even the very name) that doesn't match reality.--Father Goose 20:13, 19 September 2007 (UTC)
 * Here's a sampling of the history of this page:
 * Early January 2003
 * 9 January 2003
 * 29 June 2003
 * 6 January 2004
 * 29 June 2004 underwent major changes
 * 1 January 2005
 * 30 March 2005
 * 2 April 2005 significant changes
 * 29 June 2005
 * 1 October 2005
 * 20 December 2005 interesting edit by Radiant!
 * 1 January 2006
 * 27 January 2006 New section entitled "The difference between policy / guideline / essay / etc" inserted by Radiant
 * 3 March 2006 The words "authorized by consensus" added to the description of "guideline" and "essay"
 * 30 March 2006
 * 30 June 2006
 * 29 December 2006
 * 9 July 2007 The description of "essay" comes under scrutiny
 * 13 July 2007 The description of "essay" is amended based upon talk page discussion
 * Then, after another six-and-a-half weeks the issue of "essay" came up again, having been edited numerous times and been discussed in this talk section and the section(s) above. :Then, after another six-and-a-half weeks the issue of "essay" came up again, having been edited numerous times and been discussed in this talk section and the section(s) above. (Here's the recent history)
 * ... Kenosis 20:30, 19 September 2007 (UTC)


 * In response to TSP above,
 * "That being the case, can't we agree on a wording saying 'essays are not authoritative', then propose that any that we feel should carry authority as guidelines or policy?" I think we should do it the other way around, because I know of some pages that have authority, but of which people object to "call it a guideline". WP:AADD is the first example that comes to mind. If we can't convince the people on the talk page there that "the page is not an essay since it has authority" it follows that said definition is incorrect.
 * "a consensus decision (whether it refers to policies, essays or neither) has authority". Precisely. Therefore, stating that an essay "carries no authority, regardless of whether it represents a consensus" is false. If it represents consensus, it carries authority. As PMAnderson states, putting up a tag is the result, not the cause, of the text of the page having authority.
 *  &gt; R a d i a n t &lt;  11:05, 20 September 2007 (UTC)


 * I would tend to separate the concepts of 'has authority' and 'agrees with a consensus (that has authority)'. Wikipedia often forms consensus on things.  If these things are big, important, and universal enough, we form them into policies or guidelines, which have their own authority (for as long as consensus is to maintain them as such).  People can see they have authority because they're marked as Policies; people should (unless they have good reasons, bearing WP:IAR in mind) abide by them.  This means that users actually have some chance of knowing what is a policy they should be abiding by, and what isn't.
 * I think that "having authority" goes beyond "is generally agreed with". If someone wants a page to have authority, then it should be proposed as a policy or guideline - that is, proposed to have authority.  Then people can look at it in that light.  Someone might perfectly reasonably look at a page and say, "Yes, that makes good observations that are relevant in this case", without also thinking, "That page establishes universal rules which should be followed throughout Wikipedia".
 * I see no need for, say, WP:AADD to have authority. It gives advice, referring to policies where necessary.  I agree with it, and I think most other people will, but then I also agree with WP:WELCOME, and think it gives good advice; that doesn't mean it needs to have authority.  It's just advice.  Granting authority - or a hazy state of might-have-authority - to every piece of good advice seems to me to be definite WP:CREEP.  I'd rather say that if it's been agreed to be a policy or guideline, it has authority.  If that agreement has not occurred, it does not - no-one should feel obliged to follow it, but of course they should feel obliged to follow consensus at the time, which may in that case agree with it.  I don't see a need to say that anything might have authority, unless it has gone through some process to establish that authority.  Consensus (established per issue) is our overriding authority; policies and guidelines, and only they, are our ongoing 'stores' of consensus; and we keep them to what is necessary, to avoid WP:CREEP.  TSP 11:43, 20 September 2007 (UTC)
 * Yes, but what I object to is stating about pages that might have authority that they absolutely do not have authority.  &gt; R a d i a n t &lt;  11:58, 20 September 2007 (UTC)
 * But I don't consider that any essay has authority. I think that authority should only be seen to rest in consensus, obtained for individual issues, and in those things to which authority has been specifically granted - policies and guidelines (and bodies like the Arbitration Committee).  Saying that authority, or possible authority, may exist elsewhere, seems to me like instruction creep (which I feel is a bad thing, but don't think we need a rule about).  Consensus may happen to support an essay (or equally exist where there is no essay), but it's the consensus that has authority, not the essay.
 * Not having authority doesn't mean that something isn't right, or even isn't generally agreed to be right; but in my view the only documents in Wikipedia that are authoritative are polices and guidelines. TSP 12:23, 20 September 2007 (UTC)
 * Actually, does that make it clearer? For me it is more obvious that essays 'are not authoritative' than that they 'do not have authority'.  Is that wording any better for you?  TSP 12:29, 20 September 2007 (UTC)
 * An essay may carry authority by virtue of its own strength of argument. A logical fact may be proven by an essay.  The essay is then an authoritative reference for that fact.  I agree with Radiant and disagree with TSP.  --SmokeyJoe 14:26, 20 September 2007 (UTC)
 * Fair enough. That's not what I think the word "authority" means, but if others do then perhaps it's a word best avoided. TSP 14:48, 20 September 2007 (UTC)
 * TSP wrote earlier: “If there are essays which authors believe should have authority of themselves, they should be proposed as policies or guidelines.” I disagree because the structures of a well constructed essay, and a good policy/guideline is very different.  By a measure of literary standards, a good essay stands on its own merits, references facts and uses logic, and makes some point or tells some story.  A good guideline/policy is to the point, succinct, easily read and contains directions that are easily followed.  This is why we have to have essays and guidelines coexisting, both with possible degrees of authority.  --SmokeyJoe 14:26, 20 September 2007 (UTC)
 * I absolutely agree that good essays are not good guidelines. My perception is that this is because one is meant to be authoritative, the other isn't.  Others seem to differ on this (or perhaps simply on what the word 'authority' means). TSP 14:50, 20 September 2007 (UTC)
 * I consider it OK for an essay to be authoritative, but actually, why does it matter? Why not leave silent the issue of authoritative essays?  Mostly they won't be, and so what if a few are?  --SmokeyJoe 14:26, 20 September 2007 (UTC)
 * That is fine with me, and that is why I would prefer this page here not to claim a priori that essays are never authoritative.  &gt; R a d i a n t &lt;  14:31, 20 September 2007 (UTC)

Clarification of "How are policies started"
I think it might be valuable to clarify the "How are policies started" section. We have some good detail on Consensus that we can include here, but I think it should be a more comprehensive description of the source of Wikipedia policy. Here's a rough start:

Sources of Wikipedia policy
Wikipedia policies come from a number of sources. The most important policies are those required by the Wikimedia Foundation. Other policies are codifications of Wikipedia practices, conventions, and standards that document widespread consensus. Listed below are the major sources of Wikipedia policy, listed in order of precedence:
 * 1) The bylaws of the Wikimedia Foundation.
 * 2) Resolutions and Policies adopted by the Board of Trustees of the Wikimedia Foundation. Many Some such policies were instituted by Jimmy Wales, the founder and Chairman Emeritus.
 * 3) Office actions and other directives by Wikimedia Foundation officers, employees, and attorneys. Most office actions concern legal issues such as copyrights, privacy rights, and libel.
 * 4) Directives by MediaWiki developers relating to server load and technical issues.
 * 5) Final decisions of the Wikipedia Artibration Committee.
 * 6) Current conventions, practices, and standards that have established a wide consensus among Wikipedia editors.

Wikipedia policy and guideline pages are not in themselves the source of Wikipedia policy; they document Wikipedia policy derived from each of the above sources. By convention, the most significant and stable policies are documented in official policy pages such as this one. Policies that require more judgment and common sense by the editor are documented in guidelines.

Wikipedia polices may change as consensus changes, but policy and guideline pages must reflect the present consensus and practice, rather than attempting to lead editors toward a given result. In other words, policy and guideline pages do not create Wikipedia policy by decree; rather, they document the way Wikipedia currently works. The easiest way to change policy is to change common practice first.

One way that a Wikipedia editor can change common practice is to write a Wikipedia essay or proposal. If the essay or proposal gains wide acceptance within the Wikipedia community, it may eventually become a Wikipedia policy. Editors are encouraged to write such essays and proposals arguing for best practices and for ways of improving Wikipedia. Individual WikiProjects can also formulate their own rules and guidelines regarding certain subject matter, which might then be adopted by the Wikipedia community as a whole. Any comments? Feel free to mark up the draft. CO GDEN  20:32, 11 September 2007 (UTC)


 * Not bad at all. The last paragraph sails off the map, however, and I gather that's the one you most want.  If there's one thing the above discussions should bring to light, it's that not only does Wikipedia not have any consistent way of changing (or creating) policy, there's no agreement on what that process should be, or even how to describe it.--Father Goose 22:52, 11 September 2007 (UTC)
 * I don't have any particularly-special interest in the last paragraph. I just thought it might be a good summary of what essays and proposals are about. CO GDEN  23:05, 11 September 2007 (UTC)
 * The last two paragraphs are unnecessary, and only confuse the issues. Also, the last of the numbered methods of arriving at policy may have been the case in 2004 or 2005, but is not the case today. Today there is a process that coventions, practices, standards or other proposed mandates must go through in order to be termed "policy". That process is described at How_to_create_policy. And even that method is highly questionable in terms of its efficacy, recalling the recent WP:A fiasco. I do, though, think that putting what is presently in the "see also" section might better serve the community if it were put into the article proper rather than in the see-also section. ... Kenosis 23:44, 11 September 2007 (UTC)
 * How_to_create_policy is not policy, just an essay, for now. I think the key phrase is that consensus is a "source of" policy, which is true. Consensus, like office actions, Wikimedia resolutions, and all the others, is a source of policy. I don't think this page needs to get into any details about how policy is created yet, just where it comes from. Thus, I've added a suggested change in the heading. CO GDEN  01:27, 13 September 2007 (UTC)


 * (2) there aren't "many such", only a handful. I realize I'm nitpicking. (3) office actions aren't policy; rather, the fact that the office can act is. (4) while true, to my knowledge we have none of those. (5) is incorrect, as the ArbCom doesn't set policy. (6) is correct, and note that most policies fall in this category. I would drop the phrase about wikiprojects, and add "or guideline" to the last couple phrases. Aside from all that, the page WP:PPP may be relevant here.  &gt; R a d i a n t &lt;  14:41, 12 September 2007 (UTC)
 * I edited the draft as per your suggestions on Nos. 2, 4, 6, and about WikiProjects and "or guideline". As to #3, while perhaps office actions are no policy, they are a source of policy, which is the only thing the language claims. As to #5, the same thing, as the ArbCom is a source of policy, or at least it has the power to set policy given them by Jimbo. I don't know of any particular instance where they've exercised that power, but they have it. CO GDEN  01:27, 13 September 2007 (UTC)

There's 2 ways to go about this.... I'd prefer to emphasize the fact that we run on Consensus (with a capital C ;-) ): The wikipedia guidelines (Policies, guidelines, essays, or other) are:
 * The foundation issues, which are basically condensed consensus.
 * Other conventions agreed on between all Wikimedia projects at, on the basis of consensus between projects.
 * Current conventions, practices, and standards that have established a wide consensus among Wikipedia editors.

At limited times, these can be overridden by:
 * 1) Directives by MediaWiki developers relating to server load and technical issues.
 * 2) The bylaws of the Wikimedia Foundation.
 * Resolutions and Policies adopted by the Board of Trustees of the Wikimedia Foundation. Many Some such policies were instituted by Jimmy Wales, the founder and Chairman Emeritus.
 * 1) Office actions and other directives by Wikimedia Foundation officers, employees, and attorneys. Most office actions concern legal issues such as copyrights, privacy rights, and libel.
 * 2) Final decisions of the Wikipedia Artibration Committee.

Note especially that developers always have the final say. (for the simple reason that if a developer says "don't do that", and you do it anyway, the rule typically gets enforced by the laws of physics :-P [1]

--Kim Bruning 03:10, 13 September 2007 (UTC) [1] You: "Oh noes! Where did my image go? :-(" Developer:"Gone. I told you not to do that!". Good luck getting the foundation to recover that image, somehow :-P 
 * That formulation sounds as good to me as the proposed version. I'd support that change. I think it has the same information, with just a change of emphasis. CO GDEN  19:21, 13 September 2007 (UTC)


 * COGDEN, I respect your hard work on this; however, I am particularly sensitive to adding to the bulk of any instructional text without clear benefit to outweigh the loss of brevity/clarity. Is the value added worth the extra reading to get to the specific directions?  Since the people seeking direction here will not be involved in the upper echelon processes, why not just keep it simple and direct -- pertinent to the commoners? --Kevin Murray 03:17, 13 September 2007 (UTC)

Comment: The above “Sources of Wikipedia Policy” is an example of a fine essay. It is informative, coherent, historically interesting, and of no direct help in improving the encyclopaedia. --SmokeyJoe 06:47, 13 September 2007 (UTC)


 * I think it has very direct benefits. The purpose of a policy page is to document policy, and people need to know where the policy comes from. A lot of the major Wikipedia policies stem from Wikimedia Foundation resolutions and policies that are not cited by a single policy article. Documenting policy pages with actual citations will help solidify the pages, provide context, and make them all more informative. It will also provide backup for editors who are trying to enforce policy.


 * For example, does anybody know what Wikipedia's policy is with regard to racial discrimination against users? If you were trying to document this policy, where would you turn? If you knew that one source of Wikipedia policy was policy statements approved by the Wikimedia Foundation Board of Trustees, you'd know to look here. It would also be helpful to know that this policy cannot be circumvented by consensus or otherwise. CO GDEN  19:21, 13 September 2007 (UTC)

Revised clarification to "How are policies started?"
Here's a revised version, based on the above comments:

CO GDEN  23:03, 19 September 2007 (UTC)

Sources of Wikipedia policy
Wikipedia policy come from a number of sources. Most policies derive from consensus, of which there are four levels: The most fundamental policy principles are the Wikimedia Foundation issues, which form the core of policy in all Wikimedia Foundation projects, including Wikipedia. These basic issues, such as NPOV, are generally considered to be beyond debate. Most other policies derive from consensus, of which there are two levels:
 * 1) Wikimedia-wide policy conventions that have been established among all the Wikimedia projects. Wikimedia policy conventions take precence over policies specific only to Wikipedia.
 * 2) Current conventions, practices, and standards, established over time by consensus among Wikipedia editors. Wikimedia policy conventions take precence over policies specific only to Wikipedia.
 * 3) WikiProjects, which manage a specific Wikipedia topic or family of topics, have sometimes established their own conventions applicable only to articles and talk pages within that WikiProject.

In rare cases, the above a consensus polic y  ies may be overriden  disregarded or construed in a binding way by:
 * 1) The Wikimedia Foundation Board of Trustees, who has the ultimate say in any decision affecting Wikipedia. The Board may alter policy by adopting resolutions or policies. Some overriding authority has been maintained by Jimmy Wales, the founder and Chairman Emeritus.
 * 2) MediaWiki developers and foundation staff, through altering the MediaWiki software code or server operation.
 * 3) Office actions by Wikimedia officers, employees, and attorneys. office actions concern legal issues such as copyrights, privacy rights, and libel.
 * 4) Adjudications by the Wikipedia Artibration Committee, which are considered binding.

Wikipedia policy and guideline pages are not in themselves the source of Wikipedia policy; they document Wikipedia policy derived from each of the above sources. The purpose of a written policy or guideline is to record clearly what has evolved as communal consensus in actual practice, rather than to be prescriptive to change our behavior. By convention, the most significant and stable policies are documented in official policy pages such as this one. Policies that require more judgment and common sense by the editor are documented in guidelines.

Wikipedia polices may change as consensus changes, but policy pages must reflect the present consensus and practice, rather than attempting to lead editors toward a given result. In other words, policy pages do not create Wikipedia policy by decree; rather, they document the way Wikipedia currently works. The easiest way to change policy is to change common practice first.

Comments
Any comments on this version? From the above, I take it that the main issue is whether the added clarity outweighs the decreased brevity. Maybe this revision helps. CO GDEN  23:03, 19 September 2007 (UTC)

The fact that our best practices (policy, guidelines, essays, or whatever you want to call them) derive from consensus is fundamental.

Even NPOV is not actually considered to be truely beyond debate, nor are the foundation issues. (And in fact these *have* shifted over time due to consensus, and will shift further).


 * What's a policy convention? This is a new invented term.
 * What's a consensus policy? This is a new invented term.

The foundation, developers, and office actions provide the framework within which we work, including copyright issues, and the fact that wikipedia is an encyclopedia, etc. They do not generally interfere with what goes on within that framework, though there are occaisional situations where people try to do things that do not fit within the framework (and thus are thwarted).

Pages cannot be labeled "official policy", as we do not have an official capable of certifying pages. (the fact that pages have often been found to be mislabeled does not help either).

There are more issues (in fact many more than the previous version you proposed :-/ ) but I think that's a start for today. --Kim Bruning 20:15, 26 September 2007 (UTC)

suggestions

 * The purpose of a written policy or guideline is to record clearly what has evolved as communal consensus in actual practice, rather than to be prescriptive to change our behavior. From page on how to create guidelines, but this statement seems reasonable.
 * Added this sentence, which I think is good. CO GDEN  00:28, 26 September 2007 (UTC)
 * A riminder, per WP:PRO, that policies and guidelines are not polished-steel precision machinery, but are held together with Scotch tape and piano wire. Septentrionalis PMAnderson 05:16, 18 September 2007 (UTC)
 * I'm not sure how to implement this idea into the draft. I think some people might disagree, but I'm not sure. CO GDEN  00:28, 26 September 2007 (UTC)
 * comment: Item four on both lists - Wikiprojects cannot set policy, and ArbCom specifically doesn't set policy. Items 2 and 3 under "consensus" seem to be virtually the same thing, merge them. And the foundation issues aren't really consensus policy are they, as you realize, since you listed it in the second round, under "overridden". They are fine-tuned regularly, interpreted daily, but not ever negotiable, so that drops off as well. Leaving "consensus" as how policy is made, and three "overridden" (can we get a better handle for that?) - Foundation, developer, and Office. IMO they are in the wrong order. The "you can fine tune this, discuss how to implement or enfore this, but you cannot get rid of this" policies ("overridden") should come first. Then hey, look, its consensus. Like it's always been. KillerChihuahua?!? 00:36, 20 September 2007 (UTC)
 * Good points. I tried to address. As to ArbCom, I'd say what they do is generally apply the rules, which might mean novel interpretations of policy that don't necessarily have consensus, so I changed the wording a bit to kind of reflect that. As to the order, somebody earlier suggested focusing on consensus first, since this is how most policies come about. It's pretty rare that consensus gets overridden. CO GDEN  00:28, 26 September 2007 (UTC)
 * After reading the second draft, I must object to anything like what's being proposed here as a description of policy that would be useful on this project page. Policy, for example, may reflect the "codification" of a consensused method across the wiki, as in the case of WP:CIV. Or, in other cases, it may reflect a demand from the Board, as in WP:NFCC. In other cases, it reflects the policies set originally by the Founding Board, as in WP:NPOV, WP:VER and WP:NOR. In other cases, it may reflect a proposal of what's perceived to be a good idea for the wiki, as was the case with WP:A, which was made policy, then rejected as policy after it was made policy. TBH, I don't think this description comes close to an accurate or authoritative statement. Most of it, IMO, should be left off of this page and placed somewhere else, on a page that possibly can be linked to from this page once it is brought more into alignment with the facts of the situation. ... Kenosis 01:08, 20 September 2007 (UTC)
 * I don't think there's a consensus that any policy page reflects a codification. There are several essays that argue otherwise. I don't think this policy page can say that with any kind of authority. I think the statement as it reads now is consistent with your other concerns. WP:A is a case where a policy page did not actually describe common Wikipedia practice, so it got demoted. Hopefully, the current draft reflects your other concerns. CO GDEN  00:28, 26 September 2007 (UTC)

There have been no comments for several days, and I'm going to go ahead and implement the latest draft proposal. If there are still issues, we can always work on the problemmatic parts. If there are any show-stoppers that can't be fixed, we can always revert and bring it back here for more comments. CO GDEN  00:33, 2 October 2007 (UTC)


 * I think that Policies and guidelines is a tad bloated. Its central purpose is (?should be) documenting “What are wikipedia policies and guidelines” (present tense!).


 * The interesting and worthy issues of “How are policies started?” and “Sources of Wikipedia policy” are tangential to this. Connected, yes, but distinct and separable.  I suggest that these historical and theoretical issues should be moved to new page: Sources of wikipedia policy.  --SmokeyJoe 03:18, 2 October 2007 (UTC)


 * At approximately four paragraphs, I don't think the additions are a bloat, and they explain a very fundamental aspect of what Wikipedia policies are. I plan to copyedit it slightly, but first I want to ask COGDEN: is there anything that backs up the claim that Wales retains "some overriding authority"?  I know that was true in the past, but the current bylaws seem to grant him no special powers.  I'd downgrade his mention to "informal influence over the board", which still appears to be in force.--Father Goose 05:20, 2 October 2007 (UTC)


 * That Jimbo retains some authority is stated in Foundation issues. There might be some Board resolution on that, but I'm not sure. I guess it's possible that the Foundation issues are out of date, but I would assume someone would have correct that if it were no longer true. CO GDEN  20:16, 2 October 2007 (UTC)


 * Foundation issues is an "unofficial" page, maintained by regular Wikipedia (well, Meta) editors. That said, I find what it says about Jimbo having "some authority on certain projects by convention" correct.  Your language of "overriding authority" overstates the case (I hope); by the terms of the bylaws, it looks like the other members of the Board can oust Jimbo.  Further, his term on the board is due to expire; will he run for re-election?.--Father Goose 05:39, 3 October 2007 (UTC)


 * That was completely uncalled for and made without any broad participation from the larger community. The original, long-standing description is simply more accurate in reflecting how things have been done historically at Wikipedia. I suggest becoming more familiar with our policies and processes before making such radical and unilateral alterations to them. FeloniousMonk 05:40, 2 October 2007 (UTC)


 * It's pretty rare that any change to any page, policy or otherwise, involves "broad participation from the larger community". To insist on that would gut WP:BOLD.  COGDEN did engage in quite a bit of conversation regarding the changes first, and there was some approval, some disapproval, and a lot of commentary.  While you're free to disagree with the changes, characterizing them as "unilateral" is hyperbole.--Father Goose 07:42, 2 October 2007 (UTC)


 * FeloniousMonk, it would be helpful if you would point out the inaccuracies in the draft. Also, I think the revision incorporates the ideas of the old section. Nothing has been left out, except for the 90% figure, which is not documented and not really all that important. If you like the 90% figure, we can keep it in. CO GDEN  20:16, 2 October 2007 (UTC)


 * As to whether this revision is policy creep, I don't think so. This section is just informative. It doesn't describe policy, just describes where it comes from. Wikipedia editors do not have additional rules to follow, but the section provides helpful commentary so that users can understand what policy is, and where it comes from. CO GDEN  20:20, 2 October 2007 (UTC)
 * Cogden, I think this also raises the question of how the "Policy" template came to be attached to this page. It's actually a summary of policy and guidelines, and maybe shouldn't have the same explicit weight attached to it as policy pages do. ... Kenosis 20:31, 2 October 2007 (UTC)


 * I know there was a fair amount of discussion some time ago about the policy status of this page. I wasn't personally involved, but it makes sense to me that there ought to be a policy page about policies. Policy pages have been around for a long time, and they are generally regarded as helpful descriptions of Wikipedia convention. In fact, every policy page currently has a template header that links here. Do you think it should be demoted to an essay? CO GDEN  21:09, 2 October 2007 (UTC)
 * 5 February 2007 Radiant removed "policy summary" template ( Policy Summary )
 * 28 February 2007 Radiant added policy template ( Policy ) with the edit summary "has been in CAT:P for ages, but tag was lacking. No biggie."
 * I can't find any discussion about it. I presently imagine Radiant thought it was "no biggie" and it's gone under the radar. But it would appear that the justification for placing the Policy template was because it was in Category:Wikipedia official policy. Prior to that, the page was stated to be a summary page. ... Kenosis 23:16, 2 October 2007 (UTC)
 * I've always found it a little curious that this page is called "policy", as it contains an overview of policies and "how they work", without laying out much in the way of Official Advice. That it was in the "Policy" category probably just means someone said, "oh, this page is about Wikipedia policy, it belongs in Category:Wikipedia policy."
 * I'd support changing it back to Policy Summary.--Father Goose 05:33, 3 October 2007 (UTC)


 * It does contain sections that are policy summaries, but much of the article goes beyond that, and contains material not found on any other policy page. We might want to split the article into two: one describing the policy on policies and guidelines, and the other summarizing significant policies. But I do think it's valuable to have an official policy page that describes what policy is and what policy pages are. This page has served as kind of a "constitution" for describing how policy, guideline, proposal, and essay pages operate, and nobody has every really questioned that. I'd suggest renaming this page to Policy and moving the summaries to Key policies and guidelines. CO GDEN  02:01, 5 October 2007 (UTC)

Why on earth is this page semi-protected?
And how long has it been so? I've hunted through the talk archives and can find no explanation. (At least not that uses the strings "semi" or "protect", one of which I'd expect to see.) --Bth 21:56, 18 September 2007 (UTC)
 * My questions are: (1) Why we would want unregistered users editing our policy pages? and (2) Why we wouldn't routinely semi-protect all policy pages?  --Kevin Murray 20:17, 19 September 2007 (UTC)

check the log for the page: KillerChihuahua?!? 00:38, 20 September 2007 (UTC)
 * 09:13, 25 April 2007 Radiant! (Talk | contribs | block) protected Wikipedia:Policies and guidelines ‎ (well that didn't work [edit=autoconfirmed:move=autoconfirmed]) (Change)
 * 01:07, 15 April 2007 Centrx (Talk | contribs | block) unprotected Wikipedia:Policies and guidelines ‎ (Try unprotection)
 * I semi-protected this page because it is a common target of vandalism from IP addresses, and sees no productive edits from those IP addresses. Note that protection of policy pages is a very different matter from protection of articles.  &gt; R a d i a n t &lt;  09:31, 20 September 2007 (UTC)
 * It's semiprotected because anybody that can't obtain a username and hang around for a few days quite likely has little of value to say about WP policy and guidelines. Common sense would seem to indicate that since WP is not a blog and WP policy and guideline-related pages are not blogs, it might make sense to at least get a username first, then if one sees fit, to comment or ague or complain about some aspect(s) of policies and guidelines maybe some time afterward -- and that would be a reasonable minimum expectation, would it not?  (See WP:NOT) ... Kenosis 03:30, 2 October 2007 (UTC)