NPOV as description, not prescription

From Meta, a Wikimedia project coordination wiki

This file contains an elaborated discussion of NPOV as prescription. It was not originally raised as an issue by 24, nor is it "dead wrong" as Jimbo Wales suggests. The original file is now at NPOV as description not prescription and is available for others to extend who share this concern.


The following was written by 24. He's dead wrong.

I don't know who wrote it. I didn't. I *agree* that it is a description and that prescriptive use of those simplistic guidelines is impossible without more governance. But I did not write the original comment. Scary as this may be, someone here agrees with me...! 24

NPOV is not a "description" it is a "prescription". It is a policy that everyone here ought to follow. This is an old discussion: should the process of article writing be combative, with warring bias back and forth back and forth until one side gives up, exhausted?

Markets manage to work without that exhaustion, and they do it by having a very standard protocol of governance by which people take up their larger concerns with each other in larger, dare I say "meta", ways 24

No. Both parties in any disagreement _ought_ to try to stop fighting, and to try to achieve consensus. The miracle is: this works, it is proven, and that it helps us to quickly converge towards NPOV. --Jimbo Wales

like other markets using reputation, it just as often harms everything alive that it touches, and reflects only a shared bias called w:groupthink, an extremely well documented concept in psychology. Anonymous feedback is essential to breaking up groupthink (shared and often deliberate ignorance). The so-called convergence is towards a "spot trade" or "common bias" not to "no bias". You can minimize social conflict with this kind of prescription but you cannot make your overall encyclopedia reflect the real world outside it.24
Let me see if I understand you. If we all choose to work together co-operatively, in an effort to encompass the widest possible points of view, in a spirit of helpfulness, thoughtfulness, and compassion, then we will "harm everything alive that we touch"? I must beg to differ.
beg all you want - the philosophy you outline is called w:humanism and it is destructive to all non-human life forms. Your "work", "effort", "spirit", etc,. are guided by a w:cognitive bias and w:culture bias, but not sense of w:notation bias or limits to your wisdom. There was a terrific statement by w:Buffy Ste. Marie about this at a conference of w:indigenous peoples. She recounted a white man saying "how come you all say different things but you all agree about us?" - and the huge laugh that this got among natives with no further discussion. The white man thinks his words are real, and that they describe his actions. That is simply not true, and he does vast harm by believing such nonsense.

And if a thoughtful effort by a variety of contributors to work together for a fair presentation of all sides results in "common bias" instead of "no bias", then so be it. There is no viable alternative.

you sound like w:George Lakoff now, arguing that common bias is as good as we can get, and that there is no way to evaluate ourselves against other life forms.

Or, if there is a better alternative (for example, using voting mechanism, or rigid tools of autocratic 'governance'), then you should go start it. But most everyone here believes that the amazing power of wiki, the fact that anyone can edit THIS PAGE right now, serves quite nicely. Jimbo Wales

it does well enough, and permits certain political processes to proceed, but if we could eliminate the ad hominem argument, e.g. by everyone going anon, or simply elaborate all the visions and threats and best cases, we would have far less trouble in coordinating the governance. I hardly think a file soliciting input on "what you want" or "what you fear" out of a project is "rigid" or "autocratic", nor do I believe that voting without such an attempt at consensus would do anything but create rigid autocratic cliques. you seem like a very smart man, Jimbo, why don't you trust this "community" you've created? If you asked them all for their visions and best cases, and let the trolls (me, Larry) concentrate on threats and worst cases, you'd very shortly have a much better idea of the real consensus of the community. And I *DID* start this alternative system, via the meta and talk files, wherein I found other people with similar issues with autocracy-plus-editing.
realistically, though, the Governing Operational distinction of an IP ban or global policy of page reverts, is not something that arises from the wiki power alone. There are still conspiracies, cliques, cabals, and no small amount of crazymaking going on. If you're going to censor people, you are better off censoring them for saying that something doesn't exist because it doesn't show up in English on google, than for arguing that it does exist in French or Polish or whatever, and that no one talks in English about it, as english-speaking people think their point of view is "objective" or etc..

w:wikipedia:Neutral point of view is a description rather than a prescription. It's not a "policy" that everyone here ought to follow: I claim that every article will naturally converge to NPOV anyway. An article can ping pong back and forth when people don't agree on a topic, but eventually it will find an equilibrium point, a stage where both parties are able to live with the article, and that stage, by necessity, will be NPOV.

Not quite: it will be NPOV relative to those two opponents. There may be other opinions not yet represented. Over time however, I believe that all articles will reach true equilibria, and the only such equilibria allowed on a wiki are NPOV equilibria. 207.171.93.xxx
Two questions about that - first, how do we systematically unbias the overall equilibria to compensate for the fact that authors are self-selected and representative of people very different from the eventual readership? and, second, how do we make sure that the method we use for that makes us converge with expert opinion over time, rather than towards popular bias? 24
The authors are not here to "represent" eventual readers, so if we are not "representative" of them, then no matter. If future readers do not like the wikipedia, or detect bias, then they can edit it.
that's nonsense. It's not "no matter" since there is a clear bias here towards existing or first participants. The author most certainly must try to represent his or her best understanding of eventual readers, if only to make them comfortable enough that they understand and can comment. If obscure or obfuscatory language that is so abstract or specialized it says nothing is allowed on every page, with whatever huge vocabulary, it certainly won't cause anyone to "detect bias" nor "edit it" - but rather to simply fork the project or ignore it. 24

Don't make the conceptual mistake of thinking that "we" are writing for "them". We are them, and they are us. This is the revolution of wiki. --Jimbo Wales (please log in)

no, we are not them, and no, they are not us. At best this is a small group or clique concerned with a very narrow range of concerns, see (Ayn Rand versus Amazon Rainforest), that don't reflect the general world outside at all. That can be considered a problem, or w:no problem. If the latter, well, that's a problem. This isn't a "revolution", it's just another clique, unless you can systematically train people to "write for the enemy" not just to *convince* them, but to *solicit* their input 24
as to logging in, read what any psychology textbook says about groupthink. this project is driven by anonymous, not signed, input, anything else enables cliques and their joint ignorance. I repeat, NPOV is not an adequate policy, and it does not overcome groupthink the way that it is being applied here and now. 24

Everything below here, unless signed, was written by w:user:24.150.61.63, affectionately called "24":

Agreed, and that's a very important distinction. The single most important distinction, perhaps. None of this replaces a value system though - one still has to choose between desirable visions and undesirable threats even if they are fictionalized to the point of being totally unreal-seeming. 24
I believe that this "guild" of wiki contributors as a whole has a moral core and can best seek it out by this kind of storytelling - which is the only real control that prevents the pseudo-NPOV of pairs of combatants externalizing harms on everybody else, e.g. the w:Arab-Israeli conflict is characterized by many such equilibria - all of them seemingly bad for the next generation, bad for the current generation of kids, and bad for the olive trees, and good only for the desert that will expand as erosion lets the sand loose to cover those dead bodies. All to say:
We should distinguish between desirable and undesirable outcomes, and not let for instance "how to brew anthrax" articles have the same moral status as "how to plant an olive tree". If we allow normative functions at all, e.g. "VANDALISM IN PROGRESS" file, logins, IP bans, page locks, "rules", "prescriptions", "conventions", which we must as we are using *software* full of such normative functions, then we must have a moral core to guide how, and the better we can assess each other's cores the less pain there will be.
It seems odd that you imagine that we do not have such a moral code. We do, and our community norms are extensive and constantly growing. You may declare yourself to be somehow outside the community by denying our usage of the word, but the fact remains: you're in a community which has norms of which you are blissfully unaware. --Jimbo Wales

The argument of "natural convergence" seems very much like that of w:neoclassical economics which seems to be popular on here - and it does solve many problems, as long as there is no w:unequal-power relationship caused by one wiser-but-poorer person's slower net access or smaller vocabulary of English (which may be a second language, as it is for the vast majority of the world's people). In that case, whoever already has power is vastly more enabled by the convergence, which is the exact argument they use in the w:anti-globalization movement to argue against globalizing the rules by which the currently successful people succeeded. Presumably, we do not want the already-knowledge-rich to be telling the knowledge-poor "what they need to know", but rather, having the poor pick the brains of the rich.

I don't see how power issues enter the game: even with poor net access once a week from an internet caffee and high-school english, you can revert an article once a week, forever. This will be obvious to the regulars on the site and will show up in the article's history. AxelBoldt
yes, but an article that is reverted will simply return in a better written form - somebody with the time can always express their view more and more eloquently, and only the most absolutely relentless, merciless, and determined individuals will be able to come back to revert again and again and not be affected at all by the propaganda the author with the time has shoved in. That requires people like you and I, Axel my friend.  ;-) - 24
it is true that some people have more access to the internet than others certainly, but by the time we reach out three billionth user, this should no longer be a problem.  ;-) --Chuck Smith
it is possible to be both moral and neutral, and it's relatively easy if you know what the undisputed (not "undisputable") conventions of your tribe or guild are. I can go plant olive trees in the West Bank right now and I doubt either side would deliberately shoot at me if they knew that was what I was doing. If they knew or did shoot at me for that, people would at least know that the shooter was amoral or immoral, since no one has *SAID* that olive trees are bad, or that people who ignore the conflict to plant them are bad.

All that leaves us with the problem of soliciting description and avoiding prescription that is not consensus. There's one case of this here already:

I did what I could to try to solicit description of best cases and worst cases and get people to provide description of the status quo of this project. 207 objected reasonably that I can't set the agenda in meta, and I don't wish to, but governance is hardly one concern among many. This is a communications medium, and it needs at least to know its own value system to make ethical tradeoffs between different people's visions, which will otherwise overwhelm the project with cross-talk etc. - much as the West Bank conflict gets olive trees torn up. Do we want that?

Anyone know a better way to "get the descriptions, avoid the prescriptions"?


Wait for people's individual time tradeoffs to encourage paragraph dropping here at meta? IMHO One of the excellent traits of the medium (wiki) and our project (wikipedia) is that the subject matter is broad enough to allow spreading out and cooling off a bit. Reversion (which I need to learn how to do BTW, clearly cutting, editing and pasting is more work and out of style) should be possible in a week a two as well as 30 minutes after the previous editor completes their work.

Godel's theorem would seem to imply that even if perfect NPOV is arrived at and completely agreed to ..... newcomer's who arrive and take us up on our invitation to edit have the potential to interject new information, data, writing styles, etc. into the living work of art that is a wikipedia article. Apparently some mediation method will be necessary as the size of the project grows and the actual user base vs. number of people who created an account, found no way to delete and then abandoned the project ebbs and flows. user:mirwin


Someone other than myself wrote this, and since some have expressed a desire to discuss some of these issues without my well-known point of view becoming overly represented, I hereby reinstate the file with two comments by myself, and commit to leaving this file alone hereafter. Those who wish to read the no-holds-barred version of this, see NPOV as description, not prescription which includes a dialogue between myself and Jimbo Wales.


w:wikipedia:Neutral point of view is a description rather than a prescription. It's not a "policy" that everyone here ought to follow: I claim that every article will naturally converge to NPOV anyway. An article can ping pong back and forth when people don't agree on a topic, but eventually it will find an equilibrium point, a stage where both parties are able to live with the article, and that stage, by necessity, will be NPOV.

Not quite: it will be NPOV relative to those two opponents. There may be other opinions not yet represented. Over time however, I believe that all articles will reach true equilibria, and the only such equilibria allowed on a wiki are NPOV equilibria. 207.171.93.xxx
Two questions about that - first, how do we systematically unbias the overall equilibria to compensate for the fact that authors are self-selected and representative of people very different from the eventual readership? and, second, how do we make sure that the method we use for that makes us converge with expert opinion over time, rather than towards popular bias? 24
The authors are not here to "represent" eventual readers, so if we are not "representative" of them, then no matter. If future readers do not like the wikipedia, or detect bias, then they can edit it.
that's nonsense. It's not "no matter" since there is a clear bias here towards existing or first participants. The author most certainly must try to represent his or her best understanding of eventual readers, if only to make them comfortable enough that they understand and can comment. If obscure or obfuscatory language that is so abstract or specialized it says nothing is allowed on every page, with whatever huge vocabulary, it certainly won't cause anyone to "detect bias" nor "edit it" - but rather to simply fork the project or ignore it. 24