Knowino talk:Guidelines

This may need tightening up.

It reads at present:

Anybody is welcome to contribute to Knowino. We have a few simple rules that we as a community expect all contributors to follow:


 * You can create an account and log in to contribute, or alternatively you can contribute anonymously. We encourage real names, but inoffensive pseudonyms are acceptable.
 * Please remember that this is an encyclopedia project, and present content accordingly—in a balanced, nuanced, and accurate way.
 * Be bold and take the initiative—this is a wiki, after all! But with that comes responsibility, so please treat your fellow editors with respect.

If you see someone egregiously breaking these rules, please contact an administrator privately, and he or she will take it from there.

Problems
The first rule: is simply a statement of fact, it's hardly a rule. It's more or less a restatement of the first sentence: "Anybody is welcome to contribute to Knowino." The second part: Can really be expressesd as "Please don't use offensive pseudonyms".
 * You can create an account and log in to contribute, or alternatively you can contribute anonymously.
 * We encourage real names, but inoffensive pseudonyms are acceptable.

The second rule: Could perhaps be better phrased as "Please present content in a balanced, nuanced, and accurate way as befits an on-line encyclopaedia". But you need to define "balanced". Do you mean NPOV like WP, pro-science like RW or something else? If the project gets going this is going to be very important and I would suggest that you want to get it clear sooner rather than later when the astrologists turn up. (Actually it would probably be a good idea to blue-link to "nuanced" and "accurate" as well.
 * Please remember that this is an encyclopedia project, and present content accordingly—in a balanced, nuanced, and accurate way.

The third rule seems good to me.--Bob M 09:42, 22 December 2010 (EST)


 * I've made a few changes; what do you think?—Thomas Larsen (talk) 22:45, 22 December 2010 (EST)
 * Cool. :-) Now I think about it you could consider reworking it as "rights and responsibilities". for example:


 * Rights:


 * You have the right to edit as an IP, under a pseudonym or under your real name. Your real name is preferred but we will respect your right to privacy if you do not use your real name.
 * As an editor (or whatever term Knowino is going to use) you have the right to edit or create article which falls within the site's remit.
 * You have the right to be treated with respect.


 * Responsibilities:


 * Not withstanding your rights to use a pseudonym you have the responsibility to not create one that is sexist, racist or otherwise offensive.
 * You have the responsibility to treat other editors with respect.
 * You have the responsibility to edit articles in a balanced, nuanced, and accurate way as befits an on-line encyclopaedia
 * And so on. The original set was really a mixture of rights and responsibilities.
 * I still feel that you need to get those "balanced, nuanced, and accurate" terms defined. But I'll add a comment on that on the about page.--Bob M 08:34, 23 December 2010 (EST)

Ownership and sandbox
Something like this could be written somewhere (mostly for newbies):

You can create an article, but you cannot own it; probably others will change it. If you want first to develop a new article yourself, without help or interference from others, then do it either on your computer or in your sandbox, and then upload it to the mainspace.

--Boris Tsirelson 01:59, 6 January 2011 (EST)


 * I think the first aspect is already covered: "If you don't want your work to be edited mercilessly and redistributed at will, please don't submit it here." However, it could be displayed more prominently, if you think it's particularly important. On your second point, perhaps it would be best to add that information to the heading that is displayed when a user is about to create an article?—Thomas Larsen (talk) 02:46, 6 January 2011 (EST)


 * Or maybe creation of a personal sandbox could be done automatically when creating a new account (I remember, this was discussed on CZ), and such a note about its usage could be a part of a standard welcome. Yes, I know that the first aspect is covered, but still, it could be better to explain both aspects early, before the user starts working; the moment when the user comes to the save button is rather late. --Boris Tsirelson 05:26, 6 January 2011 (EST)


 * I've made a note about creating a personal sandbox at MediaWiki:Welcomecreation, which is the message that is displayed to newly-registered users.


 * I read through the rules again, and I think you're right: there does need to be something about articles not being "owned" or dominated by particular users. I don't think we need an extra dot-point for this, though.—Thomas Larsen (talk) 07:11, 6 January 2011 (EST)

Actually, reading through the list again, I think the rules page needs to be somewhat rewritten. There needs to be a distinction drawn between encyclopedic content and other types of content like project- and user-pages. (This distinction is also important for Knowino:Content.) I think I'll give this a try tomorrow.—Thomas Larsen (talk) 07:18, 6 January 2011 (EST)


 * Two goals of sandbox are mentioned, but the third (not less important, I believe) is not: development of a new article without help or interference from others. --Boris Tsirelson 07:26, 6 January 2011 (EST)


 * I've changed MediaWiki:Welcomecreation to reflect that.—Thomas Larsen (talk) 07:49, 6 January 2011 (EST)


 * Nice. Good night. --Boris Tsirelson 08:01, 6 January 2011 (EST)

New draft
I've created a new draft for the rules page at User:Thomas Larsen/sandbox. It's probably far from perfect, but I'd like to know whether you folks think it's an improvement over the current page.—Thomas Larsen (talk) 02:55, 7 January 2011 (EST)


 * "Don't post illegal, irrelevant, or inappropriate material." – what is the distinction between irrelevant (to what?) and inappropriate? Also, how does it relate to Knowino:Content? Why two different formulations? --Boris Tsirelson 03:41, 7 January 2011 (EST)
 * "Be bold" – yes, but... as for me, CZ style is better than WP style; I mean: unless you fix an evident mistake, you'd better first propose your change on the talk page and observe responses. --Boris Tsirelson 03:45, 7 January 2011 (EST)
 * Maybe a too special point for these short rules, but it is worth saying somewhere something like this: if you see an error, think twice: do you understand the context? It is rather usual in Wikipedia that someone observes a formula different from his/her favorite textbook, and edits the formula, not noting that the notation used in the article differs from that of the textbook; thus a correct article becomes incorrect. Even a seemingly evident mistake is not always a mistake! Another case: someone changes a notation to a better one, but does not bother to change the rest of the article accordingly; thus again a correct article becomes incorrect. Primum non nocere - First, do no harm. True, experts approval should help; but still... --Boris Tsirelson 03:56, 7 January 2011 (EST)
 * The three notes on the current version of "Rules", are they really no more needed? --Boris Tsirelson 03:49, 7 January 2011 (EST)
 * In addition, a mistake/error very common in WP (not only in science articles but in all sorts of articles) is: People don't recognize some statements because they are phrased differently than they are used to (e.g., because of different notation), and then they add the very same statements in different wordings. Many WP articles are very long because they repeat over and over again the same things in slightly different phrasings (or notations). Unfortunately, I don't know how to avoid this problem, warning may help some, but I'm pessimistic.--Paul Wormer 04:27, 7 January 2011 (EST)

@Boris: (1) The distinction was confusing and rather redundant, so I removed "inappropriate". (2) My intention was for the rules page to contain vaguer, less specific criteria for exclusion than those in the content guide–there was a reason for this, but it will likely become irrelevant, as I will explain in a moment. (3) I agree on the need to understand context before diving into an article. It's a difficult problem to address, though. (4) The notes served to elaborate on various points, but I don't think any of them were fundamentally necessary.

@Paul: I agree. We really need to think about how we can encourage boldness and yet somehow minimise poorly-thought-out drive-by edits. Expert certification will help us to deal with the problem, but won't prevent it. Any ideas are welcome!

I'm wondering whether it would be best to create a Community Standards page, similar to RationalWiki's, to replace the rules page. That would let us elaborate on points easily, and we could merge parts of Knowino:Content into it. See User:Thomas Larsen/sandbox-2 for a possible structure. There might be good reasons not to do this that I haven't considered, however...

(Sorry if I'm being a bit terse. I'm literally working off two or three hours of sleep.) ;-) —Thomas Larsen (talk) 07:23, 7 January 2011 (EST)

[Begin copy from Knowino talk:Charter]
Feel free to comment here, by the way—I don't mind.—Thomas Larsen (talk) 08:03, 7 January 2011 (EST)

I'm not sure whether I like the term "Charter", but I'm not exactly a fan of "Community Standards" either.—Thomas Larsen (talk) 08:03, 7 January 2011 (EST)


 * "It is not our responsibility to characterise disputes" – I did not know that "characterise" can mean "resolve" (rather than "describe"). Does it? --Boris Tsirelson 01:46, 9 January 2011 (EST)


 * Sorry, I only just read this now. Yes, characterisation can imply a sort of resolution—although, more vaguely, it can mean making a value judgement about something, which is what I intended. Since the term is a bit confusing, I might use a different one.


 * Also, on an unrelated note, what's your position on original research? Personally, I think limited original research should be acceptable, if it's easily testable; but the line should probably be drawn at presenting new scientific theories, philosophical arguments, etc.—Tom Larsen (talk) 06:00, 9 January 2011 (EST)


 * Wikipedia has no choice, it must prohibit all kinds of original research, even the so-called OR by synthesis, since it (by definition) has no experts to assess it (of course, some users are in fact experts, but still cannot act as experts in WP). But we have the choice, and for now I see no danger even in new scientific theories, if we take no responsibility for unapproved articles, which looks for me a natural position, of maximal advantage over WP. Indeed, what else is our principal distinction from WP, if not experts approval? But maybe others (you, in particular) think differently, want to be more cautious. Probably RationalWiki will criticize us strongly for any unscientific article, approved or not. But I do not feel much afraid. After all, who are they? They are more anti-antiscientific than just scientific... --Boris Tsirelson 11:34, 9 January 2011 (EST)


 * I agree with you, to a point. I think the key is that our articles be objective and balanced. For example, readers of our Homeopathy article, once it is created, should be presented with an article that:


 * does not support homeopathy
 * does not outright reject homeopathy
 * makes it very clear that the consensus of the scientific community is that homeopathy is unfounded, and ensures that due weight is given to that view
 * explains how homeopathy is supposed to work.


 * That's got nothing to do with being scientific or unscientific. It's a matter of encyclopedic principles: describe the dispute, don't engage in it. (Or, at least, that's how I see it.)


 * Regarding original research, I think there's a place for easily-testable claims—like whether a particular website has closed or open registration, for example. But should new scientific theories, for instance, be presented initially in an encyclopedia project? Or would it be best if they were published in an reputable academic journal first?


 * To an extent, I think Citizendium has lost many, many experts because it allows too much original research. Certainly, it does not extend support to unapproved articles, but visitors see that support as implicit. We should probably allow some original research—I've got no problem with that—but we need to draw the line somewhere. My suggestion is that we draw the line at the point where original research ceases to be empirically testable by the average readers of an article. If a claim can be backed up to a reputable source (like an academic paper), that's another story. Or perhaps I'm just confusing the issues? What do you think?—Tom Larsen (talk) 17:37, 9 January 2011 (EST)


 * "should new scientific theories, for instance, be presented initially in an encyclopedia project?" – (a) we are compendium, not just encyclopedia; (b) do we suffer from too many contributors, or too few?
 * "Citizendium has lost many, many experts because it allows too much original research" – Really? I did not know. Could you give some examples?
 * "empirically testable by the average readers" – But then we are just another Wikipedia! Such facts are more or less acceptable also to Wikipedia, aren't they? There are many facts well-known among experts but not empirically testable by non-experts.
 * Here is a good example: Line (geometry). Not empirically testable by non-experts, and difficult, probably impossible, to find in the literature. Definitely unacceptable for WP. Do you think it will alienate experts? By the way, this section is not approved; I am careful enough with approvals.
 * Thus I want to ask you: what is, in your opinion, our main distinction from WP? (Experts approval of facts empirically testable by the average readers makes no sense: it can be approved by non-experts equally well.)
 * --Boris Tsirelson 01:06, 10 January 2011 (EST)


 * There exists already a huge online encyclopedia of facts verifiable by non-experts. What is the fuss with Knowino? My answer: (1) compendium, not just encyclopedia; (2) expert certified (when not verifiable by non-experts). --Boris Tsirelson 02:12, 10 January 2011 (EST)


 * Well, a new theory presented here in full would be much too much, indeed. However, where is the border? What about a new theorem in the framework of the existing mathematics? What about a short note about a new theory presented in full on author's website? --Boris Tsirelson 02:37, 10 January 2011 (EST)

(Edit conflict.) Regarding Citizendium and original research, I'm referring to the cases where people have left the project over the content of some of its articles on subjects like homeopathy, water memory, and so on. These articles contained (and many still do) "original" and irrational claims that had not been published previously in reputable academic journals. Read some of the comments on this THE article. I could dig up more examples if you want them.

I think I've been using unclear examples, so let me give some better illustrations. In fields like mathematics, things are, for the most part, testable. $$\scriptstyle 2 + 2 = 4.\,\!$$ And if a writer wants to include an "original" mathematical proof in an article, that's fine. Other mathematicians can test it easily enough. (I see now that there's a flaw in saying that claims must be testable by average readers. Perhaps I should have said that claims must be testable by experts in the subject.) The same applies for fields like physics, chemistry, and so on—in the hard sciences, for the most part.

But now let's take an article on some famous living person. A writer gets really creative and decides to interview that person to get details for the Knowino article on him or her. The page is then developed based on that interview, which is not published anywhere else. Nobody other than the interviewer knows when, where, or how the interview was held. In fact, no other editor or reader knows whether the interview even happened. A reviewer who is interested in approving the article is stuck with one of three options:


 * remove parts of the article that are based on the "interview", and then approve the article
 * approve the article, but explicitly state that parts of it are not included in the certification—but if the whole article is based on an "interview"...
 * don't approve the article.

There's also a danger of readers being given false or even libellous information. I mean, what is there to stop me from concocting an entirely fictitious interview with, say, Richard Stallman and writing an article on him based off it? That's the kind of original research I think we need to avoid. Even if we state quite prominently that we don't necessarily condone the content of unapproved articles, that support will likely be implied by visitors.

Also, is there any reason for an individual or group to present things like new scientific theories in an encyclopedia before publishing them in a reputable scientific journal? Scientific theories are quite different from, say, mathematical proofs: they are evidence-based rather than logic-based. To test a mathematical proof, you simply have to go through all the steps, ensuring that there are no logical flaws; to test a scientific hypothesis, you need to examine the evidence, and an academic paper is probably the best way to present that evidence for evaluation.

I think Knowino's main difference from Wikipedia is encouraging experts to review articles. We just need to make sure that articles don't contain so much original research that they can't be fully reviewed by anyone other than their authors.—Tom Larsen (talk) 02:43, 10 January 2011 (EST)

(Apologies for the length.) :-) —Tom Larsen (talk) 02:52, 10 January 2011 (EST)


 * "Perhaps I should have said that claims must be testable by experts in the subject" – for me this is a very important correction!
 * "In fields like mathematics, things are, for the most part, testable" – It depends... Just today I gave a colloquium talk about a proof (not mine but of two others, very strong mathematicians) too advanced for me to completely understand and certify. Rather, I was able to give a sketch. Not just 2+2=4.
 * "encouraging experts to review articles", "can't be fully reviewed"... – this seems to accord with my view that a submission to Knowino should be intended for expert assessment.
 * "Even if we state quite prominently that we don't necessarily condone the content of unapproved articles, that support will likely be implied by visitors" – not quite prominently then... but really you are right, at least in some sense. I was several times in such situation (in real life, not Internet): I gave a rather cold opinion about someone, but then, it appeared, the very fact that I gave something was treated as approval. A possible idea is, to have one namespace for approved (at least sighted) articles, and another namespace for others.
 * In order to be specific, examples can help. Recall two, given above, one real: Line (geometry), another hypothetical: What about a short note about a new theory presented in full on author's website? Good examples show the border much better that abstract formulations. --Boris Tsirelson 10:58, 10 January 2011 (EST)
 * See also User talk:Paul Wormer. --Boris Tsirelson 12:36, 10 January 2011 (EST)


 * I've got no objections to the material in Line (geometry). In all honesty, I don't fully understand the section; I'm not a mathematician, a physicist, or a geologist. ;-) But it doesn't appear to present outrageous claims or make statements that an expert in the relevant fields would be unable to confirm.


 * I think caution is merited when linking to a theory on one's website; unless it is directly relevant to the article, and clearly the most appropriate resource, it could be construed as self-promotion. If someone wants to link to a page on his or her website that explains an aspect of a non-controversial topic in great detail (for example, a mathematical proof), that's probably fine, provided the page is educational in nature and the link is not intended to increase the author's profit or to indoctrinate a particular ideological view.


 * For example, if a group of scientists publish a paper detailing some wonderful new theory in chemistry, and it's widely accepted by other chemists, it might be appropriate in the article about the theory to link to a page on the group's website which explains it in greater detail. But it would not be appropriate for an armchair philosopher to link to his or her paper on the meaning of life, simply because there are thousands of such papers—why choose one over the other?


 * I hope that clarifies. Feel free to bring up more potential scenarios; my own thoughts are being developed in the process.—Tom Larsen (talk) 19:51, 10 January 2011 (EST)

Rights
Contributors obviously have some basic rights, like the right to a fair "trial" and the right to leave at any time. I think there are other important rights too, though, which are perhaps more subtle: for example, the right to criticise the project. Do you folks have suggestions for other rights that should be included in the "Charter"? (I'm still open to a better name for the thing; I don't really like the term "charter".)—Tom Larsen (talk) 05:43, 13 January 2011 (EST)


 * "As a rule, all material that you post on Knowino should contribute, directly or indirectly, towards our mission." Therefore, a criticism that can improve the project is welcome here; a criticism intended to kill the project may be welcome elsewhere (on a blog, RationalWiki, etc). --Boris Tsirelson 07:49, 13 January 2011 (EST)

Research
"content should be based primarily on secondary research, which should be cited rigorously; however, limited original research is permitted" – Does "secondary research" mean "secondary sources" as in WP, that is, textbooks etc. rather than research papers? Or does it only mean, external to Knowino? In the former interpretation it leaves a hole of published research papers between secondary and original. --Boris Tsirelson 01:34, 17 January 2011 (EST)


 * Well, "previously published research" is what I meant. I realise "secondary research" is a bit unclear—indeed, as I was writing the phrase it occurred to me that there might be some confusion. I'll change it now.—Tom Larsen (talk) 02:46, 17 January 2011 (EST)

Objectivity
"readers be informed, in a clear and unbiased way, of all the significant views" – different interpretations are possible; here are two. (a) If some views are underrepresented (maybe not at all) (but who decides?) then authors are asked to fix the article within a week, or else it will disappear from the main space (either deleted or moved to the user space). (b) If some views are underrepresented (maybe not at all) then everyone is welcome to represent them.

Interpretation (a) could work better in case of ideal authors. But in reality, in many cases different views are supported by different authors, and it may be difficult to find one author for all views. Also, "but who decides?" is a problem for (a), not for (b).

--Boris Tsirelson 04:59, 17 January 2011 (EST)

For example, Proof assistant; maybe someone wants to add more about some of many systems different from Isabelle (for now presented only in section "Other systems" and the links page). I have no objections, but I am not competent to do it (and not wishing to learn them). --Boris Tsirelson 05:05, 17 January 2011 (EST)


 * So, if I understand you correctly, (a) is the active approach ("Make the article represent all significant views within a week, or we'll delete it"), and (b) is the passive approach ("This article ultimately should accommodate all significant views, but doesn't have to do so right now"). I think (b) is by far the better option. The purpose of the objectivity principle is to prevent partisanship. It's not reasonable to expect all editors, even experts, to know and understand every view on any given subject.—Tom Larsen (talk) 05:25, 17 January 2011 (EST)


 * Yes. It remains to make it clear enough to every reader of the charter. --Boris Tsirelson 06:32, 17 January 2011 (EST)

[End copy from Knowino talk:Charter]

 * I've changed the wording slightly; is it clearer now? If you think extra clarification would be worthwhile, I can add a footnote.—Tom Larsen (talk) 20:35, 17 January 2011 (EST)


 * I've only found the word "ultimately" (accommodate); a subtle hint. I'd say somewhere something like "if your view is underrepresented, do not blame us; just do it". --Boris Tsirelson 01:01, 18 January 2011 (EST)


 * I've made a few more changes, and added a note at the bottom of the relevant section. Is it clear now?—Tom Larsen (talk) 02:17, 18 January 2011 (EST)


 * Yes, it is. --Boris Tsirelson 07:17, 18 January 2011 (EST)

Some fatal errors to avoid
The following is based on my experience on CZ, — my feelings and opinions (right or wrong).

When an expert sees an article written by someone else (another expert, or non-expert), usually (in 3 cases out of 4) he/she thinks: it is bad, since I could do it better (but for now I am busy).

Life or death of a wiki project depends crucially on what happens afterwards.

Case A: the expert writes on the talk page: "this article raises many serious objections, namely ..."

Subcase A1: the author, frustrated, bangs the door.

Subcase A2: a battle between experts flares up and leads to antagonization.

In both subcases a better article is not obtained.

Case B: the expert finds the time to rewrite the article. The author is frustrated. Maybe a better article is obtained, but costs us dear.

Case C: the expert proposes to the author: work under my guidance. This idea is in fashion on CZ, but really works only on Eduzendium: students work under the guidance of their teacher and get grades. As long as we do not fork Eduzendium, it will not work here (if only rarely).

Case D: the expert does nothing, feels frustrated, and the article remains bad.

And so on...

I propose roughly the following.

1. An expert should not criticize an article unless explicitly asked to.

2. If an expert does not like an article, he/she just does not approve it.

3. If an expert can write a better article (in reality, not in potention; that is, really has the time and enthusiasm to do so), he/she creates a content fork.

4. If readers definitely prefer one version to another, the latter may be deleted.

Of course, all that applies when an article is just not good enough rather than blatantly violates our rules.

Of course, all that applies to a major rewrite rather than local improvements.

In other words: yes to competition, no to conflict. Sure, someone looses a competition and is frustrated. But this is the decision of the readers, not of the competitor.

--Boris Tsirelson 15:10, 18 January 2011 (EST)


 * I think there's room for constructive criticism, like, "I think this image should be moved down," or "I find the second paragraph unclear." That kind of criticism, if sensible and well-presented, doesn't (or shouldn't) create antagonism. However, criticism that (1) is merely based on one's own individual preferences or (2) provides no suggestions for improvement—for example, "I think we should use serial commas throughout this article," or "your article sucks"—isn't constructive, and I don't think we should encourage it.


 * The content-matter dispute resolution procedures are currently based on the principle of temporary forking, so I think that may solve (or at least sidestep) some of the issues you've brought up. I don't see any reason to delete an "unpopular" fork, unless it is clearly inappropriate for Knowino.—Tom Larsen (talk) 19:05, 18 January 2011 (EST)


 * Yes, as I wrote, "Of course, all that applies to a major rewrite rather than local improvements".
 * The problem is that rather often on CZ an expert does not write himself (being busy) and at the same time discourages others (being demanding). --Boris Tsirelson 00:34, 19 January 2011 (EST)


 * Do you think it would be a good idea to add a note specifically about the collaborative nature of the wiki? I know you mentioned it before, and it's already hinted at in the current page, but it might be worthwhile making it more prominent.—Tom Larsen (talk) 00:48, 19 January 2011 (EST)


 * I guess that these "bad" experts on CZ believe that they are collaborative. :-) Just the word "collaborative" is not specific enough.
 * Also, specific factual inaccuracies should be criticized and fixed, of course. I mean rather the general structure of the article. That is, something like that: "if you believe that the structure of the whole article could be better, do not criticize the author, - he/she is a volunteer! Rather, create your (supposedly better) version as a fork." --Boris Tsirelson 01:42, 19 January 2011 (EST)

Stubs; massive collaboration?
Something about stubs should be said: are they welcome, or not?

And a related matter could be touched (or not?): do we expect articles to be coherent narrative usually written mostly by one author? Or rather to be mosaic (like wikipedia), - each one contributing a phrase or two? --Boris Tsirelson 05:07, 19 January 2011 (EST)


 * Well, regarding stubs, "Please note that we won't delete an article merely on the grounds that it isn't perfect; deletion will usually occur only when an article is in such a poor state that it would be more practical to start from scratch than to fix the problems in the article as it stands." Perhaps that could be made more specific.


 * Perhaps we should add a note along these lines: "Unless your changes are very cosmetic, it is advisable to read through the parts of the article you intend to change to make sure you are not about to break the narrative flow."—Tom Larsen (talk) 17:39, 19 January 2011 (EST)


 * Yes, I've noted that "we don't delete..." :-) but still (at least) two positions are possible: (a) a stub is "an article in such a poor state...", to be deleted; (b) a stub is basically a request to make an article, not to be deleted. --Boris Tsirelson 00:57, 20 January 2011 (EST)


 * About narrative flow, my doubt is rather this. Imagine a new article that evidently is far from being finished (not just requires polish, I mean, but contains less than a third of the final text), that is, started in a hope that others will work further. Is it a good or bad idea? Should we welcome it, or rather move it to the user space until it will be ready to the mainspace? In other words, what is the mainspace: a city or a building site? --Boris Tsirelson 01:49, 20 January 2011 (EST)


 * I really like your analogy, so I'll continue along the same lines. I think that, ultimately, we want Knowino to be a "city". But growing cities, by definition, have building sites. Now, that does not mean those building sites are indiscriminately placed or poorly organised. There are very compelling reasons to have signs up saying, "Danger! Deep excavations ahead!" and the like. To draw that analogy across to Knowino, I think the principle is that stubs, and unfinished and incomplete articles in general, are vital and important if the project is to grow. I see no reason why relevant articles shouldn't be allowed to grow from scratch in mainspace. However, imperfections should made clear to readers. If something is not covered in an article, it's probably better to have a section for it and write "To be expanded" or "Needs improvement".


 * In my experience with wiki projects (particularly Wikipedia), I've found that there is something about imperfection that motivates people to contribute. There are few people who would join a wiki project and immediately write a full article, unless they had a very specific topic in mind beforehand. I know I, for one, wouldn't do that. But expanding a section that says "Requires expansion"—that is something that your average new user will be comparatively comfortable with.


 * To summarise: Perfection is good. Imperfection is okay, provided that it's obvious to readers and they feel compelled to address it as a consequence. So mainspace should be a city with plenty of skyscrapers, plenty of houses, and plenty of well-signposted building sites.


 * Does that answer your question? I have a tendency to get off-topic.—Tom Larsen (talk) 04:49, 20 January 2011 (EST)


 * Yes, it does. Nice. Now, back to stubs (as extreme case of unfinished articles), and specifically to User:BartFraden/Bioethics. Should I move it back to the namespace?
 * (Off topic? No, I do not see it. At least, not more than myself. :-) )
 * --Boris Tsirelson 05:34, 20 January 2011 (EST)


 * Personally, I'd say no, because the page isn't really large enough to count as a stub or an article—it's more like a single-sentence definition (and, as definitions go, it's pretty comprehensive). I can see, though, how semantics could present a difficulty here.


 * What's your opinion?—Tom Larsen (talk) 06:09, 20 January 2011 (EST)


 * For now I have no opinion (otherwise I would voice it), but my "position (b)" above, – "a stub is basically a request to make an article", – applies to single-sentence definitions equally well. Surely that page is not a valuable content; the question is, whether we welcome such requests, or not. --Boris Tsirelson 06:20, 20 January 2011 (EST)
 * And in every case (be the answer "yes" or "no") we should ensure coherent behavior of our admins by a rule formulated ("if an article is less than a stub, - merely a dictionary definition, - then..."), and known to users. --Boris Tsirelson 07:21, 20 January 2011 (EST)


 * I agree that it could be helpful to have a guideline to ensure some kind of consistency. It might not be appropriate, however, to put in this page. I plan to expand Knowino:Content soon—that would probably be a better place to explain various specific article requirements.—Tom Larsen (talk) 08:32, 21 January 2011 (EST)


 * O.K. --Boris Tsirelson 08:56, 21 January 2011 (EST)