As you say, editors can be encouraged to write redlinks or stubs when other editors do not delete them as a matter of principle, from an opinion that there should be only full articles, to be written off line and then added.

A small number of determined individuals can hold out against consensus on policy indefinitely if they are not too conspicuous. (I do not mean an organized group, just individuals who separately have come to think this way).

The remedy is to make them conspicuous, and see which view is more generally accepted. It takes a corresponding number of equally determined individuals to accomplish this.

 B. How can editors be encouraged at writing something which is
    actually a redlink or stub? I have seen a user removing the
    redlink and maybe even complaining because there was one.

Massimo _______________________________________________ Wikipedia-l mailing list Wikipedia-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/wikipedia-l

2007/5/8, Massimo DZ8 frag_dz8@libero.it:

A first issue was about WP:CON, since my edits (referenced) seems to have met very little consensus so I've tried to pull out a few "focalized" questions.

 1. Is WP:CON valid on different informations? In other words,
    WP:CON obviously applies when the information given is the same.
    Does WP:CON apply when the information is different? Does it
    make sense?

Yes. WP:CON is about the selection of information as well as about its presentation.

 2. Is WP:CON allowed to "override" WP:V, WP:NOR, WP:RS? The
    implication of this are rather important: it means that if the
    average user is misinformed, the articles should spread this
    misinformation just because this is what's expected. It's
    definetly something that doesn't seem to make sense, especially
    when this badly collides with the references.

Given that a great number of Wikipedians endorses those values, if you think there's a consensus going against it, either you are wrong in that it does, or you are doing a bad job in explaining that it does.

 3. How is consensus evaluated in reference to audience? It
    obviously doesn't make any sense to count the number of votes,
    especially for highly specific readings (quantum physics?
    relativity? chemistry?)

No, you don't count votes. If there is only a single person or a fringe group disagreeing, you ignore them. If there is a sizable minority disagreeing then there is no consensus, and we should go on in finding one - either by searching new alternatives or by convincing others of our idea.

Moreover, I have a few additional questions regarding the issue I recently faced.

 A. The power of an hypermedia paradigm is allowing each reader to
    "follow its way". I have read a few about the need for
    introductions however doesn't this contradict the capabilities
    of hypermedia? Is it wrong to assume a reader would follow
    links? What is a "reasonable" assumption? This may imply that
    every article should include at least a part of others,
    something that doesn't seem optimal.

In general giving background is good. However, it is rarely necessary to give more than a few lines of it if the article is well linked. For example, in an article about someone whose only or main claim to fame is that he played a special role in the Battle of Gettysburg, I would expect to see mentioned that the Battle of Gettysburg is part of the American Civil War, but not an overview of the issues that led to the Secession.

 B. How can editors be encouraged at writing something which is
    actually a redlink or stub? I have seen a user removing the
    redlink and maybe even complaining because there was one.

There are several projects to give people ideas about articles to write about. Still, in the end it all comes down to "We're all volunteers, so you cannot force us, so the only way to ensure that the article you want is written is by writing it yourself."

 C. Supposing two references collide with each other, what one takes
    precedence? This actually doesn't happen for careful readers but
    seems common for casual readers without in-depth knowledge,
    probably because they don't recognize the context is different.

There can't be any general rule on this, but I do have to say that there are many cases where neither takes precedence, but the NPOV thing is to give both their say.

 D. What's the way to deal with users that (good faith obviously
    intended) end messing up everything?

Give them advise on how to avoid messing things up in the future. And revert their messing up while trying to keep the good of their edits.

 E. How to deal with the above issue when the new version meets
    greater consensus?

Surely if there is consensus that the article has improved, it has not been messed up?

 F. What to do when the above case contradicts sources? What if, to
    remove the contradiction, the sources are removed? (ykes!)

You're starting to lose me here...

 G. What about "article attacks"? An user once wrote as an edit
    summary "...this seems stolen..." and then pointed out over 2000
    pages to check!

Ask them for more specific info - why do they think it's stolen?

In general, I would like to give you the following advise:

It sounds like your page got edited quite furiously, and you want to undo that. That's not the way things work on Wikipedia. What you do (and that's what 'consensus' is really about, rather than about majorities) is to look at the edits, see what's good in them and what bad, and then try to write a version that incorporates the good things from both versions, so that both you and the person who changed the page would be satisfied with it. More specifically, without having the foggiest idea what page you're talking about, it seems that the new version is less accurate but more understandable than yours. Surely it should be possible to make a version that is more accurate than the new one, yet more understandable than yours?