Wikiversity talk:Blocking policy/Archive

Vandals
I think a distinction needs to be made between different types of vandals. Someone who listens to Stephen Colbert on TV and then comes here and adds some silly comment about elephants will probably stop adding useless content if we explain what this wiki is all about. Someone who is a practiced vandal and already knows exactly what a wiki is is not worth a series of warnings. --JWSchmidt 01:57, 19 August 2006 (UTC)
 * Part of the problem though, is that we have to assume good faith, and warn them, even if they are hardened, because it can often be difficult to tell the difference (I think I would know, I've made over 1500 reverts on en.wiki!) However, I think a wording change could be made to that effect. --digital_metalk 02:12, 19 August 2006 (UTC)
 * I have dealt with vandals that deserve absolutely no mercy of any kind what-so-ever. There are some that simply don't belong and have to be dealt with as the vermin that they are, as a mere warning is only going to encourage them.  Still, if somebody has been otherwise trying to make positive contributions and has a momentary lapse in judgement, I generally will be quite a bit more forgiving.  The larger issue is to deal with social vandals that are much more subtle but do things to really stir up the pot and make everybody angry.
 * BTW, as a general policy for myself, I also red-shirt every permanent block (i.e. delete their user page and user talk page). If there was something there, the individual didn't deserve any recognition at all.  Certainly I never want to see some silly WoW template for some idiot to count coup on vandalism attacks.  --Robert Horning 05:36, 19 August 2006 (UTC)


 * I agree, I don't want to see those tags here. On Wikipedia, a lot of new administrators get experience by adding these tags. It's perhaps good over there, but this is not the sort of prerequsity training that we will want to have for those who wish to work on the maintenance side of things. --HappyCamper 11:43, 20 August 2006 (UTC)


 * Why has this taken so long to implemant/reject/modify? Do we need a new system? Donek (talk) - Go raibh mile maith agaibh 20:57, 21 August 2008 (UTC)

Personal attacks
I'd rather not see anything related to "personal attacks" in this blocking policy. This notion is simply too ambiguous to properly codify. Do we really want custodians to judge whether something is a personal attack or not, and what the proper threshold is for blocking?

In my mind, the bottom line is that people who are here, are here because they take this project seriously. They don't need to be "policed" like children in a playground. If people are treated like responsible people, they will act like responsible people.

If anything, I think it might be better to refer to "professionalism" - people can simply be blocked for lack of professionalism. And this is a much better approach - we preserve the dignity of both the offender and the offendee. I'd rather be blocked for "lack of professionalism", rather than a "personal attack".

Of course, the risk is that we end up with a bunch of administrators who are very skilled at being passive-aggressive in their dealings with the community, but this is a tradeoff that we will need to evaluate, probably at a later stage in this project. What do others think? --HappyCamper 11:41, 20 August 2006 (UTC)

Votes

 * Rayc 05:24, 19 August 2006 (UTC)
 * -- sebmol ? 12:13, 20 August 2006 (UTC) Doesn't need to be a policy.
 * jes of course --Dario vet 12:21, 20 August 2006 (UTC)
 * Mirwin 12:57, 23 August 2006 (UTC)
 * Needs better wording --HappyCamper 13:22, 23 August 2006 (UTC)
 * This policy is not ready to be voted upon. --Robert Horning 22:36, 23 August 2006 (UTC)

Comments

 * I have a problem with point 4 "Any user/IP disrupting the normal functioning of Wikiversity may be blocked.", but "normal functioning of Wikiversity" soon will be defined by what wikiversity is--Rayc 05:24, 19 August 2006 (UTC)
 * This point is used on Wikipedia, and it provides a way to block disruption that isn't really vandalism--nasty messages on one's userpage, for example.--digital_metalk 13:31, 19 August 2006 (UTC)


 * Doesn't need to be a policy. We'll have to establish our own practice and precedents and write them down as they happen. Importing policies wholesale isn't appropriate nor is the timing. -- sebmol ? 12:13, 20 August 2006 (UTC)
 * This is a major action aimed at at an individual who in the perception of the Custodian(s)(concurrence of two required if I read this properly) are creating large problems. I think specific policy should be in place to facilitate review and baseline moderation.  Particularly since our percentage of custodians is quite low and will probably get lower.  Appointments all branching from a small pool of initially appointed people.
 * Wikipedia from special statistics page --> "We have 2,060,166 registered user accounts, of which 989 (or 0.05%) belong to administrators"
 * Wikiversity --> "There are 373 registered users, of which 8 (or 2.14%) are Custodians."
 * When a adversarial situation emerges it is quite common for any group to close ranks against outsiders. This group mentality should be avoided at most costs at Wikiversity.  In the future it might be useful to be able to block by school, portal or other trouble area.  At least then errors would have a more limited scope. Mirwin 13:07, 23 August 2006 (UTC)
 * There is a problem on Wikipedia where the blocking culture is as if the encyclopedia is constantly "under attack" by vandals. Just think of the sort of software package names that are available now to "protect the encyclopedia". I don't think we want to adopt that thinking here - for example, if people leave nasty message on talk pages after being blocked, why not simply protect the page, and move on? Or, if nasty messages are present on userpages, why not simply revert, and move on? Things do not need to look pristine 100% of the time, it simply is not a reasonable goal. If someone is really intent on using Wikiversity for productive purposes, they will find a way to get an unblocking. We simply need to be professional about what we do. --HappyCamper 13:17, 23 August 2006 (UTC)

Block time
*cough* - I suggest a rewording of that section entirely. :-) --HappyCamper 22:26, 23 August 2006 (UTC)

Expiry times for blocking users
I really don't think that blocking users for 1 week or even longer is necessarily excessive. I am speaking from considerable experience here when I say that. Indeed, for quite some time on Wikibooks, administrators were not only simply blocking users for just one week, it was simply an infinite block, even for anonymous users. After this policy was followed for about a year, it was moved down to a more reasonable 1 month, with more typical blocking of about 1 week being typical, and the previous infinite blocks reduced.

I know that this is not typical on Wikipedia, but at the same time this is a much smaller project than Wikipedia with considerably fewer admins. One of the motivations on Wikibooks to have slightly longer user block, (with almost everything that happens on Wikibooks with longer time frames) is both to reduce administrator overhead, and to note that the vandals really did cause some considerable damage. Posting the goatse.cx image on the front page of the project is not going to be something I would condone, and is the sign of a hardened vandal.

If you have the time, energy, and patience to really work with vandals and try to discover their motivations, perhaps a block of only 1 hour might be reasonable to "cool things off".

BTW, I still believe in infinite blocks for "registered users" that do blatant vandalism and no useful content. --Robert Horning 22:28, 23 August 2006 (UTC)


 * I think everyone is in general agreement here, it's just that the wording as it exists is open to a lot of semantic loopholes. There is the moderated idea that blocks should more or less go like this: "Block where necessary, but not necessarily blocking" - in other words, there is an element of discretion on the part of the custodian. The custodian should be able to justify the length of the block in a sentence or so. More later when I come back. --HappyCamper 22:41, 23 August 2006 (UTC)


 * I'm not even sure how a line that says "the most common block time is x" could even be justified at the moment since it would have to be determined by actual practice, of which we have close to none. I'd rather just remove the whole line. BTW, the most common block time on de-WP is indeed one hour. That takes into account that the vast majority of anonymous users use dynamic IPs so blocking them for more than one hour would cause unnecessary collateral damage. -- sebmol ? 23:06, 23 August 2006 (UTC)


 * Hence the "semantic loophole". Let's give this a rest and revisit it later. --HappyCamper 00:07, 24 August 2006 (UTC)

global block on all public Wikimedia wikis
Special:GlobalBlockList: info at Steward handbook, Erkan Yilmaz uses the Chat (try) 20:21, 21 August 2008 (UTC)
 * See also: Locked global account. --mikeu talk 14:15, 25 January 2009 (UTC)

Professionalism and disruption
Without any detailed guideline these two section look very much like blank cheques. --Hillgentleman|Talk 07:14, 1 October 2008 (UTC)


 * The term "disruption" is vague and ill-defined. If often appears to be one person's subjective opinion, ungrounded in evidence, analysis or reasoning.  In that regard, it's an open temptation to arrogant abuse of power, which is corrosive of a well-regulated academic culture.  —Albatross 11:47, 1 October 2008 (UTC)


 * Blocking itself may be an unprofessional way of dealing with subtle and complex issues. Professionals deal with issues in a professional manner.  Binding and gagging someone is hardly a professional way to treat someone who raises serious and legitimate issues that custodians may frankly not want to deal with.  —Montana Mouse 11:55, 1 October 2008 (UTC)


 * I agree that "disruption" could just mean "disagreeing with the Custodian who issued the block". In other words, any Custodians could issue a block for anyone with whom they have any disagreement.  Way too vague. StuRat 20:21, 26 January 2009 (UTC)

Recent changes
I just want to say I approve of this edit. The Jade Knight (d'viser) 10:04, 31 October 2008 (UTC)


 * I don't. "Never...unless..." is bad English, and blocking a user's talk page denies them the ability to appeal their block. StuRat 20:19, 26 January 2009 (UTC)

Blocking users for reverts
"Custodians may block users who continuously revert the edits of others without clear justification."

How is "clear justification" determined? Emesee 19:59, 31 October 2008 (UTC)


 * At the moment, I'm guessing it is up to the Custodian's discretion. Would you like to propose an alternate way of putting it, or some sort of clarification?  The Jade Knight (d'viser) 11:27, 1 November 2008 (UTC)


 * How about a 3 revert rule ? That way, Custodians don't have to determine whether the justification is clear or not, just how many times it has occurred.  Hopefully this will also prevent custodians from blocking users they don't like, for "reverts without clear justification", while ignoring such behavior from users they do like, thus favoring one side in an edit war over another. StuRat 20:16, 26 January 2009 (UTC)


 * I think custodians should have some room to decide if blocking for reverting makes sense or not. How about some conditions like [ this] that custodians could follow? --dark[[Image:Yin yang.svg|12px]]lama 00:04, 29 January 2009 (UTC)


 * It seems like there should be an "and" between the first and second point (no reasonable argument to revert AND no consensus to do so). I don't get the third point: "Participants ignore consensus".  What did you mean there ? StuRat 05:24, 29 January 2009 (UTC)


 * I mean if there is consensus to keep content, but someone keeps reverting anyways. --dark[[Image:Yin yang.svg|12px]]lama 09:37, 29 January 2009 (UTC)


 * In that case it seems rather redundant, since, if there's a consensus to keep, then obviously there's no consensus to revert. StuRat 14:01, 29 January 2009 (UTC)

Warnings
Shouldn't we have a section on warnings? I think strong, clear warnings should be given in cases of lower-grade long-term abuses, and would be highly recommended for all but blatant, intentional abuse of the site. Sχeptomaniacχαιρετε 21:07, 21 January 2009 (UTC)


 * Warnings themselves are subject to being used in a threatening or abusive manner. A warning typically takes the form of an anankastic conditional which (in many cases) amounts to a form of bullying.  —Barry Kort 21:41, 21 January 2009 (UTC)


 * Sorry, Barry, but it has become abundantly clear your ethical philosophy is based on very in-depth readings of only a few concepts (not all of which you seem to understand), an utterly unrealistic idealism lacking any basis in actual human behavior, and your own egocentrism. As a result, it's become less and less necessary to reply to you, since you are usually reposting something I've already read ad nauseum.  I'll indulge you this time, though.


 * My suggestion is about how to deal with real people (albeit in a virtual space). As a result, warnings, as well as "anankastic conditionals", are among the tools that will be needed to address problems. There's a fine line between a warning and a threat, but it does exist.  The former is a sincere attempt to deter problem behavior, while the latter demonstrates an intent to do harm (bullying).  Frequently, you can tell the difference by the focus of the statement.  If it is on the behavior (or sometimes the person) that must change, that's a warning ("Please stop this behavior, or it will be necessary to block you"), but if it's on the punishment (and usually phrased to focus on the speaker), it's usually a threat ("I'm going to block you if you don't listen to me"). Sχeptomaniacχαιρετε 22:53, 21 January 2009 (UTC)


 * The key distinction that divides anankastic conditionals isn't the divide between a warning and a threat, but the divide between powerlessness and power to cause the consequent to occur. If I (as a scientist) issue a warning, it's in the form of a prediction of a likely harmful event that will occur from natural causes, over which I have no control.  A threat (including the kind of warning contemplated in the context of this thread) is a consequential act that speaker has vested political power to cause.  Intentional harm (punishment) is well-known in the scientific literature to be a contra-indicated practice — ineffective at best and counter-productive at worst.  Woe to the anankastic control freak who fails to appreciate this crucial scientific finding.  Barry Kort 23:10, 21 January 2009 (UTC)


 * "Intentional harm (punishment) is well-known in the scientific literature to be a contra-indicated practice — ineffective at best and counter-productive at worst." Not true.  You are once again showing that same idealism and lack of breadth to your knowledge.  Punishment (and reward) is a basic concept in behavioral psychology.  Some applications of punishment are unlikely to be effective, but it has certainly long been known that punishment is a necessary tool in any social group.


 * Your definition of "warning" is also not in keeping with any common definition. It makes no difference whether the speaker has the ability to cause potential harm or not.  A warning seeks to turn someone away from a course of action, while a threat seeks to cow a person by demonstrating an intent to harm. Sχeptomaniacχαιρετε 00:39, 22 January 2009 (UTC)


 * Are you familiar with the research of James Gilligan, Suzanne Retzinger, or Rene Girard? Their path-breaking research overturns the naive myths that you cling to regarding the utility of punishment.  Intentional harm (notably encompassing state-sponsored violence) is increasingly seen as an unwise practice.  —Barry Kort 01:04, 22 January 2009 (UTC)


 * Your Google searches linked to indicate you are misrepresenting what those three have to say. They disagree with certain kinds of punishments (tending to write on criminal justice and shame-based punishments), but I see nothing to indicate they reject all punishment. Sχeptomaniacχαιρετε 01:17, 22 January 2009 (UTC)


 * Please continue your studies before you draw any conclusions. Have you read this article by James Gilligan? —Barry Kort 10:59, 22 January 2009 (UTC)


 * It only reinforces my first impression. You are attempting to take the work well out of its context.  Gilligan is talking about physical and shame-based punishment, defined as "to deliberately cause them pain--above and beyond the degree that is unavoidable in the act of restraining them," within the context of criminal justice and in the form of physical harm and imprisonment.  Since the intent of blocking here is to remove disruption, it wouldn't even fall close to his definition of punishment. Sχeptomaniacχαιρετε 18:45, 22 January 2009 (UTC)


 * Historically, Wikiversity has made a distinction between two types of situations. First, there are some problems such as obvious vandalism where custodians clean up the mess as quickly as possible and move on. When cleaning up obvious vandalism, no warnings or talk page messages are needed. The approach that has been adopted is No shrines for vandals. For other situations, I think it makes sense for custodians to be sure that they are not blocking under conditions that will lead to controversy. If there is doubt, then there should be discussion first and the community should decide by consensus if a block is needed. Another issue that should be made explicit is that a custodian who makes a bad block should not have the power to refuse an unblock request...someone else should evaluate the situation. If two custodians do not agree then it should become a matter for community discussion, not a "wheel war". --JWSchmidt 00:00, 22 January 2009 (UTC)


 * I'm in agreement regarding vandalism, which is clear-cut abuse. However, I think there is a further category of cases that may be clear-cut, but not urgent, and those that are more ambiguous.  The less urgent ones could benefit from a warning, and ambiguous situations can be taken to the community.  Warnings also have the benefit of giving those warned a chance to bring the issue before the community if they disagree with the assessment. Sχeptomaniacχαιρετε


 * Rather than a warning (i.e. a threat of administrative sanction or punishment), an alert regarding community policy is more appropriate. What you have now is not a coherent policy, but a hodge-podge of mutually inconsistent rules, each one carrying the option of a discretionary sanction by an admin who favors that rule over a conflicting one.  These inconsistent rules transform Wikipedia into a chess game:  "If you move your piece into that square, then I will respond by moving one of my pieces either to block your gambit or to entirely remove your piece off the game board.  Neener."  —Barry Kort 10:59, 22 January 2009 (UTC)

I think Sχeptomaniac's idea here is a step in the right direction: if the warned user thinks that whatever they were doing doesn't warrant a warning, they should bring it up on WV:CR for comment. If there's a reasonable consensus that what they were doing is not problematic, they should feel free to ignore the warning. They should not, however, just bring it up on CR and continue doing the same thing they were warned about. --SB_Johnny talk 14:43, 22 January 2009 (UTC)

Possible addition
OK, there seems to be some support, so perhaps an addition would look something like the following: Except when dealing with situations requiring immediate action to prevent further damage to Wikiversity, a Custodian should give a clear warning regarding the problem behavior that needs to stop before placing a block. Escalating blocks may then be placed on the editor as the behavior continues. If the editor disagrees with the Custodian's assessment, they may request a community review of the behavior in question (however, they must stop the behavior while the review is under way in order to avoid the possibility of a block). Any thoughts or edits? Sχeptomaniacχαιρετε 19:10, 22 January 2009 (UTC)


 * Looks fine to me. --SB_Johnny talk 19:40, 22 January 2009 (UTC)
 * Sounds good. --mikeu talk 20:00, 22 January 2009 (UTC)


 * ✅ text is added --mikeu talk 15:02, 25 January 2009 (UTC)


 * Let's test-drive the process of Scholarly Community Review by convening a Moot Court Ethical Review of the previous blocks. —Barry Kort 19:28, 22 January 2009 (UTC)

Comments
I strongly disagree with the phrase "Blocks may be placed on accounts or IP addresses which have consistently demonstrated a lack of professionalism and respect for the editing environment on Wikiversity." Wikiversity is for everybody also children and also people with mental disorders. If someone "doesnt have the a professionalism" in editing the wikiversity it is not a sin. I would advice to remove "lack of professonalism".--Juan de Vojníkov 19:39, 25 January 2009 (UTC)


 * Good point. I support removing the reference. I think "civility" makes more sense. --SB_Johnny talk 20:01, 25 January 2009 (UTC)


 * So, just substitute civility for professionalism, and leave the rest? --mikeu talk 20:20, 25 January 2009 (UTC)


 * What about renaming the section "Disrespect" and changing the first sentence to "Custodians may block users who have consistently demonstrated a lack of respect for the editing environment on Wikiversity or for other participants."? --dark[[Image:Yin yang.svg|12px]]lama 20:38, 25 January 2009 (UTC)


 * Agreed, "professionalism" is too high of a standard to expect, while "civility" and "respect" are not. StuRat 20:26, 26 January 2009 (UTC)

I agree, with you too.--Juan de Vojníkov 17:41, 27 January 2009 (UTC)


 * I kind of like "collegiality" since it has both an egalitarian sense and an academic sense. It's also not the name of a policy, which might be better as well. --SB_Johnny talk 17:47, 27 January 2009 (UTC)


 * What exactly are you suggesting? I don't understand your comment. The only part I can reflect on is the last sentence. Any word has the potential to be used in the name of a policy or become the subject of a policy proposal, so I don't see the point of avoiding words due to their use in a policy proposal. --dark[[Image:Yin yang.svg|12px]]lama 11:03, 29 January 2009 (UTC)

I defy anyone to find so much as an ounce of collegiality in yesterday's sham (and atrociously scripted) Bill of Attainder.

Thomas Jefferson was demonstrating collegial insight when he led the civilized world in abandoning Bill of Attainder, having recognized that it's a corrosive and corrupt tool of government that inevitably sinks any regime that comes to rely on it. Shortly after the Founders outlawed Bill of Attainder in Article I of the US Constitution, the British followed suit, abandoning both Parliamentary Bill of Attainder and Monarchial Bill of Attainder.

It's disturbing to witness you and other misguided officials reintroducing this antiquated relic from the rubbish heap of political history into the current practices of Wikiversity. Is this the kind of unwise practice you wish to teach to impressionable youth of the 21st Century? Can you imagine what would happen to anyone who tried to adopt and employ that hoary and unsustainable practice in the real world of an authentic 21st Century learning organization?

Moulton 13:49, 29 January 2009 (UTC)


 * Note. Misguided blocking is the subject of community review. --JWSchmidt 00:49, 21 July 2010 (UTC)

Responsibilities of the blocker
I'd like to see a section added on the responsibilities of the blocker. These are what I would include as info the blocker must provide when adding a block:

1) The nature of the block (is their user page protected, and e-mails locked, too ?).

2) The duration of the block. ("Indefinite" is not a duration, it's a way to avoid giving the duration.)

3) The reason for the block, citing specific policies which have been violated.

4) Evidence for the violation of policies, such as diffs.

5) Disclosure of any apparent conflict of interest. If the blocker has been having an edit war with the blockee, for example, just prior to the block, this needs to be disclosed.  (We may also want to make a rule that nobody with a conflict of interest can add a block).

I'd also like a statement that the block can be requested to be removed by the blockee, and then will be removed by a Custodian, if any of these standards aren't meant. I want to avoid the situation where a Custodian puts a block on a user, for reasons that aren't clear, then the block is left on for an extended period while the blocked person tries to figure out why he was blocked. In some cases, the Custodian may have just blocked the wrong user name, but lack of info at block time makes this difficult to determine. StuRat 20:01, 26 January 2009 (UTC)


 * The nature of the block and the duration of the block is already included in the log for the block automatically, so I don't see a need for custodians to repeat that information. If custodians were to repeat that information anyways, what do you propose custodians put instead when they really do mean indefinitely? "Forever", "Until hell freezes over", or what? Are you also suggesting that custodians must never block user accounts as well as IP addresses indefinitely? What do you propose custodians do if there are many instances in which policies were violated? --dark[[Image:Yin yang.svg|12px]]lama 23:26, 26 January 2009 (UTC)


 * The problem is that "indefinitely" can either mean that it's intended as a permanent block or that the length of the block has not been decided. I don't want it left vague like that, I want the blocker to state explicitly how long they intend the block to last.  If they mean a permanent block, then state that.  If they mean until some condition is met, then they should state that.  Vague conditions, like "until they learn to behave", should also be avoided. StuRat 04:48, 27 January 2009 (UTC)


 * About your suggesting about adding suggestion to add a rule that nobody with a conflict of interest can add a block. What if there is nobody left without a conflict of interest? I think "conflict of interest" is vague and can be gamed just as easily as you seem to feel that "disruption" can be. I've seen cases on other wikis where someone accused each person who attempted to block them of having a conflict of interest. I think the point is to find the right balance between accountability and custodians still being able to do the job that the community trusted them enough with the tools to do. I think that balance already exists in this proposal in the form of anyone being able to have the community review any block that a custodian makes. I think that is better than just allowing the blockee to make the request. I think anyone whose ever been blocked feels the block is unjustified. --dark[[Image:Yin yang.svg|12px]]lama 00:03, 27 January 2009 (UTC)


 * So, are you arguing that in the case I gave, where the Custodians was edit warring with a user, it's OK for that Custodian to block the user, so they win the edit war ? I think we have enough Custodians so we will be able to find one that doesn't have a conflict of interest with the user to be blocked.  Of course, the blocked user can claim that everyone has a conflict of interest, but nobody is likely to believe this. StuRat 04:55, 27 January 2009 (UTC)


 * I'm with Stu on this: if a Custodian is in an edit dispute, they should not use sysop tools (block or protect) to "win" the dispute. --SB_Johnny talk 11:18, 27 January 2009 (UTC)


 * No, I'm not arguing that. I think being more specific about what it is you wish to prevent would be better than using "conflict of interest" as a guide to what custodians shouldn't do. Maybe something like "Custodians must never block or threaten to block users in an attempt to influence or bias disputes, decisions, or consensus"? --dark[[Image:Yin yang.svg|12px]]lama 12:17, 27 January 2009 (UTC)


 * That's good, but I'd also like the more general "Custodians must avoids blocking, or warning of blocks, where the appearance of a conflict of interest exists". StuRat 14:44, 27 January 2009 (UTC)


 * Well I don't like the COI bit, as I've said before, I think its too vague and can be easily abused by anyone wishing to try to avoid a block. How about "Custodians must avoid the appearance of favoring those they agree with over those they disagree with when using custodian tools"? If you don't like that or think its not enough, can you think of any other ways to describe what you want without using "conflict of interest"? --dark[[Image:Yin yang.svg|12px]]lama 16:20, 27 January 2009 (UTC)


 * I guess that's OK. StuRat 03:51, 28 January 2009 (UTC)

What do you think of [ these changes]? --darklama 19:49, 28 January 2009 (UTC)


 * I had to read this sentence several times to figure out what it means: "Blocks may be escalated for behaviors that continue without further warnings." I was thinking "what are behaviors that continue without further warnings ?"  That sentence needs a rewrite: "Blocks may be escalated, without further warnings, if the same behavior continues."


 * The part you added was "Custodians must avoid the appearance of favoring participants that they agree with over participants they disagree with when using custodian tools. Custodians must never block or threaten to impose a block as a means of ending or influencing discussions. Custodians should seek a second opinion before blocking if doubts about blocking exist." That's good, but there's a lot more we discussed in this section that you haven't yet put in, such as those 4 points I started with. StuRat 05:17, 29 January 2009 (UTC)

Violating guidelines ?
It doesn't seem appropriate to me to block someone for this, as they aren't policies, after all. The word "guidelines" sounds more like a suggestion than a requirement, to me. StuRat 20:05, 26 January 2009 (UTC)

Infinite blocking on IP addresses
I notice one IP has been blocked indefinitely, by an admin here. Should IPs be blocked indefinitely at all - I'm not sure they should though. Even open proxies can become closed or so. Yes, CheckUser evidence maybe a reason to block an IP for a while, but normally, no longer than a year for non-open proxies, where CheckUser evidence is found, or a spammer/troll evading their block, but 3 years for an open proxy IP. Is that fair? That was my old wiki's blocking policy, and it seemed to work well.

Feedback is appreciated. AC --Sunstar NW XP 11:16, 20 February 2009 (UTC)


 * Recently, I've been reviewing some of the older blocks and as part of that I have reduced many of the longer ones to just one year. I tend to agree that there is not really much point in very long blocks of ips.   --mikeu talk 14:17, 20 February 2009 (UTC)


 * I agree, both because those IPs may no longer be used by the same people and because any vandal is likely to lose interest in vandalizing Wikiversity after a year. StuRat 16:02, 20 February 2009 (UTC)


 * I just went through some of the older indef blocks of open proxies. I shortened the block to 1 year and enabled allow account creation.  Then I realized that there are at least a couple of thousand blocks...  It seems that it was standard policy as part of Open Proxies back in 2006-2007 to use indef.  IMO, the only practical way to deal with these older blocks is to use a bot, though I am not sure about running a bot with a custodian flag.  The number of recent indef blocks is much smaller, and it is more practical to fix by hand.  --mikeu talk 17:35, 20 February 2009 (UTC)


 * If you can get a bot to work, great. If not, maybe you could just do maybe a dozen a day and perhaps get a few others to do the same ? StuRat 18:33, 20 February 2009 (UTC)

Offensive username blocking policies
I was just on the welcoming committee new user log and saw some interesting names, my current *favorites* being Pissonyou and Iamgay12345. What are the policies when you see this sort of blatant, offensive misuse of Wikiversity with just the user's name? Should they be blocked immediately or is good faith assumed even in these cases? I really don't see any way a name can be redeemed. Any comments here? Trinity507 03:39, 2 September 2009 (UTC)
 * I don't think "Iamgay" is that terribly offensive of a username. Pissonyou, on the other hand... The Jade Knight (d'viser) 05:07, 2 September 2009 (UTC)

Here are some examples:

If you go through the logs you might see something like this:

10:41, 5 October 2008 SB Johnny (Talk | contribs | block) (log action removed) ‎ (Unacceptable username)

In addition to blocking, a request was made to have stewards delete the account. This might be done if the username contains libel against a person, for example. The requests are made in irc #wikimedia-stewards to prevent drawing attention to an abusive username on wiki. --mikeu talk 13:11, 2 September 2009 (UTC)
 * Censorship has to be evoked for that and requires quorum in my judgement. I'm not personally offended by those user names, but those users need to realize that establishing their identity online is a once-in-a-lifetime chance: you are what you type. We can mimic w:Wikipedia:Offensive username proposal with Offensive username proposal or do something original. Let's look at Special:Contributions/Pissonyou and Special:Contributions/Iamgay12345. Hmmm. Not much there yet. Maybe User:Jtneill and others have some psychological insight about why a person would do something like that to themselves. I would also like to hear some disclosures from User:Altera vista from the Universitas/Communitas perspective. This would go to the top of Guidelines since choosing your username is your first action when arriving here. Perhaps an unwelcome template should be crafted that points to the proposed Blocking policy and this. This is not easy. We need to look at other cases on other languages and projects, also. My idea for a responsible action is to notify them first and give them a week, then block. --CQ 13:17, 2 September 2009 (UTC)
 * I support notifying first. The Jade Knight (d'viser) 18:48, 2 September 2009 (UTC)
 * Notification sounds like a good idea to me, too. I think the issue is less that it is personally offensive (though it might be to some people, it's not really to me) and more the "you are what you type" thing. What kind of persona are you creating by using those usernames for your Wikiproject identity? Doesn't really make sense to me as I don't think you'd get much respect from the community. Censorship is an issue though, especially with something like Iamgay. Some people might have a problem with that (I don't) and others would call it vandalism (I know a LOT of people who would be highly offended to see that on a page's edit history). So this is definitely a tough one. Trinity507 20:08, 2 September 2009 (UTC)

Talk page
I'd like to propose an additional section on Talk page based blocking standards that would go something roughly like:


 * When an established user (over 10 edits and two weeks of being here) is blocked, user's talk page access is not blocked except in circumstances dealing with harassment, outing, and other extreme situations, and then only on a temporary basis. Talk page access can be used to request unblocks, discuss the background and nature of the block, or to create content for other parts of the website. Language and incivility on the talk page following a block are not grounds for extending a block, and only in the extreme situations dealing with privacy issues or threats would allow for the disabling of the user talk page access, and only on a temporary basis.

I would also like to add these two clauses somewhere in the policy:


 * Standard blocks of an established user (define above) are for a 24 hour period and not incremental. Any block requiring more time must be agreed upon by the community. Blocks for IP ranges can be of various times based on the nature of the block, including repeated vandalism, harassment, and other factors.


 * Users are allowed use of only one account unless publicly linking the two, and any secondary accounts of banned users are also blocked. Only the primary user talk page is allowed access to in the case of multiple accounts.

Feel free to chime in with suggestions, tweaks, etc. Ottava Rima (talk) 00:00, 7 July 2010 (UTC)


 * I'd object to mentioning "outing" unless a policy such as Privacy policy is made official and that page includes a definition of "outing". Similarly, "banned users" should not be mentioned unless Bans is made official. --JWSchmidt 00:38, 7 July 2010 (UTC)
 * Banning is part of a blocking policy and does not need its own policy page. The privacy policy is something determined by WMF so it trickles down to the rest without needing reapproval. Ottava Rima (talk) 02:13, 7 July 2010 (UTC)


 * "Outing" is based, if I'm correct, on overall WMF policy; that user RL information is not revealed is required by that policy, it's a promise to users. We may change the policy for our local wiki, but only with notice to users and an "opt-out" possibility. I.e., we could create a new class of user, call it, say, Open User, who uses or openly declares real identity, independent contact information, etc., and we could assign special rights to such users. We could require that custodians or other privileged functionaries be such users, for example. But we have not done this, so a prohibition on outing must remain. "Outing" means revealing real-life information about an editor, without necessity (an example of necessity might be serious RL harassment of another editor). Editors routinely use, with an expectation that it is accepted, real names for editors, I might call JWSchmidt "John," for example, and if I do so, it is not hostile or harassing, because this name has been openly acknowledged, and it is not "outing." However, if JWS protested, and I knew and was aware of the protests, and I continued anyway, I'd be "outing." Generally, an editor should not be sanctioned for what may be inadvertent "outing," but rather should be warned and only sanction if the behavior continues.


 * As to multiple accounts, there is some tendency on wikis to develop rigid rules over time, losing the original freedom. In theory, Wikipedia attempted to avoid this by enshrining Rule Number One, "if a rule prevents you from improving the project, ignore it," called w:WP:IAR, Ignore All Rules. If IAR is policy, as it is on Wikipedia, nobody should be sanctioned punitively for ignoring a rule, but only protectively or unless bad faith has been clearly demonstrated. Rules, as well, should not be made that are over-rigid, particularly where possible exceptions are known to exist. There is generally a distinction made on Wikipedia between "policies" and "guidelines," where violation of a policy is considered to be more clearly unacceptable than a violation of a guideline, but, in fact, because of IAR, all policies are guidelines, they are merely stronger than other guidelines, and one violating a policy should be prepared to defend the violation as necessary for the welfare of the project.


 * Being blocked, in itself, should never be considered a proof of actual bad faith or actual misbehavior. Indeed, because of IAR, on Wikipedia, I used to say, "If you haven't been blocked yet, you are not trying hard enough to improve the project." When I was first blocked on Wikipedia, it was for boldly challenging an administrative action, and I wrote, to the blocking admin, "You don't know how happy you have made me." She, by the way, conducted herself with absolute propriety, even though, in my opinion, she erred. (I was considered to be harassing the administrator whom I had challenged, but that admin later wrote that it was all a "misunderstanding" and he became a strong supporter of my work.) Administrators (custodians here) are charged with the responsibility to use their tools according to their understanding of policy and IAR. So we should understand that by assigning the custodial tools to a user, we are trusting them to use their tools in this way. If we find that the custodian does not use the tools wisely, but causes damage and disruption even if well-motivated, then we should suspend or remove tool access.


 * Suspension would mean that a request would be made that the custodian stop using the tools, either generally or in a specific situation. We should develop policy about this, to make it clear when tool use should stop, and violation of this policy by a custodian should result in rapid desysop.


 * Back to multiple accounts: Wikipedia recognizes, and we should recognize as well, that there are legitimate reasons for an editor to have more than one account. Creating a "sock" is not an offense in itself. However, socks can be used abusively, and the following is a partial list of abusive usage:
 * To create the impression of wider support for a position than is actually the case. A declared sock, even, can be used for this; for this reason, editors should avoid using more than one account in a particular discussion, unless the account is so clearly connected with the primary account that others cannot mistakenly conclude that this is more than one editor. Example of allowed usage: I have a declared sock on Wikipedia, Abd-sock, or something like that. If I decide to use this account when using a public computer, for security reasons, it would be legitimate for me to participate in a discussion using both my main account and the sock.
 * "Bad hand account." To harass other users and engage in policy violations while protecting the main account. The violations are considered a violation by the user himself or herself, and the user may be blocked. However, if the user has successfully partitioned behavior, sometimes this kind of abuse receives only a response of a block on the sock account and a warning on the main account. The point is that if the user is able to partition behavior, there is some reasonable hope that the main account will continue as a positive contributor. However, such a sock identification should always be tagged on to the main account, through a Talk page warning, so that the record of prior misbehavior is there for review if there is repetition.
 * "Good hand account." This is preparation for the main account being blocked. However, it could be very difficult to distinguish this kind of sock from a simple legitimate sock.
 * Block evasion. If such a sock is used only for positive contributions, however, it is within administrative discretion to block or not block. General practice is to allow immediate reversion of edits by block-evading socks (or IP identified as being a blocked editor), and immediate blocks, but it is not the contributions which are excluded, rather only the ability of a blocked editor to make contributions having a right to stand and require review before removal. (We only block editors when we come to fear a continued burden of review, outweighing benefit, otherwise we would simply revert bad contributions and leave the good ones, why block?) Any editor may revert such contributions back in by assuming responsibility for their non-disruptive character or for asserting their value.
 * "Straw puppet," used to create an impression of misbehavior by another editor or to attract negative administrative attention. This is a form of "trolling."


 * And then there are legitimate usages for socks:
 * To allow usage of public computers where there is risk of keylogging, as an example, to compromise the main account password. Generally, these socks should be openly identified with a request that the account be immediately blocked if possible compromise, as shown by misbehavior, appears.
 * To operate a bot, so that the bot can be immediately blocked if there is a problem. (Openly identified.)
 * To avoid real-world consequences for whistle-blowing, as an example, particularly where the primary account is identified as a real-life identity. (Obviously not openly identified, and open identification of such a sock should be avoided as a form of "outing.")
 * To partition behavior or topics so that arguments will be seen free of prejudice coming from participation in another area. This would only be abusive if done to avoid sanctions.
 * For fun. As long as it is not disruptive, this is legitimate. However, abuse or the appearance of abuse should be avoided.
 * For educational purpose, to demonstrate a type of behavior. This should almost certainly be openly identified, or privately identified to a privileged user, such as a custodian, who responsibly supervises it, or it could be considered disruptive.
 * And a new one, recently identified: to allow a globally locked editor to negotiate for unblock with the local community, and to edit with the permission of the local community. This kind of sock must be openly declared, ordinarily. It should probably have a custodial supervisor, to closely restrict its activities to avoid disruption, until and unless the local community supports unblock and free activity.


 * Grey area on multiple accounts
 * Recently, an editor created a sock account with behavior that resembled that of a blocked editor, in some ways. Suspecting that this was a block-evading sock, a custodian requested checkuser. That's an administrative burden, so this sock created a burden, which should be avoided. Further, the sock was "welcomed" by the primary account, and there was a brief back-and-forth, thus creating, on the surface, an impression of disconnection, amplifying the impression that the sock was that of the banned editor. When the admin reported the results, the user then attacked the administrator for a privacy violation, while asserting no underlying justification for the additional account other than what might fit above as "fun." Warning the primary account about the sock should probably be sufficient for this situation. The warning might be private, but since the connection was publicly identified, and then referred to all over the wiki by the primary account, there is no alternative but simply making sure that the accounts are openly linked. It is possible that this account has been used to possibly create an impression of two users, for those unaware of the connection. The existence of grey areas is a reason why custodians must have discretion. --Abd 15:35, 28 July 2010 (UTC)


 * "Outing" is based, if I'm correct, on overall WMF policy <-- No, this is wrong. The WMF privacy policy concerns release of information from WMF servers by WMF functionaries. "so a prohibition on outing must remain" <-- There has never been such a prohibition at Wikiversity. Such a prohibition was proposed as policy two years ago. However, if JWS protested, and I knew and was aware of the protests, and I continued anyway, I'd be "outing." <-- Using someone's name is not outing, it is normal behavior in a learning community. --JWSchmidt 18:00, 29 July 2010 (UTC)


 * "So we should understand that by assigning the custodial tools to a user, we are trusting them to use their tools in this way" <-- Wikiversity Custodians clean up spills (obvious vandalism). All other use of custodial block and delete tools is by community consensus, with a particular sysop simply doing the work as directed by the community. The Wikiversity community needs strong protections against the documented poor judgment of a few rogue sysops who have disrupted Wikiversity. If such a sock is used only for positive contributions, however, it is within administrative discretion to block or not block." <-- Wikiversity welcomes all good faith contributions. It would be destructive to Wikiversity and would alienate honest Wikiversity community members if there was a policy calling for mindlessly deleting good faith contributions. "Suspecting that this was a block-evading sock, a custodian requested checkuser. That's an administrative burden" <-- The account did nothing wrong and should not have prompted any Custodial action. --JWSchmidt 18:56, 29 July 2010 (UTC)

blocks by non-Custodians
I think there should be a section about blocks imposed by non-Custodians. Something like:

"Stewards can block vandals at any time. Any other blocks imposed by non-Custodians must first be discussed with the Wikiversity community."

--JWSchmidt 00:55, 7 July 2010 (UTC)
 * I believe stewards would need to op themselves to block. Founder, however, had that ability without opping (but that is not true anymore I believe). Perhaps you mean above "non-community approved Custodians"? Now, those like mike.lifeguard are community approved sysops and stewards. Ottava Rima (talk) 02:12, 7 July 2010 (UTC)
 * No they don't have to. Perhaps look at the block list and you'll see the most regular stewards (Jyothis' and Pathoschild's) blocks. Diego Grez 02:20, 7 July 2010 (UTC)
 * Those would be global actions. I don't recall any local blocks from them without having to add the sysop flag. (see example: # (show/hide) 00:39, 7 July 2010 Jyothis (Talk | contribs | block) blocked Spiteclear (Talk | contribs) with an expiry time of infinite (account creation disabled, e-mail blocked, cannot edit own talk page) ‎ (crosswiki abuse) ) Ottava Rima (talk) 02:50, 7 July 2010 (UTC)

Problem with JWS's proposal is that he is attempting to impose local jurisdiction on stewards. If there is to be such a policy, it would be on meta, as a policy regarding the actions of stewards with respect to local wikis. Quite simply, Wikiversity is not an independent, sovereign entity, though it does enjoy a measure of independence (more than it used to enjoy). Do not implement in policy what the policy-making body cannot enforce. The community here cannot enforce any restrictions on stewards, and we cannot, through a Wikiversity discussion, decide upon and effect any sanction against a steward. If we have a complaint, we can go to meta. We could also go to meta and ask for a policy change there. For all I know, there is already some policy in place, possibly being disregarded.

Further, if there is a problem with a steward block, any local sysop can reverse it, there is no over-riding "freeze" that a steward can impose to prevent this. All the steward could do would be to desysop the custodian, which is a fast track, quite possibly, to losing the steward bit. Stewards are trained not to jump into catfights like that. Only if the steward were actually imposing some local but ineffective consensus would the steward get away with it, or, alternatively, that there simply is not enough local community awake and active to oppose. In which case a local policy would be of no help at all. Stewards aren't required to know every local policy!

Bottom line, JWS is reacting to events where a steward or equivalent came to WV and acted without getting local consent first. As long as Wikiversity remains under the "wing" -- or "thumb" -- of the WMF, this possibility must exist, absent an alternate path to determine and handle emergencies. JWS seems to want absolute freedom. It's not going to happen, not because someone is preventing it, per se, but because absolute freedom in organizations is an oxymoron. --Abd 22:11, 21 July 2010 (UTC)
 * "JWS seems to want absolute freedom" <-- Abd, what are you talking about? The proposal is for the Wikiversity community to be involved in decisions at Wikiversity. --JWSchmidt 22:39, 21 July 2010 (UTC)
 * No, this is not merely as you described it, "involved." The Wikiversity community is involved in that it may comment on meta, generally. However, the proposal creates a rule (must first be discussed with the Wikiversity community) when the WV community does not have authority over the "non-custodians." I.e., stewards or staff, as examples. We may request such discussion, but we do not have the freedom to impose it. Now, we could certainly create such a policy here, and then claim that a steward is violating local policy. And this would then invite discussion of closing down Wikiversity, because we would be creating possibly disruptive local policy; interfering with stewards is disruptive, unless done within due process. We could, indeed, negotiate such a policy, but such discussion -- and the policy itself -- should be on meta or with the WMF directly. --Abd 15:44, 28 July 2010 (UTC)
 * the WV community does not have authority over the "non-custodians." I.e., stewards <-- Totally wrong. "Stewards do not make decisions....Their task is to implement valid community consensus." --JWSchmidt 19:04, 29 July 2010 (UTC)

Recent changes 2010
This recent edit introduced many problems. Misguided new language: "or emergency, where delay would result in a necessity of extensive cleanup" <-- What "emergency", besides large amounts of ongoing obvious vandalism, requires an immediate block? Misguided new language: "be satisfied that an editor realizes that a community norm is being violated, such that a reasonable editor would expect the possibility of being blocked" <-- Given the misuse of the block tool documented at the community review, the Wikiversity community cannot trust sysops to make unilateral block decisions about anything except obvious vandalism. The recent edit attempts to introduce a disruptive blocking policy from another wiki that is not suited to a scholarly learning community where decisions are made by community consensus, not by rogue sysops. --JWSchmidt 17:37, 29 July 2010 (UTC)
 * Thanks for your response, JWS. Let's deal with one issue at a time, okay? What I introduced comes from over twenty years of on-line activity, including moderation and dispute resolution. Not from Wikipedia.

Emergency
Wikis depend on volunteer labor, generally, and some kind of operating consensus. Incivility poisons the collaborative environment that wikis depend upon. There are some people, and this tends to depend upon age and social background, who can tolerate high levels of incivility without harm, there are others who are very intolerant of it and will exit rapidly any environment where incivility is tolerated. The level of incivility to tolerate is thus a complex judgment. Ideally, none is tolerated, and incivility is treated as a sign that something is breaking down. Identifying a "bad guy" and blocking him or her is not a particularly safe approach, because incivility may be related to some complex interactions and we may only be seeing a particular product of that. On Wikipedia, a classic cabal technique, with the "climate change" situation currently under arbitration, was to greet newcomers with a "bad POV" with incivility, including incivility from administrators, tag team reversion, often using assisted editing tools and no clear edit summaries, and no discussion, followed by rapid blocks from cooperating administrators based on resulting incivility (complaint about being treated badly!), or on revert warring. One editor, and these editors were almost always isolated -- because the previous ones had been blocked and banned! -- is a sitting duck to be sucked into 3RR violation.

Where administrative abuse is subtle like this (the cooperating administrator was not necessarily an editor of the article itself), where it involves several cooperating editors, it's very easy to look at each situation and only see that here is a single editor, perhaps a single-purpose account, going "against consensus." When neutral administrators intervened, the cooperating admins would wheel-war, and if a complaint was filed, there were enough members of the cooperating faction to show up to prevent consensus from being formed. They would then claim exoneration even though the actual !votes, in several cases I counted, were certainly not a supporting consensus for their position.

These are wiki problems caused by a lack of efficient dispute resoluton process. The idea that "decisions are made by consensus," is inadequate to address it, and even establishing an ArbComm on Wikipedia has largely been unable to cope with it. There are a number of classic solutions, and "discuss everything" isn't one that works. Believe me, it's been tried many times. Communities that depend on consensus for decision-making, purely, burn out, I've seen it many times, up close. It starts out by being very exciting, people are amazed to discover that, with sufficient discussion, it seems that any disagreement can be resolved. Partly that's true and partly it is an illusion, because many people will simply give up, decide that opposing what they think is a majority -- or even a stubborn minority -- isn't worth the effort. So resentments can build up over the years. People find that they don't have time for all the meeting process, over and over for every decision. Consensus organizations become highly conservative and can find it difficult to adapt to new circumstances; I've seen situations where "consensus' had become "minority rule," where the status quo favored a minority, where even a majority of people complaining about the status quo, being hurt by it, couldn't change anything.

Bottom line, it is impossible for "emergency" to be decided by consensus. "Emergency," then, is a matter of judgment by those who have the power to act. If such people make poor judgments, and do not respond to correction by the community, they should be removed from power, which requires a sensible removal process. Many people will affect themselves to power, almost instinctively, so trying to remove an abusive sysop on Wikipedia can be like pulling teeth. I've seen sysops who blatantly violated recusal policy, who used obscene language to insult editors they were blocking, who used their tools to promote their own POV, and undeniably so, who still received nothing more than an admonition when the matter, with great effort (read "disruption") was taken to ArbComm. The committee did not require, even, any show of understanding from the admin. So, big surprise, the admonition accomplished little but that his offenses simply became a bit more subtle. Not much more!

The wiki system depends for administration on distributed authority. It is possible to design a system where sysops only use the tools under very narrowly defined circumstances, but this will, guaranteed, leave a consensus community without the means to respond to emergencies. What kinds of emergencies? Incivility is a big one, so is outing or other harassment, including RL harassment; there are also legal threats, revert warring or wheel warring. Sysop power is a kind of police power, and sane societies do two things: allow police discretion, and review police actions and remove abusive officers from power. A police officer is empowered to do almost anything perceived as necessary for the protection of the public. They are not limited to precisely defined categories. Don't confuse this police discretion with conviction for a crime and resulting imprisonment. A police officer may temporarily jail someone, but cannot punish them, legally. If they do, they will normally lose their job.

So the analogy on a wiki would be that administrators may block, but only the community may ban. In some cases the "administrative corps" may stand in for the community, hence the defacto ban that exists on Wikipedia, that of a blocked editor whom no admin is willing to unblock. That becomes circular, though, where it was assumed that a single administrator could not, then, unblock this "banned" editor, I argued strongly that these were simply blocks, not bans. A ban should require a consensus, period, and supermajority one, which, by the way, didn't exist for our favorite model of a disruptive editor here. I didn't do a head count yet, but the ban discussion was about 50-50. But that's a different decision from a block.

A block is merely an enforced "Stop it!" Absent emergency, it should always be preceded by a warning, and if the editor disagrees with the warning, fine. An editor should never be blocked merely for disagreeing with an administrator. But if the editor then disregards the warning and repeats the problem behavior, the administrator may block. The disagreement that is routine (who likes to be restricted?) is not sufficient, to, in itself, create a recusal requirement. However, once the admin has blocked, the admin should either unblock upon being convinced that the danger is past, which could include an agreement from the editor to not repeat the behavior until community discussion shows it is allowable, or should defer to the decision of another, uninvolved administrator, if that is possible. If there is a shortage of administrators with a user who has been a long-time problem, at the least the second administrator should be a different one. Examples of admin abuse that I've seen recently on Wikiversity has been a block by an admin, who then declines the unblock. That's not an emergency. The admin should adequately explain the block reason, with sufficient evidence, and may request that any admin considering unblock discuss it first (though it's better, on principle, for the blocking admin to walk away, having provided simple evidence devoid of passionate argument, which tends to harden conflicts), but should never decline the unblock request. If it sits there, it sits there, until a neutral admin declines, or, alternative, unblocks. And in any case, any blocked editor should be helpfully assisted to appeal. A long-term danger is that administrators will tend to mutually support each other. That's a good thing! But it can also lead to unfair decline of unblock requests. Thus there should always be an appeal to the community.

Or to an executive or judge or arbitrator. There are many ways to do it. Ultimately, the community is the core resource of any wiki, but the problem is getting sufficient community attention to problems without having every problem become thoroughly disruptive. In my view, the community should prepare for incidents like those of 2008 and March 2010, by developing mechanisms for rapid negotiation and expression of broad consensus. It can be done. And with that, this community could easily have negotiated with the WMF to obtain whatever consensus was actually found.

There are major political principles involved here, that cut to how society evolved over time. Early social structures were probably almost always relatively egalitarian and consensus-based, where tribal units were small. However, small tribal units could not compete with the efficient organization of strong-leader societies, which generally, as large societies developed, were able to easily overpower and absorb (or exterminate!) small consensus societies that they met. Branfman et al show an example of an egalitarian society that was able to survive for hundreds of years in contact with the Spanish, the Apache, but, in the end, the Apache were vulnerable when the larger society became a little more sophisticated.

So there is a natural evolution, small consensus -> oligarchical or single leader -> hybrid democratic. Democratic societes are generally whomping dictatorships, but still maintain aspects of oligarchy, and certain use central-control models, even democratic organizations, i.e, organizations to promote democracy, frequently are internally authoritarian. It's considered more efficient.

My own view is that we can move beyond this to hybrids that return to consensus as the goal, but which use deliberative, protected process to negotiate it, structuring the massive discussion necessary for consensus into a fractal structure so that it only expands as needed. There is quite a bit of this I found incorporated in Wikipedia dispute resolution process, but without formal structure, big surprise, w:WP:DR has been far more bypassed than used. I was able to use it to actually resolve disputes on Wikipedia, turning editors who were about to get each other blocked due to incivility and revert warring, into cooperative friends. But few were watching. And it's much easier to haul an editor before a administrator's noticeboard than to enter the extensive discussion that is required to find true consensus.

But having a corps of mediators who are trained in dispute resolution, and who can also serve as mentors when needed, guiding editors to collaborative behavior (as well as protecting them!), could greatly reduce the need for blocks.

Bottom line: emergencies are defined, routinely, by emergency services, i.e, by the police, ambulance, fire, etc. On wiki, sysops who have the tools. Not by courts or by public debate and discussion. Don't trust them with the tools? Remove them from power, or, possibly, restrict the tool use. A sysop who is great with, say, deletion cleanup, knowing when to summarily delete (which should always be appealable, but it's a bad idea to require discussion first in all cases, highly inefficient, much easier to allow deletion and then only review contested cases), and when to refer to a discussion, may not be good for, say, intervening in civility problems. While the software doesn't allow discriminating on this (true?), any good sysop would voluntarily accept proper restrictions. --Abd 19:34, 29 July 2010 (UTC)

I have collapsed the above because it is so long. If anyone wants to extract and summarize important points here, immediately below (not indented), in part or in whole, that's welcome, please sign a summary; I'll edit it if I disagree with anything or need to clarify, with reference to a diff or version of the original. --Abd 20:11, 30 July 2010 (UTC)


 * "it is impossible for "emergency" to be decided by consensus" <-- Wrong. Wiki communities have long ago identified repeated vandalism as the only emergency that warrants quick sysop action. Custodians are only empowered to block vandals without community discussion. All other use of the block tool should follow community discussion. "Sysop power is a kind of police power" <-- If you want to be a wikicop, go play the wikicop MMORPG at Wikipedia. At Wikiversity, sysops are Custodians who clean up spills (obvious vandalism) and do other jobs that are requested by the community following a community discussion. "the community may ban" <-- I doubt if the barbaric practice of banning has any role to play in a scholarly learning community. If you want to have bans at Wikiversity then make Bans an official policy. "Examples of admin abuse that I've seen recently on Wikiversity has been a block by an admin, who then declines the unblock." <-- This is a matter for community review. It would be useful if you would comment there. "Thus there should always be an appeal to the community." <-- The "appeal" to the community should come before the block, in community discussion before the block. "prepare for incidents like those of 2008 and March 2010, by developing mechanisms for rapid negotiation and expression of broad consensus" <-- Wikiversity just needs official blocking and deletion policies that say the community decides, by discussion, on blocks and page deletion, before the block and delete tools are used. Abd, please revert your misguided edit and propose your ideas on this page and allow community discussion of your ideas before trying to alter the proposed policy. Your attempt to bring the destructive methods of Wikipedia into this community of collaborating learners is misguided. --JWSchmidt 09:21, 30 July 2010 (UTC)
 * Your still living in the past. Just recently another person suggested that Custodians should be empowered more or else English Wikiversity should be shut down. -- dark lama  13:17, 30 July 2010 (UTC)
 * Darklama, I reject the idea that we should be guided in our actions by misguided bullying. --JWSchmidt 13:32, 30 July 2010 (UTC)
 * I don't consider the way it was phrased bullying, and Wikiversity is free to be guided by whatever actions it wants. You may (or may not) be a member of the Wikiversity community, but you are not the Wikiversity community. -- dark lama  13:38, 30 July 2010 (UTC)
 * Making a threat about closing Wikiversity after being gamed into deleting a harmless learning project (designed to find an ethical way to improve Wikimedia projects) was bullying and a way to intimidate the Wikiversity community. --JWSchmidt 13:45, 30 July 2010 (UTC)
 * What you are talking about isn't the recent suggestion I'm talking about apparently. I'm not aware of any learning project being deleted as a result of that comment. -- dark lama  13:56, 30 July 2010 (UTC)

This is what happens with JWS discussions, they spin off endlessly into subarguments, with no consensus being acknowledge on even one single point. Only what is seen as wrong by him gets attention. This process is guaranteed to perpetuate disruption and disagreement. I encourage JWS to start with what is agreeable, and to start to assume good faith, which means sensible meaning, instead of projecting false meaning and then arguing against it. The question must become "what's right?" instead of "what's wrong."

So, single point: JWS quotes me and responds: If so, then vandalism is not an "emergency," which means, in the context of my comment, an unanticipated situation, not covered by policy, that, in the view of the custodian, requires immediate action or damage will ensue. That sysops or other privileged users (such as moderators on mailing lists) are allowed to make such decisions, ad hoc, is so routine that I'm astonished that JWS thinks the reverse. If this is the only situation requiring "quick sysop action," where is that documented? It is also, by the way, an unenforceable restriction, because any sysop who sees the project threatened by immediate danger is quite likely to act anyway, no matter what policy says, that that's what Ignore all rules means. Sometimes people think IAR is just some Wikipedia quirk, when, in fact, it is clearly the old common-law rule of Public Policy, that trumps the literal interpretation of law on occasion. And then there will be review, of such actions, if anyone objects to them and follows due process instead of just frothing at the mouth, and, quite clearly, a sysop who more than rarely defines "emergency" outside of reasonable limits -- and there can be very little a priori consideration of this, by definition, since "emergency" here means a perceived necessity not covered by policy -- cannot be trusted with the tools. The defect on Wikiversity isn't abusive sysop action, it is lack of systematic and effective review and decision about such action, after the fact. There will always be abusive action, at some level.
 * "it is impossible for "emergency" to be decided by consensus" <-- Wrong. Wiki communities have long ago identified repeated vandalism as the on[ly] emergency that warrants quick sysop action.

In a review of sysop actions, whether or not recusal or other policy should have prevented the action should be considered, as well as the wisdom of the action itself. There are various conditions that can be determined by the community:
 * The action was correct and the admin was not prohibited by rule from action. Action and admin confirmed.
 * The action was correct but the admin was violating, say, recusal policy. Action confirmed, admin commended for it, and whether or not there was an emergency may be considered. If there was no emergency, the admin might be warned against taking such actions, for they can cause problems even if correct. If the finding of emergency by the admin is confirmed, though, the admin should again be commended, additionally because of the boldness of the action, for admins should know that when they act against rules, they are taking a risk of losing their tools.
 * The action was not correct, but the administrator reasonably considered there to be an emergency. The reasons why the action was not correct should be explained to the administrator, and, provided that there is no fear that this kind of action will commonly repeat, through assurance from the administrator showing that he or she understands the issues better now, lifting of the tools would not be normal. Administrators are volunteers, not professionals, and perfection is not properly expected even from professionals.
 * The action was not correct and there was no reasonable determination of emergency. Here the continued possession of tools by the administrator should be questioned. There is a possible ameliorating condition: if the administrator made an emergency decision using tools, but also recognized that this could be controversial and immediately consulted with the community, bringing the action to the attention of other administrators, and not wheel-warring or showing other forms of tenacious clinging to position, the administrator might be admonished for overreaction to a non-emergency, but proceeded correctly nevertheless. Only if such actions were common, creating substantial disruption, would removal of tools properly ensue. At issue would be the level of damage caused by the improper action. if it was improper but harmless, the only worry would be that in a similar situation, where deeper damage might be caused, the admin would repeat the error. A showing that the admin understands the error, again, could be very important. On Wikipedia, a common process error is to assume that if an admin is admonished, it will be sufficient to correct the situation, when, in fact, admonishment that is not acknowledged and accepted can harden the situation, where the admin feels that the community has unreasonably failed to understand him. I've suggested on Wikipedia that ArbComm much more easily suspend admin privileges, pending the receipt of assurances that recusal failure (typically that's the problem) will not repeat. Voice in the wilderness, I'm afraid, so far. Eventually, though, they'll get it. "Suspension" could be as simple as a direction not to use tools at all, or not to use tools in some particular area, and it could be in expectation of voluntary compliance: violation would result in immediate desysop, but, in the mean time, the person would continue to be, overall, trusted.

Because JWS has seen recusal failure, without the community reviewing it and coming to consensus about it, he would like to limit administrative discretion. That's the wrong solution, entirely and, rather obviously, unenforceable. Rather, what we need is something new: much more efficient and rapid review, by the community -- or by a representative or representatives of the community -- of controversial decisions by administrators. And I hope to be making suggestions about that. I am, by the way, quite happy for any of my decisions, as a provisional custodian for almost two months, to be subject to review, just as I've seen problematic decisions by other administrators during that period that I think should be reviewed. The discussion of these things, including possible errors, must be divorced from a very bad habit of blame and condemnation, which is a poor way to treat volunteers. It is essential, however, for any custodian be open to criticism from the community (and even from blocked editors, within limits), or else custodianship can easily become -- can be predicted to become -- abusive. Custodians should therefore become less sensitive to criticism than normal users, it is probably one of the necessary qualifications. But not thick minds! Historically, even wise autocrats permitted and encouraged criticism within respectful limits, and carefully considered it; when they didn't, their dynasties ultimately fell, the results were disastrous. --Abd 20:06, 30 July 2010 (UTC)
 * "projecting false meaning and then arguing against it" <-- Abd, what are some examples of "projecting false meaning"? "JWS thinks the reverse" <-- Abd, please provide evidence to support your claim about what I think. "If this is the only situation requiring "quick sysop action," where is that documented?" <-- Sadly, Wikiversity needs a good blocking policy to explain common sense to some people. Abd, if there is some other need for quick blocking, then please explain that need; list one thing besides repeated vandalism that requires an immediate block. "he would like to limit administrative discretion" <-- Existing Wikiversity policy limits how Custodians can use the block tool. A few rogue sysops have violated policy and misused the block tool. The Wikiversity community now needs protection against further disruptive use of the block tool. "what we need is something new: much more efficient and rapid review, by the community -- or by a representative or representatives of the community -- of controversial decisions by administrators" <-- Wrong. The Wikiversity community first decides by consensus if a block is needed, then, if needed, a Custodian imposes the block. "Custodians should therefore become less sensitive to criticism than normal users, it is probably one of the necessary qualifications." <-- Custodians should meet their obligation to explain their actions. --JWSchmidt 17:43, 1 August 2010 (UTC)


 * To respond to questions with any depth can take far more words than the question. So barrages of questions like this, if answered, can rapidly expand discussions into impossible tomes. I'm going to answer, each question, but within collapse. My condolences if you think you have to read this. The possible value of it is that it causes me to review my understanding of wiki common law, for those who might be interested. --Abd 03:39, 2 August 2010 (UTC)

JWS:"projecting false meaning and then arguing against it" <-- Abd, what are some examples of "projecting false meaning"?
 * I had in mind the usage of "emergency" above. That is, of course, what we were discussing. "Emergency" in this context means a situation considered to be urgent, requiring immediate action, by someone with the ability to act, and not covered by guidelines or policy. If it is covered by policy, the action is properly guided by policy and is routine. On all wikis, policy is generally kept relatively vague precisely to allow discretion. JWS projected a false (i.e., unintended) meaning on this, above: Wiki communities have long ago identified repeated vandalism as the on[ly] emergency that warrants quick sysop action. In this, "emergency" is conflated with "quick." Handling vandalism is indeed clear and routine, but something preventing the routine handling of vandalism could be considered an emergency (example: sysop is blocked, policy suggests that sysop not unblock self. But sysop sees vandalism that should be immediately reverted, and there are no other active sysops. So sysop unblocks self, handles vandalism. Properly, emergency over, sysop might reblock self, but simply refraining from any controversial use of tools pending consensus on the block would be adequate....) "Vandalism" is not the kind of emergency contemplated. Suppose a sysop sees a budding flame war and this could result in the loss of a valued user or other damage (many people I know will leave a community without comment where incivility becomes common, sometimes a single incident is enough. A community discussion can take days or longer, by which time the damage will be done. Sysop cannot remain on-line and simply revert uncivil edits. So sysop may block, even without warning, through warning is greatly preferred, with block only if the warning is ignored. Such blocks should be temporary and not allowed to take the place of discussion and consensus, long-term.

JWS:"JWS thinks the reverse" <-- Abd, please provide evidence to support your claim about what I think.
 * Actually, JWS is free to tell me what he thinks if it contradicts what I say. I am obviously not an expert on what he thinks, I only infer meaning from his words, and the expression "[he] thinks" is colloquial for "his expressed position is." This is an example of how routine conversation, with ABF interpretation of the words, becomes a fertile bed for more dispute. What was the issue here? I had written:


 * If so, then vandalism is not an "emergency," which means, in the context of my comment, an unanticipated situation, not covered by policy, that, in the view of the custodian, requires immediate action or damage will ensue. That sysops or other privileged users (such as moderators on mailing lists) are allowed to make such decisions, ad hoc, is so routine that I'm astonished that JWS thinks the reverse.


 * What had JWS written that led me to say that he "thinks the reverse"?

JWS:''Wiki communities have long ago identified repeated vandalism as the only emergency that warrants quick sysop action. Custodians are only empowered to block vandals without community discussion. All other use of the block tool should follow community discussion.''


 * This was the reverse of what I'm claiming is routine everywhere, including but not only Wikiversity, and JWS made it clear that he was talking about universal practice, when, in fact, I know of not one wiki that follows his interpretation. What JWS here has done, by not cleaving to the actual controversy, but rather to some supposed mind-reading offense, as he implies, repeated over and over, causes discussions to spin out. My comment was simply a clear summary of JWS's expressed position, the reverse of what I had stated from my own experience of other wikis, as well as my knowledge of Wikiversity policy. Custodians have discretion, and if we don't want that, we must provide for rapid and efficient process to desysop for "abuse of discretion," because there is no way of preventing discretion from being exercised. If there were a way, i.e., by bots or something, we wouldn't need sysops! I actually proposed, on Wikipedia, something that might satisfy JWS. I suggested that much clearer recusal guidelines be written, that any registered user, under certain conditions, be allowed to protest an actual administrative action, and that the individual custodian would be required to recuse from further action, absent emergency, until community process allowed it. It was considered preposterous, but for shallow reasons. This could not be abused by wikilawyers, and abuse of this right to say, attempt to make oneself unblockable, would rapidly lead to an indef block, probably after about three demands for recusal. "Recusal" does not mean "unblock," it means to abstain from future blocks at least pending discussion and permission to resume. (With many admins, there is little reason to set up conditions of some personal battle between an individual admin and an individual editor, but with very few, conditions can be more difficult.)

JWS:"If this is the only situation requiring "quick sysop action," where is that documented?" <-- Sadly, Wikiversity needs a good blocking policy to explain common sense to some people. Abd, if there is some other need for quick blocking, then please explain that need; list one thing besides repeated vandalism that requires an immediate block.


 * The barrage of questions abstracts statements from their context, and without context, it all becomes word salad, in this case with rancid dressing. What had JWS written?

JWS:Custodians are only empowered to block vandals without community discussion. All other use of the block tool should follow community discussion.

He was citing policy, as he has frequently done, implying that policy restricts custodial blocks as he states. He also cited himself, in his "Problematic actions" Community Review. He is acknowledging, here, without saying so explicitly, that policy does not prohibit blocks for other than vandalism, taking the position that custodians are only "empowered" to do what is specifically stated in policy, when, in fact, with no restraint process, they are empowered to do almost anything they want, that's part of the problem. The solution, though, is not to restrict discretion, it is to provide for efficient process to restrain custodians who abuse discretion. The discretion has always been necessary and will remain so. As policy is developed, to be sure, the scope of discretion will narrow some. But, guaranteed, unless some alternate method of handling emergencies is developed, it will not become so narrow as to lead to automatic desysopping for blocking other than vandalism.

Many of the blocks he's complaining about are blocks for block evasion. I.e., blocked editor -- it always seems to be Moulton -- edits in spite of block, as IP, so a custodian blocks the IP without discussion, and perhaps without any consideration of whether or not the contribution is "positive" (which is a very complex judgment that shouldn't be in the hands of custodians specifically, except ad-hoc and temporarily, under emergency conditions). It appears that JWS would have the custodian evaluate each edit, and, if it is a "good-faith contribution," by some unspecified standard, requiring the custodian to make a content judgment rather than a behavioral one, then the edit should be left. This, in fact, is a standard that would make blocks meaningless. If each edit while blocked must be evaluated, then there is no reason to block! Rather, any editor could revert a bad edit. The purpose of blocks, in fact, is to prevent the need to evaluate edits of an editor, once it has been considered to be a waste of time. That's true for vandalism as anything else.

Instead, this is the "wiki solution": Once an editor has been blocked, it is standard that all edits of that editor, identified as such (and Moulton is usually cooperative here, and I meant that, i.e., we should appreciate how easy he makes to identify his edits), can be reverted without examination. But any editor may examine those edits and revert back in the edits on their own responsibility. Because these block-evading edits were being routinely deleted, as they still are, and because JWS was not reverting them back in, but was, rather, repeatedly complaining about "censorship," I conclude that his purpose is not to bring in the useful contributions of Moulton, but rather to continually protest the situation by stirring up as much fuss as he can. While I can appreciate that, as a motive, and if I've correctly divined it, it's not going to fix the problem. Taking steps to establish a new, and better protected, collaborative environment will. JWS's hysteria must stop, as part of this, but blocking JWS, per se, isn't the solution. I proposed a topic ban, and, properly administered, and with reasonable cooperation from JWS -- who should be able to choose a mentor willing to accept him -- he could still engage in criticism, but within due process and with collaborative restraint. Same as Moulton, in fact.

It is the fact that we had no coherent process for restraining Moulton except the blunt and relatively inefficient instrument of the block tool, no way to negotiate with him that left him as a peer participant, duly respected, that caused what could have been a relatively minor flap to spin into two years of disruption, and counting. The problem is not Moulton, or Jimbo, or Adambro, or JWS (or myself, some are now suggesting that I'm the cause of recent disruption, which is preposterous, overall, even if I did make mistakes, which remains to be seen), it is the lack of efficient and coherent consensus process, or other means of making rapid decisions that anticipate or invite true consensus.

And at this point, I simply have no more time to give to responding to these questions. Perhaps I'll return. Right now, I'm being threatened with a block by Ottava for reverting him once, when I was clearly following policy. Cool. JWS, will you strain at a gnat and swallow a camel, or are there principles behind your position? --Abd 03:39, 2 August 2010 (UTC)

--Abd 03:39, 2 August 2010 (UTC)

Conflict of interest

 * The recent changes to the proposed blocking policy constitute a total change from the policy that was developed by community consensus over the previous four years. People who have violated Wikiversity policy and disrupted the Wikiversity community by their misuse of the block tool should not be writing Wikiversity policy for blocking. As discussed below on this page, this matter is under community review. "What you are talking about isn't the recent suggestion I'm talking about apparently" <-- Darklama, the comments by User:Daanschr were misguided. Daanschr repeated the false claim that Wikimedia Ethics/Ethical Breaching Experiments "was intended to disrupt other Wikimedia projects". The Wikiversity community knew about that page and knew it was a harmless learning project, but it was deleted by an outsider who made no attempt to first engage in Wikiversity community discussion. The deletion was instigated by a Wikipedian acting at Jimbo's Wikipedia talk page, someone who falsely claimed that the "Ethical Breaching Experiments" learning project was harmful. That project was a useful search for an ethical way to improve Wikimedia projects. Darklama, please explain why Daanschr's misguided discussion comment is relevant to Wikiversity blocking policy. --JWSchmidt 13:42, 1 August 2010 (UTC)
 * In the same sentence JWSchmidt notes that this is a proposed policy then suggests that the recent changes by Darklama "constitute a total change from the policy that was developed by community consensus over the previous four years". If there was community consensus then this draft could have been made policy. That it wasn't suggests there wasn't clear community consensus. Anyway, consensus changes and the incidents that have occurred in the last four years may have changed what the community feels this policy should say. I really don't see why JWSchmidt is arguing that those who he suggests "have violated Wikiversity policy and disrupted the Wikiversity community by their misuse of the block tool" shouldn't be involved in developing this policy. What he seems to forget is that whether it becomes policy will depend on whether there is community consensus that it should do. Therefore, if the community does feel that any of it is inappropriate then they will reject it and demand changes. I don't really consider it to be a problem for those who JWSchmidt declares to have violated policy to be involved in writing policy. One of the other strange things here is how JWSchmidt argues for the necessity of a policy on blocks whilst at the same time he is apparently able to say that certain people have undeniably violated policies relating to the use of blocks. If there was already clear policies on the use of blocks then we wouldn't need this. JWSchmidt is right that some of our existing policies do deal with blocking but that isn't in any great detail and leaves custodians, rightly or wrongly, with a great deal of freedom in how they use blocks.
 * To conclude, JWSchmidt arguments here don't really fit together. On the one hand, he states the necessity for a blocking policy whilst on the other he seems to suggest we don't need one because he is able to state that certain people have undeniably violated existing policy on blocks so shouldn't be allowed to be involved in writing this proposed policy. His suggestion that people should be excluded from participating in policy development according to his demands also doesn't really make much sense when the community as a whole will have the final say. Adambro 14:07, 1 August 2010 (UTC)
 * "If there was community consensus then this draft could have been made policy" <-- Not true. Since 2008 a few rogue sysops have systematically disrupted Wikiversity policy development by preventing needed policies from being made official. Those rogue sysops have instead enforced policies from other websites, often in violation of existing Wikiversity policy. Now that a new call has gone out for needed policy development, policy-violating rogue sysops are re-writing Wikiversity policy in ways that are disruptive to the Wikiversity community and its Mission. Adambro wrote, I really don't see why JWSchmidt is arguing that those who he suggests "have violated Wikiversity policy and disrupted the Wikiversity community by their misuse of the block tool" shouldn't be involved in developing this policy. <-- Here is an analogy: should convicted felons be allowed to impose a new law legalizing their crimes? Of course not. "What he seems to forget is that whether it becomes policy will depend on whether there is community consensus that it should do" <-- Adambro, there cannot be community consensus for anything that disrupts the Wikiversity community and deflects Wikiversity from its Mission. Adambro, please explain how your violations of Wikiversity policy and misuse of chat channel operator tools and your failure to explain your disruptive actions supports the Wikiversity project. "JWSchmidt is right that some of our existing policies do deal with blocking but that isn't in any great detail and leaves custodians, rightly or wrongly, with a great deal of freedom in how they use blocks." <-- No, such freedom does not exist. Use of the block tool is simple: it is for preventing repeated vandalism or it is used by community consensus. Adambro has incorrectly claimed that, "there is no policy on how blocks should be used", but that is not true. There are four existing Wikiversity policies that explain exactly how the block tool can be used, starting with Custodianship and including Civility. Adambro is one of the rogue sysops who have violated Wikiversity policy by using the block tool in ways that are not prescribed by Wikiversity policy. The Wikiversity community needs an explicit policy on blocking that will protect Wikiversity from further misuse of the block tool. --JWSchmidt 15:00, 1 August 2010 (UTC)
 * "Here is an analogy: should convicted felons be allowed to impose a new law legalizing their crimes? Of course not." So where has Darklama been judged by the community to have committed any "crime"? Nowhere. You don't have any right to disrupt policy development by unilaterally deciding particular individuals who you disagree with shouldn't be entitled to participate in that work. I note that you again suggest "exactly how the block tool can be used" is clear yet also suggest "The Wikiversity community needs an explicit policy on blocking". If the project needs an "explicit policy on blocking" then surely you are saying it currently isn't clear at the moment which voids the idea that you can so definitively declare someone has violated the policies relating to blocks. You arguments here just don't fit together. If "The Wikiversity community needs an explicit policy on blocking that will protect Wikiversity from further misuse of the block tool" then how can it currently be clear how the block tool is to be used currently?. If it was, we wouldn't need a blocking policy as you suggest we do. It doesn't seem to make sense. Adambro 15:28, 1 August 2010 (UTC)
 * where has Darklama been judged by the community to have committed any "crime"? <-- Adambro, what crime are you talking about? "You don't have any right to disrupt policy development by unilaterally deciding particular individuals who you disagree with shouldn't be entitled to participate in that work." <-- Adambro, I have not disrupted policy development. "unilaterally deciding" <-- Adambro, I have not decided anything. "If the project needs an "explicit policy on blocking" then surely you are saying it currently isn't clear at the moment which voids the idea that you can so definitively declare someone has violated the policies relating to blocks." <-- Wrong. The problems is, that you and a few other rogue sysops have gotten into the bad habit of violating Wikiversity policy by misusing the block tool. The needed version of the proposed blocking policy would help the Wikiversity community deal with people who ignore the existing policy on use of the block tool, existing policy that is distributed within four existing policies. Second try: Adambro, please explain how your violations of Wikiversity policy and misuse of chat channel operator tools and your failure to explain your disruptive actions supports the Wikiversity project. --JWSchmidt 15:45, 1 August 2010 (UTC)
 * Your comments are noted. I won't cause the discussion of the proposed blocking policy to be distracted any further by me responding. Your above comments mostly repeat similar comments elsewhere which I have seen and will respond to, or not, as appropriate. There are appropriate venues to raise concerns about my contributions or those by Darklama. Every discussion we participate in should not be seen as an opportunity to repeatedly make the same points. Adambro 16:25, 1 August 2010 (UTC)

Community review
Note: this policy is currently the subject of community review. --JWSchmidt 15:34, 30 July 2010 (UTC)


 * The recent alterations to this proposed policy do not address the issues that are under community review. Users who have violated policy by calling for and imposing unjustified blocks have a conflict of interest when they try to alter this proposed policy. --JWSchmidt 15:57, 30 July 2010 (UTC)


 * And users that have been blocked don't have a conflict of interest when they try to alter this proposed policy? -- dark lama  16:12, 30 July 2010 (UTC)
 * At Wikiversity, Custodians "block users from editing as prescribed by policy and community consensus". The block tool is for preventing vandalism. At Wikiversity, "blocking occurs in response to obvious and repeated vandalism". Darklama, when you blocked me you violated Wikiversity policy. Your past misuse of the block tool means that you have a conflict of interest with respect to creating policy for blocking. In my case, all of the blocks imposed on me were done in violation of Wikiversity policy and were bad blocks. I have every right to make sure that Wikiversity has a blocking policy that protects the community from continuing misuse of the block tool. Darklama, please undo your disruptive edit. --JWSchmidt 12:34, 31 July 2010 (UTC)
 * I find how you keep quoting "policy" that you made policy without discussion and consensus, funny in a fishy sort of way. -- dark lama  13:04, 31 July 2010 (UTC)
 * I and everyone else also have a right to make sure that Wikiversity has a blocking policy that protects the community. Not everyone is likely to agree what protection is needed. That is why there is a need to find common ground for what is agreed on. As long as you insist your way is the only way, people will continue to move forward without you. -- dark lama  13:09, 31 July 2010 (UTC)
 * "you keep quoting "policy" that you made policy without discussion and consensus" <-- Darklama, list some examples. The only thing "fishy" is your view of consensus, a view which does agree with the Wikiversity guidelines on consensus. "Not everyone is likely to agree what protection is needed" <-- In particular, those who have disrupted Wikiversity for the past two years by misusing the block tool are unlikely to protect the community when they try to re-write Wikiversity policy so as to allow their disruptive policy violations. "As long as you insist your way is the only way" <-- Darklama, all I insist is that destructive practices from other websites not be imposed on the Wikiversity community, particularly by a few rogue sysops who have a history of violating Wikiversity policy by their misuse of the block tool. --JWSchmidt 13:16, 1 August 2010 (UTC)

I'm happy to give you examples of when you didn't ask the Wikiversity community if it wanted the proposals to be made policy and gaining consensus for them. Here you go: Custodianship, Blocking policy, Rollback, and Verifiability. Rollback did eventually gained some consensus, but after most incidences that you say were violations. Nobody appears to have asked the community to adopt Civility as policy either and you do not appear to be responsible for that, but you keep quoting it as if the community had accepted its adoption. Consensus was adopted with only 3 people supporting it. Policies are only as good as support behind them. Policies with only 3 supports shouldn't be taken seriously as a community policy, let alone policies without even that. Reliable sources seems to be the only policy where discussion did happen for which you might of been right and not alone in deciding it is a policy. However I think the Wikiversity needs to do more to strengthen the integrity of all current policies, because the majority of people seem to ignore them, which suggests there is little support behind any of the current policies. Perhaps Wikiversity is still too small to be having any policies at all. -- dark lama  14:16, 1 August 2010 (UTC)
 * "you didn't ask the Wikiversity community if it wanted the proposals to be made policy" <-- Darklama, you have a mistaken view of how consensus is established. There is no requirement that someone "ask the Wikiversity community" before the tag is placed on a policy. If a policy is needed by the community then anyone can place the  tag on a page. If there is a valid reason, then the  tag can be removed from a page. In all the "examples" you listed, there was discussion and consensus for the policy. Civility is a guideline and it is now proposed that it should be official policy, since you won't follow the guidelines. "shouldn't be taken seriously as a community policy" <-- A needed policy, for which there is no argument explaining why the policy is harmful, should be policy. People who remove the  tag from a page without giving a reason explaining why the policy is harmful are disrupting policy development and disrupting the Wikiversity community. Darklama, when you enforced that kind of disruption by imposing a bad block you violated Wikiversity policy. Your past misuse of the block tool disqualifies you from now altering the proposed block policy. Darklama, please revert the proposed policy back to this version and recuse yourself from further editing of the proposed policy. --JWSchmidt 13:21, 2 August 2010 (UTC)

explain change to proposed blocking policy
Adambro removed a sentence from the block policy: Custodians cannot block editors again for 30 days when the Wikiversity community has unconditionally unblocked an editor. However, he left in place the reverse situation, thus imbalancing the policy toward blocking, which is the reverse of standard understanding. Accordingly, I have edited the passage to this:


 * The Wikiversity community may decide to conditionally or unconditionally unblock an editor or to deny an unblock request. Custodians should not block an editor when community consensus has determined that the editor should be unblocked, except upon a showing of new block-worthy behavior; similarly, custodians should not unblock an editor except under conditions which respect the concerns shown by the community in previously denying unblock, and which protect against further disruption. In particular, any block/unblock discussion should be closed and implemented by an uninvolved custodian, and that custodian may reverse the original close based on new evidence or arguments, either to block or unblock.

Adambro's concern about the sentence was correct, it improperly protected an editor for 30 days, denying the custodial discretion that is essential to functional wiki process. What I wrote is far more flexible; and if we develop a clear recusal policy, and better dispute resolution, almost all serious problems around blocks should be avoidable. The concept that a discussion is closed by a person, who is then able to reverse the closing decision on new evidence or argument, without requiring a possibly disruptive new discussion, is one that is a reality on some wikis, such as Wikipedia, but is often not understood, with some closers thinking that the community made the decision, not them. In fact, precedent is strong that a close should be based, not on !votes, but on evidence and cogency of arguments, and there is a classic example on Wikipedia where Jimbo closed a discussion against the !vote of, I think, 86% of commenters; I saw another MfD there where a close was similar. It went to Deletion Review, was overturned on a technicality, and then went on, being reopened, to precisely the same conclusion without a shift in the !votes.

What a closer could decide in the first place, the closer can return to and redecide upon new evidence and arguments, and we should generally trust neutral custodians to do this, which is why it's so important for a closer to be neutral; the closer should not be one firmly aligned with one side of a dispute. None of this is intended to make it difficult; the reverse, actually. If it is impossible to find a neutral custodian -- our custodial "cabal" is small -- then it's possible that the community could accept a non-custodial closer, who would simply *request* whatever custodial action was needed, which would normally be respected. (This is like a mentor for an editor, who can request that the editor be blocked or unblocked, and who can thus guide the editor while having enforcement power. Ideally, this mentor should be accepted by the editor, it will be more effective and less likely to result in disruption.). But usually that won't be necessary. --Abd 04:03, 2 August 2010 (UTC)

I added the 30 days bit. The intent was to encourage people to move on and give people time to calm down, as has been suggested by some participants recently, when the Wikiversity community decides that there should be no conditions for unblocking. In my experience recusel can cause problems on a small wiki because people can easily abuse the the recusel process to the point that everyone would need to be recused, and there isn't enough people participating in the community processes to have a clear consensus that a person is not involved and thus does not need to recuse. To use a chess term, recusel on a small wiki often leads to a stalemate situation. I doubt this problem is really unique to small wikis, but perhaps by the time wikis grow they have ironed out how to resolve a stalemate situation. -- dark lama  12:54, 2 August 2010 (UTC)
 * Recusal can create problems on a small wiki, but so can failure to recuse. The solution is the principle of emergency action with immediate referral for post-hoc review by the community (which can be done by someone with a recusal obligation, which should generally be disclosed in the review by the acting sysop if it is not obvious), and that review would include a consideration of whether or not the situation was actually an emergency. A custodian is dangerous who frequently takes emergency action found contrary to consensus in the absence of a genuine emergency (where waiting for a small community to find consensus would cause a serious problem). --Abd 14:42, 2 August 2010 (UTC)
 * Are you suggesting that Custodians should immediately recuse and initiate discussion when a block is performed without prior discussion? That would mean that an attempt at discussion would happen all the time whether before or after, if I understand what your suggesting. I can agree with that. -- dark lama  15:01, 2 August 2010 (UTC)


 * No, I'm not saying quite that, but it could sometimes be close. Recusal requirements have to do with involvement in dispute, or the appearance of same. That is, we should avoid the appearance of bias as well as actual bias. It can get difficult with a small wiki. We need, in any case, a decent recusal policy. What I've seen in the past in suggesting this is immediate objection that it would be wikilawyered to death. That's an error, partly based on a lot of experience without a good recusal policy. A good policy makes it easier for everyone except for truly disruptive users who get blocked just as quickly.
 * So, let's assume that an admin has a recusal requirement, from the history with a user or some other condition that presents an appearance of bias. The goal is to make a conflict with an individual user be between, not the individual admin and the user, but a matter for the relationship of the user to the community. A good example of what not to do is for a custodian to deny an unblock request after blocking the same user. Rather, we routinely expect that some other custodian would deny (or accept) it, and we hope for a neutral admin, who isn't biased by prior activity. Custodians are human, and when humans make decisions, we tend to become attached to defending them. That perpetuates conflict. Hence we expect that a blocking custodian, either directly and immediately or, possibly, later, if the user puts up an unblock request (immediately is better!), provide clear evidence for the block preferably with a diff or diffs, and, preferably, showing a disregarded warning. That then allows the blocking custodian to walk away, and make no more efforts to keep that particular user blocked. I can't say how much grief I've seen caused by a custodian who is attached to a prior block decision, it can start the whole community fighting. This concept that once you have blocked a user, you would be considered to be under a recusal requirement with respect to deciding that user's unblock request, is an example of routine recusal.
 * But what about emergencies? I've elsewhere argued that we cannot definitively define "emergency" in advance, we can only review it. We can provide for certain common contingencies, which, then, aren't "emergencies" as I'm using the word. It means "unanticipated situation which requires immediate action." In any case, fundamental wiki policy allows custodians to act for the protection and welfare of the wiki, contrary to explicit rules. But whenever this is done, and it should only be done in an emergency (otherwise it could be discussed first!), and especially if we are talking about recusal failure, where recusal policy would require the admin to abstain from action, it should be accompanied by an immediate request for independent review, not just on the user Talk page, but on a regularly watched noticeboard. It should not wait for an unblock template. If the custodian does this, they should be protected from negative consequence for error, unless the error rate has become worrisomely high. Blocking an editor, the custodian would properly instruct the blocked editor how to prepare an unblock template, it should be easy, and there should be a link to a page about how to do it, with some decent advice (Like, "Don't focus on the custodian's alleged error in blocking you, it's actually almost irrelevant. A neutral custodian looking at this will be interested primarily in whether or not your editing presents a threat of disruption. Look for ways to assure this custodian that any problems won't repeat, or that you will discuss and find consensus before, say, vandalizing the main page.... i.e., assure the custodian that you will follow collaborative process. Be polite and friendly, and we know that can be difficult when you've just been blocked ... but it can be done. Try to trust that the blocking custodian, even if he's obviously a bloomin' idiot, is doing what he thinks is best for the wiki." ... or something like that. A light attitude that also recognizes how painful it can be to be blocked -- it can be horrific -- may be best.)
 * So, with the emergency block, it's very important for the blocking custodian to detach, and really surrender to whatever a reviewing custodian or the community decides. We see a failure to do this when a custodian wheel-wars to maintain (or reverse!) a block. It's almost as bad, in effect, when a custodian vigorously argues that a user is a troll, only wants to demolish the project, and is completely worthless and an ****** to boot. Present the evidence and trust the community. The community won't always be right, but neither will you. Bottom line, how well this all works is dependent upon how well the community is at reviewing actions. In the end, that's where the most difficult problems lie. Recusal policy should be simple, and easy and efficient to follow. It can avoid a great deal of the unfortunate personalization of conflict that takes place. I've been there, improperly blocked, but by a very skilled admin who knew how to detach. I was left facing, not her, but the community. If I'd railed against how stupid she was, or how she had violated policy by dredging up RL stuff, etc., I might never have been unblocked.... and if she had railed against my obviously disruptive agenda (which another admin was busy doing), well, there would have been a lot of cleanup to do. I managed to heal the rift with both those admins, they became friends. It was harder with the guy, though, and ... I saw him run a fair amount of recusal failure.... he had sense enough not to block me, though, after I'd first confronted him about an abusive block of another .... I hope this is useful to you, Darklama, or to others. --Abd 20:52, 2 August 2010 (UTC)


 * I am not particularly crazy about the phrase: "Custodians should not block an editor when community consensus has determined that the editor should be unblocked, ...". Every editor is expected to abide community consensus, it shouldn't really be necessary to say so here.  I will see what I can do to improve it. Thenub314 13:21, 2 August 2010 (UTC)


 * Darklama and Adambro should not be editing this proposed policy. Their past misuse of the block tool and their past violations of policy are under community review. See Community review. --JWSchmidt 13:29, 2 August 2010 (UTC)
 * Custodians are allowed to participate in the development and expression of policy, and even custodians who have lost their tools, even for misbehavior, are so allowed. JWS is making up rules as he goes that have no basis either in policy or in actual practice, here or elsewhere. If JWS believes that those custodians should not participate in policy development, the procedure would be for him to propose and find consensus for a topic ban. He has found practically no support with his "problematic actions" filings. Some of the actions he reports are indeed problematic, but his filings are so unfocused and so voluminous that they are next to useless. --Abd 14:48, 2 August 2010 (UTC)


 * JWSchmidt, where is the harm in letting them participate? There are other non-custodians here like you and me to double check their work, and it doesn't become policy until passed by community consensus.  If anyone tries to slip something in to allow some game of the system I will be very vocal about it in my opposition, and so would many people I think.  I think they should be encouraged to be bold and take part.  We need all the hands we can get. Thenub314 15:25, 2 August 2010 (UTC)
 * "where is the harm in letting them participate? <-- People who have disrupted Wikiversity by misusing the block tool should not be allowed to change the proposed policy on blocks. The policy proposal should be reverted back to this version, and Wikiversity community members with no conflict of interest should be allowed to develop the policy. The changes that have been introduced by those who have a conflict of interest are not in the best interest of the Wikiversity Mission and contradict existing blocking policy. --JWSchmidt 18:19, 2 August 2010 (UTC)
 * I have to disagree, this is a wiki and they should be allowed to edit the pages they choose. It is up to the community to decide if the any version of this page should be accepted as policy or not.  I don't agree it should be reverted back to some other point, but that is my just my opinion. If others feel differently they can change it as they see fit, and I will do likewise. Thenub314 17:00, 3 August 2010 (UTC)
 * Thenub314, do you really think that a few editors can violate the existing policy on blocking, greatly disrupting the Wikiversity project and its Mission, and then simply re-write policy in a misguided attempt to facilitate their further disruption of the Wikiversity Mission? --JWSchmidt 17:36, 3 August 2010 (UTC)
 * I am assuming good faith on their part, so I do not accept they are "attempt[ing] to facilitate their further disruption of the Wikiversity Mission". Before you point out the current community review, I am fully aware of it, so safe your fingers from typing.  Since people are pulling out legal terminology such as recusal and dure process to describe the situation in that community review, let me remind you of another legal tenant.  They are innocent until proven guilty.  Until the community review concludes and has as part of its closure that these custodians should not edit policy, then I fully accept and appreciate their help, and I hope they reciprocate.  Before you call me naive, let me assure you I will watch all edits with an eagle eye and make sure that no one can abuse the system.  I hope everyone else does the same. Thenub314 21:25, 3 August 2010 (UTC)
 * A clear policy is easier for everyone to work with including Custodians. Custodians involvement in the policy development process can help to ensure policies that they are expected to follow are easy to understand and easy to follow. Custodians are human and like anyone might do, they seek guidance from policies. However when there is no policy, a policy offers no advice for a situation, or a policy offers vague advice, people end up policing themselves, including Custodians. If a policy places too many restrictions than nobody will want to volunteer, thats true whether the policy effects everyone or just Custodians. Involvement for everyone can ensure policies are neither too relaxed nor too restrictive that an a single person finds a way to game the system to get what they want or prevent others from getting what they want. -- dark lama  15:49, 2 August 2010 (UTC)


 * Yes, Darklama. As to the policy, I made another change based on the fact that a custodian who appears to be "contradicting" a prior community consensus may be, instead, implementing the sense and concerns of the consensus, having judged that conditions have changed. Generally, if in doubt, discuss, but discussion itself can be disruptive, and there can be reasons for going ahead and acting, particularly if the matter is going to be reviewed and disruption from the action itself (i.e., a previously blocked editor going on a rampage) is not expected to be significant. There is a danger which should be avoided, the idea that the "community consensus" actually makes decisions. Technically, the community does three things: provide evidence, argument, and opinion, for the guidance of a closer, who theoretically makes a decision based on arguments and evidence, not !votes. However, where it is clear that there is an overwhelming preponderance of votes in one direction, a custodian should be very careful about closing in another direction. I've seen such closes, however, on Wikipedia, and they were confirmed later. Sometimes !voters simply have not paid adequate attention to the issues. The danger, then, is that an inadequately informed "consensus" is taken as a confirmed status quo, without having a genuine, deliberated consensus behind it. The standard wiki solution is that closers can revise the close, which works if the original closer is available and was actually neutral. The closer is thus a single person with whom one can negotiate, if that's needed. It is almost impossible to negotiate with a community, or, at least, what can be negotiated is quite primitive. What if people come in and !vote, based on what is before them at the time, and then new evidence comes in, possibly leading to a contrary conclusion? Only a closer, or some process like that, can balance these things. I reviewing closes on Wikipedia, I generally looked at the later votes as the most controlling. With AfDs, there would be a rash of !Delete votes at the beginning, mostly based on what seemed to be knee-jerk reactions, without research, without considering attempts to fix the article, or because they didn't like the author or just wanted to rack up some !AfD votes. Later votes, because of the prior discussion, probably, as well as increased time for people to become aware of the discussion, tended to be more grounded and informed, and sometimes completely reversed the sense of the !voting. Some of this is due to scale on Wikipedia and does not particularly apply here, but there is still participation bias here, it's obvious. Most users don't watch process pages. Nor should they. --Abd 17:36, 2 August 2010 (UTC)
 * I think I'm not understanding what you are saying about Wikipedia, but I guess I agree with you in general. I used "apparent contradiction" in more places in the proposal and changed the order in hopes of showing how a decision can have negative effects a bit more. -- dark lama  17:50, 2 August 2010 (UTC)
 * Well, the Wikipedia stuff is just about the contradiction between decision by !vote (which is deprecated with few exceptions) and decision by a closer. No claim is made that we should adopt any particular Wikipedia policy, but WP experience can be a clue sometimes. Your change is reasonable for now; what should be avoided is tying the hands of custodians, while making it clear that "changing consensus" can be a problem. The real conclusion is that a custodian can do practically anything the custodian believes in best for the wiki, provided that, if it could be seriously controversial, the community is rapidly consulted; a custodian who makes bad decisions about this will face disruption and review, one who correctly anticipates consensus should not be faulted for doing so! The question in some review or feedback would be whether or not the custodian's decision was reasonable in context, and whether or not it properly leads to fear of future abuse. "Violation of policy" alone isn't quite enough, because any good custodian will at least occasionally violate the letter of some policy. It must be a harmful violation, not one that either saves trouble or disruption, or reasonably could have been thought to do so. --Abd 18:16, 2 August 2010 (UTC)
 * Involvement for everyone can ensure policies are neither too relaxed nor too restrictive that an a single person finds a way to game the system to get what they want or prevent others from getting what they want <-- No community can allow people with a serious conflict of interest to manipulate policies and introduce policies that damage the Mission of the community. People who have disrupted Wikiversity by misusing the block tool should not be allowed to change the proposed policy on blocks. The policy proposal should be reverted back to this version, and Wikiversity community members with no conflict of interest should be allowed to develop the policy. The changes that have been introduced by those who have a conflict of interest are not in the best interest of the Wikiversity Mission and contradict existing blocking policy. --JWSchmidt 18:25, 2 August 2010 (UTC)
 * JWS, do you have a conflict of interest in making changes to the proposed policy on blocks? -- dark lama  19:53, 2 August 2010 (UTC)
 * Darklama, you tell me. Who are the people who have disrupted Wikiversity by misusing the block tool? Given past misuse of the block tool, Wikiversity needs protections against further disruption of the Wikiversity Mission. This is a subject for community review. --JWSchmidt 17:44, 3 August 2010 (UTC)

Enforcement and self-reversion
I have made a proposed change to the proposed policy page, on block enforcement and the concept that identified self-reverted edits do not violate the purpose of blocks, and therefore should not be considered block evasion unless they are, in themselves, grossly disruptive. They are, in fact, like any other self-reverted edit, for it is commonly considered that prompt self-reversion avoids sanctions for what might otherwise have been a blockable offense. Examples would be revert warring, incivility, or, say, making extensive changes to policy without discussion! Had I made this change as a permanent one without self-reverting, and if I repeated such behavior after warning, I could be blocked for it!

Because this is a draft and is self-explanatory, it will probably require substantial editing, even if accepted in outline. It does outline, to some degree, actual practice, but the practice is ragged and uneven. This policy could provide guidance to users and custodians to the end of both improving content, reducing disruption over block enforcement (it is common for users to complain about ordinary block enforcement, i.e., the reversion of "harmless edits" by blocked users, but the fact is that no unreverted edit by a blocked user is truly harmless, and, on Wikipedia, topic-banned editors (quite similar to being blocked, but topic specific) have used blizzards of "harmless spelling corrections" to frustrate ban enforcement. Self-reversion was invented to provide such an editor on Wikipedia with an alternative that was simple and easy and that did not complicate enforcement. Since the editor, in fact, did not want to simply make spelling corrections, the editor was outraged by the proposal. After all, why should he revert a perfectly good edit! However, his intent became completely visible and he was site-banned for a time.

To facilitate working on this, I am creating a subpage, Blocking policy/Enforcement and self-reversion, and editing of this proposal to make it ready for prime time should be there, discussing on the attached Talk, with conclusions brought back here. Thanks. --Abd 15:37, 21 August 2010 (UTC)

I originally made this change here, and self-reverted to avoid making a major policy change without prior discussion. If anyone believes that the policy is immediately better with the change, it could be reverted back in. Thanks. --Abd 17:30, 21 August 2010 (UTC)
 * Nobody commented here or on the subpage, so today I went ahead and implemented the changes, in two parts. Please read them carefully. The section on block enforcement is fairly routine as an explanation of block enforcement. Self-reversion is process that has been seen to work, as a way for an editor to show cooperation and to encourage cooperation in return. It makes a block less of a slap in the face, less of a door slam, and more of a regulatory tool requiring that work be reviewed. Most truly disruptive editors will not self-revert, so, for them, this is moot. --Abd 02:32, 9 September 2010 (UTC)
 * Support must be demonstrated. A large amount of people making blatant supports is necessary for such changes. Ottava Rima (talk) 02:37, 9 September 2010 (UTC)
 * You just made that up, Ottava, this is not a policy page, it is a proposed policy page. You just effectively rolled back three edits, the first was probably mostly uncontroversial, about how blocks are enforced, the second was copy editing of a section, the third is, indeed, a new proposal, but one based on solid experience. I'd made other edits directly to the policy before, there is no policy prohibiting editing of a proposed policy, and it is preposterous to imagine so. I'll follow due process, you can be sure. I'm reverting you once. I advise that you work on the material, accept what is acceptable, and only reject what is not, please don't just do a blanket revert. The page, overall, doesn't become policy until explicitly accepted as such by the community. Please do not obstruct the development of the proposed policy. --Abd 02:48, 9 September 2010 (UTC)
 * If you do not like consensus, please feel free to find another website. Ottava Rima (talk) 02:56, 9 September 2010 (UTC)
 * Consensus? I love consensus. What I don't like is a sysop imagining that their own highly biased and warped opinion is consensus, ignoring plenty of signs. There was no consensus here, that's for sure, because there was no discussion. When there is no discussion, in spite of an invitation in the proper place for it, it is acceptable to act as one proposed. That is all that happened here. Had this been a policy page, Ottava would have been completely correct to insist on consensus first, or at least some sustained majority opinion. But it isn't a policy page, which means that the text does not enjoy consensus yet. See below.
 * Try as you keep trying, you cannot impart your fantasy upon reality. I have not opined on the proposals or matter, nor have I been involved in the discussion. You put forth dramatic changes that people were opposed to. You put them back after you were warned. Such behavior is completely unacceptable, and your fantasies are not enough to justify your participation here. Knock it off. Ottava Rima (talk) 18:58, 9 September 2010 (UTC)
 * Ottava, you revert warred here. Now you acknowledge that you didn't care about the content, you made a judgment that it was against consensus without having an "opinion" on it? How did you pull that off? Did you check the actual content to see if it matched what had been proposed here? Regardless, it now doesn't matter here. If you want to participate in the development of this policy, you are welcome. If not, please don't disrupt the process like that any more. --Abd 20:56, 9 September 2010 (UTC)
 * Removing disruptive additions by a disruptive user who does not have consensus for anything he has done or tried to do on Wikiversity is not revert warring. You were warned not to do anything like that and did it anyway. If you don't want to abide by our procedures and standards, please just stop completely. Ottava Rima (talk) 21:27, 9 September 2010 (UTC)
 * Take it somewhere else, please. This has nothing to do with this policy page. --Abd 21:37, 9 September 2010 (UTC)
 * I think you vandalising this policy, pontificating inappropriately, and being overall nasty has very much to do with your behavior on this page and your inability to discuss things appropriately. You were warned and kept it up. You obviously have no respect for directions or codes of conduct. Ottava Rima (talk) 21:48, 9 September 2010 (UTC)

I am personally of the opinion that the self reversion idea is not a workable large scale policy. It seems to me it just begs the formation of cabals where a blocked editor has an unblocked friend automagically revert back in his edits. If the editors are clever about how the choose to try to disrupt, then this could get very messy. Thenub314 05:01, 9 September 2010 (UTC)
 * Thanks, Thenub. That fear is at variance with the reality of the situation of blocked editors. If one editor incautiously reverts back in the edits of a blocked editor, either these are good edits, bad edits, or they are mixed. If they are bad or if they are mixed, that restoring editor would be waving a big red flag that says "block me." Because the blocked editor has already been blocked, presumably. And, if the situation is as Thenub imagines, that editor is being watched. Thenub hasn't watched self-reversion work, and I'd guess, he hasn't looked at the examples of it working. There is no example of it failing to work, i.e., if an editor is sufficiently cooperative to self-revert, so far, there are two kinds of community response. One is to ignore the self-reversion and block IP or the editor (for a topic ban) anyway. The result: the editor stops editing entirely, or stops self-reverting, and starts using socks or learning how to evade IP blocks. If this gains content for the project, it could well be questionable content. If the editor edits openly -- a self-reverted edit as I've recommended it would have, if I did it, say, in the original edit, "Will self-revert per block of Abd." And then the revert would state, again, that it was self-reversion per block of Abd. If the IP is not blocked, it might be followed. If this is used for a topic ban, it is easier to track and review. Go for "absolute block," and what you get is Scibaby, well over 500 socks and counting, a huge amount of wasted effort, including lots of time by highly skilled administrators, checkusers, and very substantial collateral damage. I was told by a functionary that much of the work of the Office was sorting out people who were blocked because of the massive range blocks Raul654 used to use to exclude Scibaby.
 * Would that have helped with the Scibaby situation? Maybe. Maybe not. It will not work with truly disruptive editors, whose very purpose is disruption. Self-reversion was actually invented as a way of dealing with "harmless edits" by a topic-banned editor, who rejected the idea. Why? Because his purpose was disruption! He was trolling to be blocked for a harmless edit, so he could then raise a fuss about it.
 * What Thenub misses is that this "cabal" he imagines, could already disrupt and create quite a mess. The question is whether or not self-reversion will make it better or worse for the blocked editor, for custodians, for editors who favored the block, for the others, if there are such, and for the project. Self-reversion is proven to produce good content through cooperation between editors who had been opponents. It can do that even if the edit is an error, there was a great example of that on Wikipedia. The restoring editor neglected to look at the result from the permanent link to the edited page; the edit was an attempt to fix bad reference formatting, and it was itself in error. However, nevertheless, the edit got attention and bad formatting that had been standing for quite some time was quickly fixed, probably more efficiently than if I had proposed the edit somewhere. I made that self-reverted edit, and, for my trouble, was blocked. (At that point, there was reason to believe that self-reverted edits would be acceptable, including an arbitrator opinion.) Later, of course, that sysop lost his bit, over the general situation vis-a-vis me. --Abd 17:39, 9 September 2010 (UTC)

What about the other two of the three reverted edits?
There has been some focus on self-reversion. However, I made three edits that were reverted:
 * re Block enforcement
 * copy editing in What is disruptive? < Is there any objection to the restoration of this edit?
 * about self-reversion.

These were in three different sections, and could have been considered separately, and could be restored separately, I believe. The self-reversion section follows logically from the Block enforcement section, but the Block enforcement section still works without it. Above, you can see that I created a page, Blocking policy/Enforcement and self-reversion, to work on the two major matters, excepting for the copy editing edit, which was ad hoc, yesterday. Please discuss that copy editing edit here, is it okay to revert it back in?

Unless someone concludes that the proposed policy is immediately better by simply reverting a section or sections sections back in, it might be better to examine the material on the cited subpage, using the attached Talk page to discuss it as we go. If those who work on this can agree, we can then bring this back here for final review before incorporating it. Or we could recognize that this page isn't policy, doesn't have consensus as-is, and can be worked on in the ordinary fashion. Or we could do both. I'm going to update the subpage according to what I actually put in last night, that was reverted. --Abd 17:53, 9 September 2010 (UTC)


 * I do not like either the first or third edit as they place into policy the self-reversion philosophy, the second is largely unnecessary. if anything we should limit the number of places to discuss the block to one or two key places that can be more widely watched.  To me Custodian feedback is more a appropriate place for discussion of some situation is over with, an edit/action that one didn't like, etc.  It doesn't strike me as the place to start the conversation "What's with the block?" or "I want to be unblocked." Thenub314 21:45, 9 September 2010 (UTC)
 * Thanks, Thenub. Of course there are issues around the first and third edits, much more the third than the first. The first edit does not "place into policy the "self-reversion" process," That's the third edit. The first edit talks about block enforcement and the rights of other editors with respect to the contributions of blocked editors. It doesn't mention self-reversion. I kept these separate. I was absolutely not expecting a blanket revert. Yes, the policy is, in my view, vague about the process to be used to question a block as improper, after the fact. It seems that the present process is this: discuss it with the blocking custodian, during the block but also possibly after; go to Custodian Feedback, and if that fails, go to Community Review (which is the only non-emergency process which can desysop here). But I don't understand your questions at the end. We should describe the unblock process, i.e., the placement of an unblock template, which should solicit a neutral custodian to review the block. --Abd 23:32, 9 September 2010 (UTC)


 * For the first edit it was the section "Any editor may revert, on sight, any unreviewed edit from a blocked editor, without regard to the content. It is not the content, however, which is being blocked, it is the editor. Any registered editor, seeing an edit from a blocked editor, may review the content and revert it back in, on his or her own responsibility." is a part I object to. Granted it didn't suggest a self-revert, but it was still part of the "revert then review" philosophy.
 * For the second edit, Now that I see what you were going for I still disagree. I don't mind the policy describing how to request an unblock, but I don't think it is a good idea to try to spell out in policy how to handle a bad block, this should just be handled on a case by case basis. Thenub314 02:14, 10 September 2010 (UTC)


 * Okay, maybe we are getting somewhere. You object to the statement that editors may review the content of contributions of a blocked editor, and, taking responsibility for it, revert it back in. This is standard practice. This has nothing to do with a "bad block," why do you introduce that idea? No, this part has nothing to do with self-reversion, except that this allowance, which is standard, at least on Wikipedia, sets up conditions where self-reversion can make sense. That's all. It's true and useful to know even if nobody ever self-reverts.


 * Dealing with this on a 'case by case basis', if by that you mean different rules for each blocked editor, is utterly impractical. The whole point of block policy is a kind of 'one size fits all' structure, for the simplicity. I.e., blocked is blocked; in theory, blocked editors are not supposed to edit at all. But they do, and what do we do with that? It's pretty settled, the edits may be reverted on sight, but there is no requirement that they be reverted. It's somewhat traditional to strike them (use strikethrough) if they are, say, !votes, especially if someone else has voted after. They can simply be left in place. In this sense, Thenub, it is not only case by case, it is edit by edit. It's standard wiki stuff.


 * The first edit, then deals only with standard block enforcement stuff. Then it deals with a common misconception. Obviously common, some around here may have it, the idea that there is something intrinsically reprehensible about seeing a reverted edit from a blocked user, deciding it is worth exposing it to view, and restoring it. There isn't. Does this "help the blocked editor," which is an argument that has been made? It might or it might not. What it helps is the project, in the view of the restoring editor. Who is not blocked. If an editor makes a lot of bad decisions in a matter like this, they will soon be blocked themselves.


 * Lurking underneath some of the opinions about how to deal with edits of blocked editors are ideas of punishment, which are entirely contrary to wiki principles.


 * We block to protect the wiki, not to punish, and we can use edits from a blocked editor if we decide that they benefit the wiki. If this editor was known for, say, using sources misleadingly, then those concerned with that are likely to look at these edits very closely if someone brings them back in. In practice, the only people who will see the edits, usually, are recent changes patrollers -- who will usually leave them alone unless they are spelling corrections and they decide to revert them back in -- and people interested in the involved articles, who are quite likely to be familiar with why the editor was blocked.


 * Yes, it's true that the issue of restoring edits of blocked editors is a foundation for the dependent concept of self-reverted edits. But first things first. Is edit one correct as to existing policy? If not, specifically how? Again, your response talked about bad blocks? What, indeed, does that have to do with the issue?


 * We cannot have one rule for "bad blocks" and another rule for "good blocks." The software doesn't know the difference. None of us are obligated to enforce a block that we consider "bad," nor even one we consider "good." There are some sysops who think it is their duty to immediately revert all contributions of a blocked editor, when recognized, and to immediately block the IP. Regardless of the content. That is allowed, but not required, by policy. Once we have the foundation agreed upon, that editors can restore decent contributions of a blocked editor, which is not only actual practice, but I've been doing it here, openly, for some time, and it is absolutely clear that the community doesn't object to it, we can then move on to consider possible extensions.


 * Above, you said that the minor change edit was "largely unnecessary." None of it is necessary, just as spelling corrections aren't "necessary." But that's not the question. Do you oppose restoring that edit? All of it or just some of it? --Abd 03:46, 10 September 2010 (UTC)


 * For reference, here are Wikipedia policies on the issue of handling contributions by banned editors: Edits by and on behalf of banned editors and Enforcement by reverting. Wikipedia is often dealing with articles with strict requirements. Wikiversity, generally, has far looser requirements for content. (Quite a bit looser than Wikibooks, Thenub, please be aware of the differences.) --Abd 04:04, 10 September 2010 (UTC)


 * In reference to the first paragraph I think you have missed my transition to where I was speaking about a separate subject. It takes place with "For the second edit..." After which I am no longer discussing self reversion, as it is not mentioned in the second edit.  I mention bad blocks because you were discussing blocks being improper.
 * In reference to the second paragraph: I meant claims about improper blocking should be handled on a case by case basis as they are rare. I did not intend to imply all blocks should be handled on a case by case basis. The communication seems to have gone so astray I do not see how to comment further. Thenub314 05:28, 10 September 2010 (UTC)

Okay, let's start over. Basic rule of deliberation, one issue at a time! --Abd 14:31, 10 September 2010 (UTC)

The edit to 'What is disruptive?'
[*copy editing in What is disruptive? < Is there any objection to the restoration of this edit?]I'm being told that changes to the proposed policy should be discussed first. The edit to this section was the kind of minor editing that I've seen take place commonly without discussion. For minor changes, it is customary to make the edit, and if someone disagrees with it entirely, they revert. If someone disagrees with part of it, they revert part and accept part. And so the section, over time, becomes better-written, in theory. But this entire change, consisting of three changes to English usage and one adding a clearly appropriate link, was reverted, and without any reason being given. I intend to revert that back (just that section) if there is no objection to the content changes. What I saw above was not objection, but general comment on the section, now that I reread it, and perhaps a disagreement with it. But that's a disagreement with the standing section, not with my edit to it. And I disagree with the objection. Blocking policy should cover (directly or by reference) process for appealing blocks. That isn't about "bad blocks," it should be just as clear and just as open to appeal "good blocks." --Abd 14:31, 10 September 2010 (UTC)

Two additions
1. This section from Wikisource has sound advice. I recommend a localized version with a necessary warning provided first.

2. Adding a prohibition on blocks when it comes to disputing warnings or removing warnings from talk pages. As Wikipedia says, removing a warning is acknowledgment of the warning, and a warning is no a black mark to tarnish others with. The talk page is the person's castle and they have the right to remove what they want. Additionally, to not allow blocks to be extended from someone disputing the block. Too often admin make bad warnings to bait someone into being blocked, then taunt the user in a manner to indef them. It is not what blocks are to be used for. Ottava Rima (talk) 01:43, 13 January 2011 (UTC)


 * This version and the additions take my suggestions above in a way I am comfortable with and I would support this wording and language. Ottava Rima (talk) 02:49, 13 January 2011 (UTC)

That version isn't bad, but I do have some quibbles, not important enough to address immediately. It's correct that removing a warning is acknowledging it, and perhaps we should have some detail like this, because Wikiversity custodians are not necessarily familiar with general wiki traditions, some of us have little other-wiki sysop experience or haven't followed these issues elsewhere. Generally, a single admin shouldn't increase a block for the user responding badly to being blocked. It's generally better to let another custodian handle it, if possible, so that the matter doesn't become a personal conflict. Hardly anyone takes being blocked well. We should make it very clear that custodians, if they feel it necessary to block, should rigorously avoid any incivility. I've often considered that every block should be accompanied by an apology. "I'm sorry, but my duty to protect the community and project leads me to find it necessary to block you for 24 hours to prevent a repeat of what I warned you about. If you have any questions, I'll be watching your Talk page, and you may also email me. If you will agree to ... I can lift this block immediately. Otherwise, here is how you can appeal this block...."

I was amazed to see some admins on Wikipedia block users, writing obscenities and various unnecessary accusations. And nobody did anything about it, usually.

In addition, a user should not be blocked for angrily rejecting a warning or insulting the admin. A warning admin, seeing that, should simply affirm that the warning stands. Blocking should only happen if the warning is disregarded by acting contrary to it. Again, hardly anyone likes to be told that they can't do what they want to do. It's just a normal response. --Abd 03:52, 13 January 2011 (UTC)


 * I think we can take this a step further. Not exactly "don't warn" as has been previously suggested/discussed before, but suggest ways to behave/act appropriately or request they behave/act/do something a certain way in a firm "tone". That can help to avoid the whole not liking to be told what not to do business. However some people also don't like being told what to do either and there is still the potential to come off sounding bossy, rude, or uncivil too. -- dark lama  04:15, 13 January 2011 (UTC)
 * Warnings only serve as tools to hurt someone. Instead, an admin should say "could you work on ____ articles for a while and ignore this topic so the dispute cools down?" or "could you try to look at some positives in their position and point them out to help establish a compromise?" Similar things are equally necessary. Ottava Rima (talk) 04:26, 13 January 2011 (UTC)
 * In theory, "don't warn" is a good idea, but doesn't seem to work in practice. Sure warnings can leave people feeling hurt, but at the same time people can feel hurt when they think not enough was done or a person wasn't stern enough before blocking. From what I have experienced, people who say warning shouldn't be done usually seem to end up being the first to say you should of warned first after a block was done, and/or seem surprised when a block still seems to be necessary. I think "don't warn" cannot work unless people are willing to define when enough has been done, or are willing to accept a custodian's own judgment as to when they have done enough to try to address the issue before blocking. -- dark lama  04:51, 13 January 2011 (UTC)
 * I think warnings serve a purpose, but not as first contact (the Wikipedia way). Care should be taken with the design of a warning template, which can go along way in making that the warning comes across as help rather than a ticking bomb.
 * Deleting a received warning of course makes perfect sense, warnings are not intended to attract attention from visitors. (The same goes for block messages, by the way, IMHO.) Regards, Guido den Broeder 13:37, 13 January 2011 (UTC)
 * They should replace "warnings" with "jeopardy reports" that advise someone how close they are to falling off the edge of a cliff. Or (in the US) they could call it Cliff's Notes.  —Caprice 14:03, 13 January 2011 (UTC)
 * My concern with warnings is that people use warnings instead of figuring out how to fix a problem. Instead of compromise, people seek warn. Instead of finding solutions, people want to remove those who point out a problem. Remove the tools of war and easy domination, and people will have to respect another user. Warnings are like putting templates on articles - they show that you recognize a problem, but they are like giving up in attempting to fix it. We need fixers, not labellers. Ottava Rima (talk) 15:11, 13 January 2011 (UTC)
 * We can provide the means by which a "better" way can be used, but unfortunately any approach can be used to dominate and war with. The problem isn't usually the means, but the people. This issue is like gun control, some people say guns are bad, and some people point out people kill people not guns. -- dark  lama  16:46, 13 January 2011 (UTC)
 * "any approach can be used to dominate and war with" Then the easiest solution is to remove and ban the warriors who lack educational/academic backgrounds and ensure that they stop corrupting an educational website with their ways. Ottava Rima (talk) 16:59, 13 January 2011 (UTC)
 * Your comment seems to suggest a bias against non-educated people, as if only non-educated people can or will start wars and try to dominate over other people. Going down that road would be to exclude the very people Wikiversity aims to help, as Wikiversity is a website to educate people not just for the already educated. To use my analogy again, you seem to be suggesting that only people that don't know how to use a gun kill people with guns. Police officers are trained to use guns and kill people with their guns too. Sometimes the easiest solution is not the most reasonable solution. I suggest the first step, as Guido den Broeder hinted above, should be to redesign/replace any warning templates we have. The second step might be to require that our templates use an assertive tone rather than an (passive-)aggressive tone and offer reasonable alternatives as a means of compromise. -- dark lama  17:18, 13 January 2011 (UTC)
 * "non-educated people"? Maybe non-educators. An educator is a job/occupation with the primary desire to teach. Very few people have been involved in trying to help others learn and that has been a trend since 2 years ago. I believe that those who have a background in education should be the only ones who run an educational community, and many have pointed out that giving power to those without such background has led to many of our problems. Ottava Rima (talk) 18:32, 13 January 2011 (UTC)

I think "do warn" AND "don't warn" must both be avoided. Hopefully that sums up my position a bit better. Maybe replace "warnings" with "cautions" that informs a person what is at issue, what they should/can do instead, and how continuing down this road can be hazardous for their health. To use a silly example to illustrate what I mean:

-- dark lama  16:38, 13 January 2011 (UTC)

Block evasion
In my latest edit, I have removed this section because I felt that due to its length, extreme wordiness, excessive use of commas and ambiguous interpretation that it should not belong on this page, at least not yet, until some significant rewording has been made. I'm also going to cite my bot-account-example once again, because this question never seems to get addressed. TeleComNasSprVen 19:43, 8 April 2011 (UTC)


 * The section TCNSV removed (under the title of this Talk section):
 * Users are expected to respect blocks, following due process for appeals, and block evasion is its own offense, causing work for custodians. Users are thus subject to extended blocks if they evade a block by editing without logging in or using an alternate account, and IP blocks and range blocks are often used to interdict this. Sometimes editors have been able to edit in spite of being blocked, without negative consequence, if they take steps to insure that their edits are not disruptive and do not complicate block enforcement; self-reversion is one process that has been used for this.


 * This describes, minimally, actual practice, and some of the reason for it. What part of this is not actual practice? If the wording is poor, surely it could be improved, and if it is too wordy, it can be summarized.


 * TCNSV's diff about his bot example is misleading. First of all, it's an edit to my Talk page, it's not easy to follow, and he's making some different point with it, but he summarized with Remember, blocks are against the account, and bans are against the user. That is based on certain exceptions. Generally, when an account is blocked, the user is banned, effectively, for the duration of the block, and block evasion is treated as an offense, with the block frequently being extended as a result of any evasion.


 * This argument by TCNSV is an example of using the existence of exceptions to, then, radically reinterpret the rule. The exceptions are, not necessarily exclusively, bot accounts (which may be blocked without creating any kind of sanction against the user), obsolete accounts blocked to prevent possible compromise or abuse, with no intention of sactioning the current account of the user. Sock accounts are also sometimes blocked without intention to sanction the main account. My account User:Richard P. Feynman was recently blocked for edits considered disruptive, and the sysop did not extend this to my main account (though the autoblock did effectively block me for a time, I simply used a different access), and the same was done with one or two of JWS' sock accounts. But if I were blocked, you can be sure that, if I used User:Richard P. Feynman to evade the block, or edited by IP, in any way considered disruptive -- and the bare fact of editing might be considered disruptive, self-reversion is a way for a blocked editor to possibly avoid a conclusion of "disruption" -- the IP or other account(s) would be blocked.


 * TCNSV is correct in this: blocks are against accounts, they are technical features of the software, which allows blocking specified accounts or IP addresses or ranges. But if a block is used to protect the wiki against disruption by a user, it is the user who is effectively banned for the time of the block. That's spelled out on Wikipedia, I believe, but it's just common sense. In the exceptions, the user is not considered generally disruptive, but a bot is a problem, or there is possible abuse of a sock account, by the user or someone else. To a user blocked because a custodian thinks that they are being disruptive, the difference is academic. --Abd 20:44, 8 April 2011 (UTC)


 * Blocks have always been used as a technical function to protect wikis against disruption by a certain account. That's how it is defined. The only time when "a block is used to protect the wiki against disruption by a user" is when it merges into or explicitly becomes a ban, instated by the blocking sysop. Bans protect the wiki against a certain user. TeleComNasSprVen 22:01, 8 April 2011 (UTC)
 * I assure you, TCNSV, almost every block performed by an administrator is intended to ban the user involved from editing, for the block period, to "protect the wiki against [the] user," yet the word "ban" is not used. What you describe as "the only time," implying that it's some exception, is actually the normal and, by far, most common case. An individual admin will not use the word "ban," normally. One did with me, setting a "topic ban," and enforcing it with a block, without community confirmation, and, in fact, got himself desysopped over it. Topic bans are not decided by admins, except for informal ones (long story). --Abd 23:55, 8 April 2011 (UTC)


 * "Oh, but this is just argument for the sake of argument! Let's close and archive this thread and then forget the section previously inserted into the page never existed altogether." TeleComNasSprVen 19:13, 9 April 2011 (UTC)
 * Let me guess, TCNSV, you are collapsing arguments and comments from various discussions on other pages into one, as if all this is about you and me. It's not. It's about the policy. Please discuss the policy and the suggested additions or changes. Thanks. --Abd 19:58, 9 April 2011 (UTC)
 * Why should I? After all, with no other people like custodians etc. monitoring the situation, it's just us hashing out it at Wikiversity talk:Bans and Wikiversity talk:Blocking policy, with no compromise or consensus whatsoever. TeleComNasSprVen 23:42, 9 April 2011 (UTC)
 * You are making a common error, TCNSV, thinking that Wikiversity takes place in real time, and that nobody is watching. Rather, our discussions stand, normally, for many years, and may attract comment for a very long time. Those "discussing" or "monitoring the discussion" include many who may not see this today! The basic cycle runs for maybe a week. As to "no compromises," speak for yourself. Lots of compromises are being made, by both of us. I'll let others judge who is more assiduously seeking consensus, but I assure you, it's my goal, and you are not excluded from 'consensus' unless you exclude yourself. --Abd 00:08, 10 April 2011 (UTC)

Meanwhile, we have the proposed text above. TCNSV has not actually objected to anything specific in that text. I could guess -- and did -- that he's objecting to Users are expected to respect blocks, following due process for appeals, and block evasion is its own offense, causing work for custodians. Does he think this is incorrect? If so, how?

Exceptions exist, but they do not negate the rule. The bot exception is not going to cause confusion to a newcomer, but newcomers often think that if one account is blocked, it's fine to create and edit with another, I've seen this happen many times where the user did not seem to realize that it was any problem at all. And sometimes it's not! If nobody cares, nobody cares! --Abd 20:14, 9 April 2011 (UTC)


 * Your unfounded claims go without evidence. I have stated my reasons for removing the text in both the edit summary and in the header for this section. This is yet another example of applying double-standards to editors: you expect me to read everything that you write, and yet you don't read what I write. Please reread the reasons again and come back when you are going to seriously discuss the policy at hand. TeleComNasSprVen 23:42, 9 April 2011 (UTC)
 * I have read everything you wrote here. I'll read it again before editing the page again. You have not made any clear objection to the specific text I asked about in my last comment, other than comments (previously) that describe something that's preposterous, the idea that an admin blocking a user is not normally expecting the user to abstain from editing, that they have to use the word "ban" to create this expectation. You wrote:


 * The only time when "a block is used to protect the wiki against disruption by a user" is when it merges into or explicitly becomes a ban, instated by the blocking sysop.


 * That is completely divorced from what administrators actually do. Perhaps this is due to your shallow experience on WMF wikis, but, with the odd exceptions that have been noted, every block expects the user to stop editing, and users who continue to edit face sanctions, routinely, and you have even cited a case for this, yourself: User:Wikademia, whose blocks were extended because of socking. The way you have interpreted blocking policy, to divorce it from any kind of banning other than single-account, completely fails to explain this. If it's only an account that is blocked, why then the blocks of socks, and why IP and range blocks? Even though a "ban" had not been declared?


 * On the contrary, "ban" is not a term normally used by any administrator when they block, and blocking admins do not determine bans on Wikipedia. I actually was involved in a WP case that made this clear, because I'd been "topic banned" by an administrator, who believed that he could unilaterally impose a strict ban, it was a page ban, so that he could then block me for any edit to the pages, even if the edits were harmless. When he tried to enforce this, during the RfAr, he almost lost his bit immediately, as an emergency desysop by ArbComm, but this was one very popular administrator, and they waited till the conclusion of the case! You may have confused, in your thinking on this, a block imposed by an administrator pursuant to a ban, either by ArbComm or by the community. A closing administrator will frequently do the block, if it's a site ban that has been decided. --Abd 00:31, 10 April 2011 (UTC)

Reversing a consensus
. This is one of the most difficult wiki problems, and it is an older, more general problem, existing in consensus organizations for a long time. Such organizations frequently start with high levels of personal freedom, and activity is restricted only if a consensus appears that it is harmful. (I'm not defining "consensus" here, but it normally refers to supermajority; however, on WMF wikis that gets really complicated, because supposedly decisions are not made by votes but by arguments. However, if a majority of active participants favor a position, having equal power to act, at the beginning, finding a consensus against it will be unlikely.

If consensus at one time is considered to be established, taking a consensus to overturn it, the community becomes increasingly "conservative," such that it can easily arise that most participants favor some position, but it is rejected because of the former "consensus." However, a result of this situation is a gradual weakening of the organization, because disaffected members may leave. They begin to consider the community decision-making process "impossible."

A more sound position than the persistence of consensus is that if it can be shown that a position no longer has consensus, that position loses its "consensus" status. However, this may not represent a consensus in the reverse direction. The situation is returned to the original one, that what was not prohibited was permitted. (Or what was required is no longer required, because it would not now be required.)

The problem with this is that, of course, continual re-discussion may be needed. Hence human communities develop executive bodies that have the ability to continually re-assess situations.

The practical question here would be that a community consensus has been established to block. However, the community doesn't actually make decisions. Rather there is a discussion, and a close. The closer determines the result. In theory, votes don't count, but arguments do. The closer is not just counting votes to determine a majority (or supermajority). Rather, the closer is, consistent with his or her sense of the community's position, assessing the arguments. The closer may decide contrary to the majority of votes, that's well established, but should do so only in expectation that a broader or deeper discussion would support the closer's position. None of this is easy to codify, to nail down.

What is clear is that an established consensus should be respected, but that doesn't set up an absolute requirement. We need, with blocking policy, to have clear guidance for custodians as to what they cannot do; beyond that, it's discretion, there is no "must" for volunteers, as to *required* action. Even if we have clear policy, though, custodians should have discretion. Bottom line, if a custodian sees an emergency situation, with damage being caused (and sometimes remediation of damage is not possible; a blocked user might, for example, go away and never check the wiki again; incivility can have similar permanent effects), the custodian may, and we could say "should" block to stop the damage. Unblocking is a relatively easily remediable action, normally, if a custodian pays attention to what then happens.

What should not be done by a custodian is to unblock and then leave on a wikibreak, declared or undeclared, leaving others to handle the possible resulting damage. However, as long as there are other custodians active, an unblock is normally relatively harmless, because if damage appears, the user can quickly be blocked again.

I don't think we are ready to declare policy on the point of unblocking after a consensus for a ban. We almost never have such a situation, as to local bans -- though it's happened. The present disagreements are over bans declared elsewhere, particularly at meta. The only thing new about this is that a meta RfC declared a "global ban." Globan bans are not new, nor is meta consensus regarding them new either (as to specific cases). What's new is the use of meta user RfC for it; previously these were only considered to control meta status, and meta was not considered to have any authority over local wikis. An appearance of control existed with the global locks, but those are really over SUL accounts, not detached accounts and not locally supported socks. (I.e., an account only active on one wiki, with no disruption there, and no meta RfC regarding that account and possible harm being done globally by it.) --Abd 19:02, 13 November 2011 (UTC)


 * Overall I find the above logic, unconvincing, and I preferred the previous wording, which allowed for an unblock under community set conditions. Thenub314 19:36, 13 November 2011 (UTC)
 * What are you suggesting? I think we both agree that discussion should happen first. I'm guessing you believe that for a block to be maintained there should be a consensus for maintaining the block within the current discussion, rather than a consensus for unblocking within the current discussion. -- dark lama  20:38, 13 November 2011 (UTC)
 * Perhaps the unclarity is because I did not quote what I'd removed from what Darklama had inserted:
 * However if there is a community consensus for the block, the community must first come to a consensus to unblock, which can include conditions that must be agreed to. This was new.
 * I revised this to:
 * However if there is a community consensus for the block, an individual custodian should not unblock without discussion, and a custodian should never unblock against a maintained community consensus.
 * @Thenub: My change was of Darklama's change to the previous wording, which was, as relates to your comment:
 * The Wikiversity community may decide to conditionally or unconditionally unblock an editor or to deny any unblock request. Custodians are encouraged to start a community discussion about new or continued concerns before acting in apparent contradiction to previous community consensus. Custodians who unilaterally unblock previously blocked users may be seen as ignoring community consensus and thus behaving inconsistent with the Wikiversity mission or scope. Whenever possible any block/unblock discussion should be closed, and the result implemented by, an uninvolved and impartial custodian.
 * That language is okay with me.


 * @Darklama: No, that's not exactly my view, but there is some resemblance. If we could efficiently know true community consensus (or even absolute majority opinion), it would be binding. Given that we don't have methods or procedures for doing that (they are possible, and one of the long-term goals of the Assembly is to suggest and demonstrate --in the future -- possible ways), we must rely upon custodial discretion. Discussions advise custodians, but any given (small-scale) discussion may stray far from true community consensus. Bottom line, policy should reflect what is solidly established, which should be consistent with actual practice -- or either the policy or the practice should be revised! It is probably impossible for policy to cover all situations. Having policies and practices that are incompatible, however, is highly damaging to the security of the community. If there are protections available in policy, users will properly expect them to be real and enforced. A long-standing principle is that blocks are abnormal, allowed only due to necessity, and necessity requires a continued consensus. But efficiency requires that matters not be continually re-argued. The conflict between these principles is handled through custodial discretion, and by the actions of a community of custodians, not through rigid policy. However, in the other direction, policies that protect users should generally be interpreted strictly in favor of users being allowed to edit freely -- as long as there is no ongoing danger.


 * A discussion may show, not a consensus for unblocking, but a lack of consensus in the community for maintaining a block. If the community is apparently divided, should the user stay blocked? or is being unblocked the default? If the discussion is very small-scale, it says little about the real status of the issue in the community, but if there is wider participation, what then? There is no way, without structural changes, to nail this down. --Abd 00:20, 14 November 2011 (UTC)


 * Incredibly ill-formed with bad reasoning. There doesn't seem to be any support for these ideas at all. Ottava Rima (talk) 02:39, 14 November 2011 (UTC)

Given the lack of agreement about the section, I've removed it for now. It may not even be necessary. Perhaps a separate section or page could be considered for dealing with local unblocks for global bans. -- Jtneill - Talk - c 04:23, 14 November 2011 (UTC)


 * Do you understand what has been objected to? The part you removed was not about global bans at all. I don't even understand how you came to the conclusion it was about global bans at all. -- dark lama  10:18, 14 November 2011 (UTC)
 * Perhaps I didn't/don't understand the intent of the section describing proposed policy for reversing consensus about local blocks - feel free to replace/improve if it seems important. The paragraph on this seemed contentious to me based on my reading of the discussion above, hence my suggestion that perhaps it could be left out altogether. I realise that unblocking a local block is not the same as unblocking a user banned globally, hence the suggestion that perhaps a separate section or policy could be useful, based perhaps on what emerges from Community Review/Global bans. -- Jtneill - Talk - c 10:42, 14 November 2011 (UTC)
 * My intention is to help guide people with solutions that have worked like first discuss and address any concerns when considering an unblock because that is what has worked in the past and when people have failed to do this in the past the Wikiversity community has usually been split, discussions have usually been unproductive, and in the mist of it all usually the one thing people could agree on is that discussion should have happened beforehand.
 * I have no intentions or plans right now as far as global bans go, unless waiting, doing nothing, and ignoring it for the most part counts as a plan. -- dark lama  13:09, 14 November 2011 (UTC)
 * Thanks - My sense is that we have the same or similar intentions here. My suggestions also reflect an intention for simplicity, which I also think we share. I've made some further minor copyedits to the unblocking section. Currently, it seems reasonable to me and clearly spells out the need to discuss and find consensus for unblocking first, particularly if blocking was supported by consensus. I think the current unblocking text could be further simplified e.g., there is some repetitive content about the places to discuss unblocking (user talk, RCA, and CR). Perhaps the three current paragraphs could be simplified to two paragraphs. -- Jtneill - Talk - c 13:50, 14 November 2011 (UTC)


 * Just to be clear on my end - I believe that to unblock requires consensus to unblock, not consensus to stay blocked. I believe that changing it to consensus to keep block would lead to unnecessary wikilawyering, disruption, etc. Ottava Rima (talk) 16:22, 14 November 2011 (UTC)

Is consensus required for unblock?
I removed this language from the draft policy: ''Unblocking requires a consensus in favour of it. If there is no consensus as to whether a user should be unblocked, they stay blocked. ''
 * The old position, and default across the WMF wikis, was that maintaining a block requires consensus. Specifically, the lack of full agreement among those holding the block privilege. Otherwise any administrator may unblock. The policy requires consultation, but not a surrender of authority to a blocking custodian, whose decision becomes an irreversible action without support by consensus. If there is consensus for block, then a showing, in the future, of no consensus for block would allow a custodian to unblock. This has all been demonstrated by examples, and the reverse demonstration is only what used to be called a "defacto ban" on Wikipedia, i.e., user was blocked and no administrator is willing to unblock. Overall WMF policy, however, clearly distinguishes such a "defacto ban" from an actual ban, which requires a discussion and consensus for ban.
 * Are we going to need a Community Review to establish clear policy? Maybe it is time, in this and in other areas. I've been thinking that Wikiversity is a safe place to develop educational resources. If users can readily be blocked without clear violations of policy, maintained after warning, it's not safe. --Abd (discuss • contribs) 14:14, 27 March 2014 (UTC)
 * What the policy requires is *consultation*. That does not mean "agreement." That is, a custodian should attempt to ascertain the block rationale before unblocking, and a custodian should be responsible for determining that unblocking will not harm the wiki, more than maintaining the block will.
 * There is a tendency to write policy as normative without regard for actual practice. There is also a tendency to then act without regard for policy, but according to the custodian's own opinion. A mature policy would be created by (1) review of actual practice and (2) discussion of possible norms in the light of actual practice. Hence I will begin to draft a review of actual practice, at User:Abd/Wikiversity blocking practice. Comment on the talk page will be welcome at this stage. If this becomes part of a Community Review, of course, it will be open to editing. --Abd (discuss • contribs) 14:32, 27 March 2014 (UTC)
 * I've been reviewing the history of this page. We should realize that there has been great disruption on Wikiversity caused by lack of a policy on blocking. We have seen recusal failure and accusations of recusal failure. Custodians have been sanctioned for blocking under conditions they considered "emergency," where the custodian immediately consulted the community, unblocks have sometimes been arbitrary and harmful, wheel-warring has been common, and it's been quite a mess.
 * The page was tagged as policy by JWSchmidt, https://en.wikiversity.org/w/index.php?title=Wikiversity:Blocking_policy&diff=581868&oldid=577610 6 July 2010‎], reverted by Darklama, quickly, claiming "no consensus here either."
 * Excepting edits by Jtneill and Darklama, which were clearly accepted by JWSchmidt, there had been no significant changes to the page for a long time. Darklama had edited the page, and he was not reverted by JWSchmidt. JWSchmidt had last edited the page 29 January 2009‎. By marking that page as policy, JWS was implicitly accepting all the intervening changes, including those by Darklama. By only one standard was there "no consensus," which is a wikilawyering standard that a Policy requires a Community Review. (The policy was not followed by some administrators, basically those who resisted the establishment of policy. But they did not attempt to correct the policy, if it was incorrect, and the reason is fairly obvious: the community would not have supported those deviations and, generally, when they became widely known, they were not supported.)
 * JWSchmidt did file an omnibus Community Review, a singularly ineffective way to handle policy consensus. It was mentioned in a section of his general complaint about custodian practices here, See this page, section on Community Review. The review itself was way over-broad, Community Review/Problematic actions.
 * Reading these discussions, in that period, JWS seems obsessed. Yet not so long before, JWS could have been a bureaucrat simply by saying "Yes," see . His comments in that period seemed moderate, and, we should remember, he was one of the founders of Wikiversity.
 * His first edit in our wiki history, 17 March, 2004. Notice that this page started on Wikibooks. (The first edit to Main Page was not until 15 August 2006. Looking at the early edits to that page, I see User:Cormaggio, still a crat but inactive, and User:JWSchmidt.


 * What happened to him? It's a convoluted story, complex. However, a major part of it is that the community did not make the development of policy a priority. Instead, it accumulated custodians. Custodians tend to be more active than other users, and especially around central organizational pages. And custodians develop a particular point of view, which can differ from the general view of the community, which includes many users who never pay attention to the "drama" of the organizational pages, as long as it isn't their resource up for deletion, or their account being blocked, or threatened with same.


 * Administrators (and others who are centrally active) generally believe that they have special rights to determine the welfare of the community. It's not unique to wikis, this is a common trait in human organizations, I call it the "dictatorship of the active." It's not that it's wrong, it is that the active can develop a special point of view that is not true consensus.


 * I think JWS never dreamed that the community would accept what happened to him, and when it did -- it did not stop it -- he completely lost his balance. Imagine what it is like to have a 'crat tell you to "go fuck yourself," and that 'crat then blocks you, not just once, but again and again. See the JWS block log.


 * It is not that there was no basis for the blocks. JWS also complained bitterly about me, because I suggested a topic ban, with clauses allowing him to do pretty much whatever he wanted, but with a mentor approval, a mentor of his choice (but acceptable to the community as well.) I think he'd go for that, now, but am not proposing it yet, because Wikiversity is not safe.


 * In that block log, I see, that aside from a dumb self-block at the beginning, JWS, was
 * Indef blocked by SB_Johnny, 19 September 2008. Unblock by same after discussion.
 * Blocked by Mu301, 15 January, 2009. Unblocked by Countrymike after 5 days for lack of discussion by blocking admin.
 * Blocked by Mu301, for two weeks, 15 January 2009. Expired.
 * Blocked by Adambro for one day, 26 March 2010.
 * Blocked by Darklama for one day, 7 July 2010 (disruption by converting proposal into a policy without consensus), Unblocked by Diego Grez.
 * The "disruption" was the edit to this page, calling it a policy. In the discussion above, JWS asserts that custodians have a conflict of interest when editing this policy. I'll confirm that below. Darklama not only reverted JWS, but blocked him. And this was tolerated.
 * Blocked again by Adambro, 11 July 2010, unblocked by Ottava Rima. "no consensus for block."
 * Blocked again by Darklama, 6 March 2011, one month
 * Reblocked by SB Johnny, 10:25, 20 March 201, expiry 6 April 2011, to shut off talk page access
 * Reblocked by Mu301, 13:29, 20 March 2011, to extend block to indef.
 * Obviously with no further disruption at that point!
 * Reblocked by SB Johnny 26 March 2011 to restore talk page access.
 * Unblocked by Geoffry Plourde, 9 April 2011 "due to expiration of original time and per comments on user talk page")
 * Reblocked by SB Johnny 16 April 2011, indef.

So, JWS was blocked for making an edit that reasonably did reflect defacto consensus. I was harassed and threatened with block for editing the page (way back). Yet we have custodians editing this page with no restraint, and without seeking consensus first. There is very obviously a double standard being applied.

There is much, much more to this story than I've detailed here, but a quick summary of what happened to JWSchmidt was that his high hopes for Wikiversity were dashed. Wikiversity could, in my opinion, still be what he hoped for.

Now, why do I bring up all this? It is not to condemn custodians for doing what may be reasonably common practice, or at least not prohibited by policy. It is to examine what happens when we don't have policy! In some cases, we have policy, but who enforces policy? Routinely, it's custodians. Hence in areas where custodians have extra power, it is crucial that this be restrained by policy. Or else we get continued disruption and confusion, it's human nature.

I wrote Recusal, in this version: permanent link. It was edited with apparent approval by Adambro. It was gutted by Darklama.. Eventually, I noticed this and reverted back to Adambro's version. Darklama reverted. Note that Darklama and some other custodians routinely violated even the gutted version. His version was utterly impractical, destined to be violated by any custodian who cares about the wiki. I suspect that, at the time, I was being attacked for, or had been attacked for, say, blocking Ottava Rima. I had blocked under the emergency provisions of the proposed policy (and had immediately taken the block to the community). Darklama, then, was writing policy to prohibit me from acting. Yet it can arise, as the proposal still notes, that nobody is available who doesn't have a potential conflict of interest. This occurs especially when a user has often been blocked, such as JWSchmidt. Darklama made recusal requirements apparently strict: "When people can agree quick action is needed, you may take temporary action to address other people's concerns with the understanding that the decision may be reversed later and anyone may undo the action. You must not decide or act otherwise when expected to recuse." Basically, Darklama is requiring consensus ("people can agree") before action. He did not himself wait for that, frequently.

Police, in general, are empowered to act promptly when the public welfare is threatened. However, police do not make final decisions, judges and juries do. Police are only allowed, on their own, a limited freedom to confine and restrain others, pending judgment by other than the police.

The absence of a clear policy has long damaged Wikiversity, and we can see that this absence is not merely the result of lack of interest, it is actively maintained by some in the community, and that is still going on. Indeed, we see these users show up after long absence to again protect policy from alteration in any way that would effectively restrain them.

I have two basic choices here, given what I'm seeing: give up and leave Wikiversity, or begin to draft a Community Review. I'm opting for the latter, but will do it in my user space and will use the Assembly process for that. It will take time. Anyone may participate in the Assembly process, subject to clerking (and if anyone disagrees with a clerk, they are free to start their own parallel process; presumably when those who participate assess it as ready, a regular Community Review will be started, and multiple sources may be merged or be mutually or unilaterally linked).

We cannot allow custodians to unilaterally determine (or, more to the point, block) policy that restrains custodians. --Abd (discuss • contribs) 22:19, 20 April 2014 (UTC)

Spamming
Addition proposed by User:Goldenburg111 on Community page:
 * 5. Spamming, especially accounts do userpage spamming, as they have no purpose to contribute positively.

(14:26, 19 April 2014‎ by Goldenburg111)


 * That is what usually goes around, even through Wikimedia Projects. And it pretty much happens here. --Goldenburg111 19:44, 19 April 2014 (UTC)
 * We do see user page spamming, but "spam" still needs definition. There are many obvious cases, and Goldenburg is thinking of those. If a single edit is not spam, is the same edit repeated on many wikis therefore spam? Why? The issue is an assumption of motive. That assumption usually works. That is, most examples of repeated content are, indeed, what we readily recognize as spam. However, it's not necessarily that simple. For example, we have seen user pages created recently with gibberish followed by some message that looks like spam, like "Tramodol." Except there is no web site link. I don't think this is spam at all, it's vandalism or possibly the sending of coded messages. It's "vandalism."
 * There are many examples of userpage spam that is likely to be actual spam: some content describing the user, and then a link to a web site selling something. Spam is disruptive. We all agree that linkspam, with no redeeming purpose, is disruptive and should be blocked, and the use of the spam blacklist may be appropriate. But suppose a user registers an account and discloses that they work for a web site, and the site is named.
 * Is that "spam"?
 * If we are going to make blocking and deletion be based on "spam," we need a clear definition. The lack of that has allowed "spam" to be extended to "anything promotional" and it can be extended beyond that, to "anything POV and repeated."
 * The paragraph above includes 'Users who vandalize and spam should be blocked immediately before considering anything else.'. How does this addition improve on that?  -- Dave Braunschweig (discuss • contribs) 19:48, 19 April 2014 (UTC)
 * So we need a definition of "vandalize" and "spam." We all may think we know what these are, but Goldenburg, of all people, should know that there are cases where what seems like clear vandalism to one user may seem like "educational activity" to another. Hence, when we block for vandalism or spam on Wikiversity, we will often leave Talk page access open, unless it is abused. I.e, if a user, say, links to a web site on a mainspace resource, and it seems to a custodian to be irrelevant, spam, intended to promote the web site, then the custodian might possibly block the user (we actually are unlikely to do that unless the user insists, i.e. repeats the edit after warning and without discussion.)
 * My own practice, when I blocked for other than blatant, in-your-face vandalism and spam, was to warn, block if repeated, and invite discussion on the users' talk page. True spammers, it seems, never do this. They don't discuss, they don't appeal a block. True spam doesn't have an engaged user behind it. The name of the game is to get as many incoming links out there as possible, as cheaply as possible. The accounts are throw-away.
 * So, ideally, block templates should provide for those contingencies, so that it is very little work for a custodian to handle vandalism and spam while still allowing good-faith users to amend or explain behavior.
 * If spam or vandalism is blatant, if it is repeated across pages, here, or cross-wiki (and I frequently check), we don't create user talk pages just to warn against what the user would already know is inappropriate and unwanted. We may still leave talk page access open. We risk little by doing that. If it was user page spam, our practice is to delete the page, usually, but we could consider blanking as an option, if the case is at all marginal. When the user page is actually being vandalized, as happened recently, we delete the page to leave it open for the real user. --Abd (discuss • contribs) 20:30, 19 April 2014 (UTC)
 * Thanks Abd, : I think it is best to have it separate, as of vandalism and spam, and it needs to be more specified. So vandalism and spam can be each one point. --Goldenburg111 20:37, 19 April 2014 (UTC)