Jump to content

Wikipedia talk:Five-block rule

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Examples that seem to make this policy necessary:

[1] [2] [3] [4]


Support

[edit]

This came out of a discussion out of groups that seem to specialize in wide ranging, low-level edit wars. Needless to say, despite the fact that some of the users ARE somewhat useful when it comes to editing, the sheer amount of disruption and hand-holding needed to keep things from breaking down just does not make it a net positive for the project. If someone has 5 separate blocks for the same type of offense, it's pretty obvious that they're not going to be abiding by our rules. Rather then continue to try to mediate and protectthese problem users, admins should be encouraged to eliminate those whose disruption outweighs their benefit SirFozzie 01:20, 28 August 2007 (UTC)[reply]

And luckily, if they *really* belong, the ban can be challenged. MessedRocker (talk) 08:21, 29 August 2007 (UTC)[reply]
+1. Sounds like a good idea to me. People shouldn't have to waste their time on offenders that have no intention of changing their behaviour. A five-block rule seems like a fair (or even lenient) way to differentiate between such cases. --Seans Potato Business 16:32, 6 November 2007 (UTC)[reply]

Thoughts

[edit]

In principle the core idea is a good one. Wikipedia isn't a playpen, and much admin time and user time goes into unproductive work due to the impact of a number of people who refuse to get the point (WP:IDIDNTHEARTHAT). I'm happy to have a limit on how much wthe community (and readers) should bear the brunt of this before one says "You are not indispensible to the project" to an editor.

I am worried about a few of the ways it works. I'd feel a lot more comfortable if these were addressed, to reduce the risk of miscarriage of fairness. Changes in summary:

  1. Blocked 5 times in a given time period (6 months? 12?) by multiple admins for the same issue.
    Rationale - ensures it's not just one person's view; if its genuine any admin should be able to get another eadmin to agree to block. Limited time period means that someone who does the occasional item (one attack every few months) has a way to have "old" blocks "fall off" the scale so to speak, which is fair. if its old, not so important; this is more to catch intense habitual repeaters.
  2. Disagree with indef blocks. Be happier to see it authorizing longer blocks than admins usually use, of several months.
    Rationale - Reluctant to have as policy a page that mandates indef blocks like this (subject to exception) as the norm, although I can see the rationale. Too risky of "not enough real cause". That's my real concern I suppose.
  3. The real problem is not so much repeated reuse of same issue, but inefficient handling of edit warriors generally. A lot of folks dont get sanctioned when they probably should be, or get to repeat when WP:AGF should run low, and so on. Admins as a whole seem to be either very aggressive, or very unsure, about how hard to "hit" problem editors........

Overall, core idea is reasonable, actual view reserved until I see more discussion of the detail. FT2 (Talk | email) 01:33, 28 August 2007 (UTC)[reply]

To point one, I concur, it should be more than one admin making the blocks of the user in question... if not the block ing admin his/herself should be taking a step back and asking for help. As for the time period I thought i'd been clear... there isn't really a deadline, but blocks with gaps of more than a year in between, the earlier offenses dont count.
To point two, the problem is 1 month, 6 months, 31 months, these types of users are the ones who wont care about block length because their behavior does not change, when time expires, they'll be back to the same game again. Or they'll be creating sockpuppets with which to avoid their block.  ALKIVAR 01:39, 28 August 2007 (UTC)[reply]
Yes, but you're creating a formal statement that effectively and automatically judges they wont ever change and therefore with no review they can be indef blocked, after any 5th event. And over a long time frame -- if there can be a year between events, that's 5 blocks in up to 4 years, if taken literally. If it's going to be automatic, as suggested, then it's unsafe to also assume they'll never change. The problem is this policy is catching not just "these kinds of users" (the ones that wont change) but others who for whatever reason get multiple similar blocks in a long time-frame. A few months prescribed block and a known period to have acquired 5 blocks in, is safer, if this is to be fairly standard and automatic consequence. FT2 (Talk | email) 01:46, 28 August 2007 (UTC)[reply]

Proposed reword to fix this:

"If a user has been blocked 5 or more times within [a year | 18 months | period of time] by different administrators for identical categories of infractions..."

"The next block should be placed for an indefinite length."

[edit]

Perhaps that should be "can". I don't want to tie my hands with a block that is hard to undo (1 admin can't do it). Voice-of-All 01:40, 28 August 2007 (UTC)[reply]

Also, can the blocking admin reverse himself and unblock (I know it's semantics, but it says 'lone admin')? Voice-of-All 01:41, 28 August 2007 (UTC)[reply]
Reversing yourself shouldnt need explicit permission imo... I can officially state that if need be. Can vs Should... yeah i'll go ahead and make that change now.  ALKIVAR 01:43, 28 August 2007 (UTC)[reply]
"Should" -> "Can" addresses one of my main concerns too. Thanks. FT2 (Talk | email) 01:52, 28 August 2007 (UTC)[reply]

Overturning

[edit]

At present any admin can overturn another admin's block. Hence "request for unblock" for example, or unblock-l, neither of which (by communal agreement) require consent of the blocking admin. This often works as a safeguard against misuse. In general we trust administrators not to overturn without good cause. In this case the block is based on a pattern, not one incident, and so it is a good thing to be more explicit.

Also do we really want to make arbcom the only recourse? Over-stringent. Surely usual unblock processes are adequate and better suited to handle objections. Per WP:BLOCK:

"If you wish to contest a block, see Wikipedia:Appealing a block for further instructions. Except in cases of unambiguous error, admins should not undo other admins' blocks without prior discussion; see below."

This is probably all that's needed. Proposed reword of criteria:

  • Except in cases of unambiguous error, admins should not undo other admins blocks under the 5 Block Rule without prior discussion (although the blocking admin can reverse their own block).
  • A user blocked under the 5 Block Rule may appeal via any of the methods applicable to blocks in general. See: Wikipedia:Appealing a block.

FT2 (Talk | email) 11:10, 28 August 2007 (UTC)[reply]

It was my intention to make it impossible to shop for an unblock. As it stands now its possible to find a softer admin unfamiliar with the situation who is willing to reduce or remove a block from a user regardless of conduct simply because said blocked user promises to behave (shortly before reverting back to the behavior they were blocked for anyways).
Forcing arbcom to give a simple yay/nay of support for an unblock wouldn't require a whole lot more work from them (it wouldn't clog up the already extensive case log) and it would work as a strong deterrent for unblock shopping.  ALKIVAR 00:26, 29 August 2007 (UTC)[reply]
Understand however disagree with the idea. "Someone else might not understand hence make shopping impossible" is likely to include unfairness; then arbcom either must put in time to deal with it (weeks delay then evidence etc) or its stalled.
I'd much rather refer it to usual unblock process, with a note that this is a serious block, and admins proposing to unblock should make sure there is genuinely good cause and set appropriate conditions prior to unblocking. Breach is then grounds for instant reblock with a note "reblocked after 5BR + final chance". That one will probably be perm without arbcom's incvolvement.
But the draconian nature will catch some people unfairly. If that's genuinely to be considered as okay now, then I'd want to see explicit discussion and consensus first on that point, because we don't have any rule of that kind so far anywhere. FT2 (Talk | email) 00:53, 30 August 2007 (UTC)[reply]

A tempting idea, but a warning template would be better

[edit]

This idea is tempting, but I think that in practice it could serve to inflame difficult situations rather than restoring order.

First, we do already have a practice of escalating the length of blocks, and that usually works quite well. There are judgment calls to make, but an insufficiently short block for a persistent offender may lead to a review at WP:ANI, and likewise with an excessively long block. There are checks nad balances in place which help these things work well.

Secondly, I'd be afraid that a five-block rule could have all sorts of unintended side-effects. One would be to make anyone blocked for a minor offence fear that this was a step on the way to a long block, so even a very short block could be perceived as being a step on a ladder towards severe punishment, rather than the small preventive measure intended.

Thirdly, this reminds me of the Three strikes laws enacted in some U.S. states, which have had some controversial results, including the application of severe punishments for relatively minor offenders. Wikipedia can't simply lock people away, and something perceived as unfair in usually going to lead the continued presence of mightily aggrieved editors.

Finally, and most importantly, a five-block rules can appear to be punitive rather than preventive, which undermines the whole ethos of WP:BLOCK.

However, I think that a five-blocks warning template would be a good idea, to remind an editor that they appear to be getting into trouble too often I would suggest that a 5-block warning:

  • should be dated, so that checks could be made on recently-applied instances, drawing in otherwise non-involved admins for a review
  • should only be applied by an admin, and only after that admin has confirmed that there were indeed five previous blocks for the same offence within the specified period, and that none of those blocks was overturned
  • should never be applied by an admin who has previously blocked the editor being warned. This would reduce or remove any suspicion that an admin perceived to be hostile has set out to "frame" the offending editor. (I have never seen an admin try to do that, but I have seen overwrought editors who fear that they are being targeted)

Such a warning template seems to me to be potentially helpful in bringing a difficult situation to an agreed solution. --BrownHairedGirl (talk) • (contribs) 16:56, 28 August 2007 (UTC)[reply]

What about guidance on WP:BLOCK simply clarifying blocks escalate for repeat performance. At present there are far to many cases of 4 x 24 hr block rather than say 24 hr, 24 hr, 48 hr, 1 week ... ? Thsi would be deterrent and protection and less able to be abused maybe. FT2 (Talk | email) 23:15, 28 August 2007 (UTC)[reply]
Sorry to be so contrary... but to respond to your first point no block escalation DOES NOT WORK. In practice it is much more common to see multiple 24hr blocks for repeat offenses because no one bothers to look at the block log or talk to prior blocking admins. There is too much WP:AGF assumed. Once someone has proven there to be no good faith behind their behavior AGF can and SHOULD go out the window.
In response to your second point, significant gaps between offenses nullify the earlier offense. If you've gotten 5 blocks in 3-12 months... your clearly not following the rules. They also have to be similar offenses dished out by more than one admin as well, so as to prevent someone being forced out the door by 1 angry admin. The fact that multiple admins must see similar offenses occur is plenty of oversight that "minor offenses" wont be treated excessively.
Thirdly, Wikipedia is not a democracy, we most certainly can lock repeat offenders away from editing, and we do so ON A REGULAR BASIS. It's called a community ban, and we've got a process for appealing those too. This proposal has an identical appeal process... "Take it to arbcom".
Lastly, A simple block notice doesn't do a damn thing. Read this user talk page and tell me block notices work.  ALKIVAR 00:39, 29 August 2007 (UTC)[reply]
I do understand the reasons underlying this well-intentioned proposal, but the comments here don't persuade me that this is the best way to resolve the problem.
Some editors pay no attention to block notices, but others definitely do, so the example cited here is just one example among a range of possibilities. If the escalating block principle isn't being used, then more admins should be reminded to use it. However, the 5-block rule proposal goes too far the other way, replacing an under-used principle of graduated penalties with an over-rigid system which could well have unintended outcomes.
The fundamental problem with this proposal is that it creates inflexibility, and would treat in the same way some very different cases. A generally civil editor who has occasional lapses into 3RR over a 6month period would be treated the same as someone who goes on a rampage of edit-warring until-blocked, and that doesn't make sense. Much better to allow admins to exercise discretion, and to refer to WP:ANI if a review is needed. --BrownHairedGirl (talk) • (contribs) 17:43, 29 August 2007 (UTC)[reply]
The problem isn't that block escalation doesn't work. It's much more that admins don't systematically use it. I concur with Alkivar over the usefulness of a "5 strikes" rule of some kind, and with BrownHairedGirl over the inflexibility and related issues of this way of setting it up. An attitude that repeat blocks (and repeat occurances of bad conduct) get treated seriously, is good. But the inflexibility in this does not say to me that we'll be stopping trolls from wasting time appealing or finding softer admins. It says to me that we'll be blocking people who we possibly shouldn't be, with little realistic recourse (wait a few weeks, then take arbcom time up).
The element of "don't trust admins" is really what bothers me most. Forbid unilateral unblocking - fine. Require discussion - fine. But the inflexibility and distrust that even multiple admins can't unblock, it's got to go to arbcom and only arbcom, that sounds wrong in a proposal. If the concern is we don't trust multiple admins, but only arbcom, to make long term block/unblock judgements, then we have a real problem.
If admin commonsense and block use is the issue, that needs addressing first. Compensating for bad block judgement or light use, by this method... you know? The high level of inflexibility (as BHG calls it) or rules over judgement, is misplaced. FT2 (Talk | email) 01:08, 30 August 2007 (UTC)[reply]

Why do we need a rule?

[edit]

Admins should already take into consideration previous block history in cases of problematic editors. We can already block people for increasing lengths when they are repeat offenders. So I don't see the need for some new "rule" here. This purpose seems like it would be well served by just adding more advice to Wikipedia:Blocking_policy#Duration_of_blocks. Let's not create yet another guideline/policy page unless we have good reason. Friday (talk) 15:54, 30 August 2007 (UTC)[reply]

Hmm, I'm tempted to agree. We have enough rules as is. This could be served with an extra line to the main blocking page. Voice-of-All 03:05, 1 September 2007 (UTC)[reply]
If this rule is meant to mean a mandatory indef block after 5 block then I think that is harmful in that it reduces admin discretion. If it simply means to allow for that, then we really already have the option. I see no need to codify this, if we have been handling this without a policy(as I believe we have been) then creating a policy is not a good idea. 1 != 2 02:01, 17 October 2007 (UTC)[reply]
I think it's because folks look at each block as its own creature.. In some examples, if I blocked someone for 72 hours for editwarring and they had a history of such things, I would be accused of drifting into "Punish instead of prevent" SirFozzie 03:19, 17 October 2007 (UTC)[reply]
I concur that we don't really need a rule there. It suffices to add a footnote to WP:BLOCK that blocks can sometimes be longer for repeat offences. Which, well, they can. A strict "five strikes you're out" isn't necessary, and Wikipedia has its share of "well-meaning hotheads" who are generally productive but manage to land themselves a block per month since they're, well, hotheads. >Radiant< 12:45, 11 November 2007 (UTC)[reply]