Wikipedia talk:Rehabilitation of offenders

This policy would only apply to users (not IPs), right? Griffinofwales (talk) 01:26, 6 July 2009 (UTC)

I hadn't really thought about it. It might as well apply to IPs, it often isn't even the same offender then, so they certainly shouldn't have old blocks held against them. --Tango (talk) 01:59, 6 July 2009 (UTC)

Yes, but old blocks are a deciding factor in new blocks with IPs because IPs are not allowed to banned from editing. If IP editors don't want blocks being held against them, they can always create an account. Griffinofwales (talk) 02:09, 6 July 2009 (UTC)

A 6 month old block shouldn't be a deciding factor for blocking an IP because there is a very good chance (especially if there are no more recent blocks) that it is a different person. --Tango (talk) 03:20, 6 July 2009 (UTC)

Checkusers and ARBCOM should be allowed access, with a possible creation of a new title and/or 'crats being allowed access also. Griffinofwales (talk) 01:31, 6 July 2009 (UTC)

Most Arbs are checkusers, aren't they? They certainly all have easy access to checkusers if they need to ask about a user. Checking spent blocks shouldn't be a routine thing, they should be ignored unless they is a specific reason for needing the check them. I don't see why crats would need it. --Tango (talk) 01:59, 6 July 2009 (UTC)

If a case went to ARBCOM, the arbs might want to check the block history. Maybe not 'crats though, it was just an idea. Overall, this sounds like a good idea, but it should not become policy because there are so many exceptions to the rule. Griffinofwales (talk) 02:09, 6 July 2009 (UTC)

The whole point is that they shouldn't be using old blocks to make their decisions, they should consider things that are relevant to the present day. --Tango (talk) 03:00, 6 July 2009 (UTC)

I think the best way to achieve that goal is to educate editors/admins, rather than restrict the information available to them. I used to routinely ignore old blocks when deciding how to deal with an editor, and I'm sure many admins do the same. They are useful to know about though, for example if an editor has had a 3RR block you can assume that they know about the 3RR policy. Kevin (talk) 03:59, 6 July 2009 (UTC)

Support

1. Support - Especially were this combined with a "near-" zero (post-warning) tolerance policy for incivilities, the policy-tweaking proposed here sounds reasonable. ↜Just M E here , now 02:30, 6 July 2009 (UTC)
2. Support with two modifications: 1) 31-hour blocks are popular, so it should be blocks of 36 hours or less, and 2) it's six months or one year with no additional blocks during that period. --Carnildo (talk) 05:41, 6 July 2009 (UTC)

   My proposal was "less than 24 hours" not "24 hours or less". That category was intended to catch things like 3 hours blocks with 24 hour blocks being considered more serious. The numbers can be debated, though. I'm not sure about the no additional blocks thing. If we could make it "no additional blocks for the same offence" then I'd quite like that, but I don't think that is technically feasible (esp. considering a lot of blocks aren't for a specific offence but for a last straw after lots of different offences). If I was given a 24 block for vandalism 13 months ago and a 3 hour block for 3RR 2 months ago, is that vandalism block any more relevant than it would have been without the 3RR block? If there are additional blocks they will generally be for ever increasing periods anyway, so it will be clear that there were previous offences (and the longer blocks will stay around longer). --Tango (talk) 15:11, 6 July 2009 (UTC)

3. Support - I like Carnildo's point, so consider me open to suggestion around timeframes, but the principle of rehabilitation should be encouraged. Since Arbs can still see the full log, people who engage in switch and bait behaviors will be caught and rehabilition not made possible.--Cerejota (talk) 20:12, 6 July 2009 (UTC)

   This proposal has less to do with rehabilitating users (changing behavior) and more to do with hiding records of past behavior. --JBC3 (talk) 22:33, 6 July 2009 (UTC)

Oppose

1. Oppose, not necessary, previously banned or blocked users can be allowed back under existing criteria. Each case needs examining on its merits. Jezhotwells (talk) 02:35, 6 July 2009 (UTC)

   You seem to have misunderstood the proposal. This is not talking about allowing people back, it's talking about blocks that have long since expired. --Tango (talk) 03:00, 6 July 2009 (UTC)

2. Oppose. On top of general concerns against increase of meaningless censorship (user actions that preceded the block usually remain in public logs unless they are oversighted)... Why six? why not three neither thirty-three? In real life periods of "spent sentence" have a meaningful correlation to human lifespan (apart from US system with its millenial terms :)) and are weighed on a case-by-case basis. Here all editors are different and each editor evolves at a different pace. In six months some become wikilawyers and successfully pass to adminship, others are still mastering WP:V classes. Some are stuck in juvenile troublemaker mode forever. Six-months term is unnecessarily long for some and too short for others. NVO (talk) 02:47, 6 July 2009 (UTC)

   The censorship isn't meaningless, the meaning is clearly explained in the proposal. The numbers are, as I said, negotiable, but I think they should work. There is no "case-by-case" weighing under the UK act I linked to, there are arbitrary numbers. If the sentence is less than a certain length then after a certain time it is spent. If someone continues to be a troublemaker then they will get newer blocks that will probably be longer, so the old ones being spent doesn't really make any difference. --Tango (talk) 03:00, 6 July 2009 (UTC)

   Real-life "spent term" does not mean erasing or censorship of public records. Former convict assumes full rights, he's relieved of whatever kind of legitimate police attention he had, but the public records remain in place, and noone confiscates old newspapers from the libraries to obliterate the facts... "know your neighbor". The opposite approach has been enforced by German courts: clean start is a clean start, and neighbors have no right to know, neither inquire. Perhaps it works in Germany but I doubt it's relevant here. NVO (talk) 03:58, 6 July 2009 (UTC)

   At least in the UK, the public do not have access to criminal records. You can dig through the records of court cases if you want, but that's all. This isn't a proposal to oversight all mentions of the block, just to remove it from the log. --Tango (talk) 04:02, 6 July 2009 (UTC)

   Tango: "meaningless" perhaps is incorrect word, "futile" may be more correct. That is, usually cause of a block is retained in the archives, so unless the whole thread of events was oversighted, some track remains public. And some folks would go at length to uncover it (see Wikipedia:Requests for adminship/Davemeistermoab although it's not block-related). The offended party (if it existed) will remember the case, etc. Some memory will always remain. I'd rather accept block logs as public records easily accessible to anyone, rather than see "skeletons from the closet" popping up here and there to unsuspected audience (i.e. disrupting normal process). Net wrong from the latter far exceeds net wrong from block log. NVO (talk) 07:01, 7 July 2009 (UTC)

3. Oppose. The last thing that we should do is reinforce the misconception that blocks are intended as "punishments" (as this proposal incorrectly describes them). —David Levy 02:52, 6 July 2009 (UTC)

   It doesn't matter what words I used, this policy should reduce the punitive element. The block remaining on the block log after it has expired (which is when its preventative purpose has concluded) is a form of punishment. --Tango (talk) 03:00, 6 July 2009 (UTC)

   No, the block becomes a form of punishment when its record is deemed a shameful blemish to be expunged for the lesser offenders (on a sliding scale) and retained as a "permanent black mark" for the rest. —David Levy 03:10, 6 July 2009 (UTC)

   If they aren't already seen as a shameful blemish then this policy isn't needed, if they are already seen that way then this policy will do no harm. Do you really think they aren't currently seen that way? If so, what explanation do you give for the disproportionate reaction people have to short blocks? --Tango (talk) 03:18, 6 July 2009 (UTC)

   Some users mistakenly regard blocks in that manner. As I said, the proposed change would reinforce the misconception (or worse still, render the perception accurate). —David Levy 03:24, 6 July 2009 (UTC)

   I'm not sure it would reinforce that perception, I would hope it would make it clear that the perception is wrong by showing that we don't even need to know about old blocks since their purpose is over. I'm not sure what you mean by the perception being accurate or inaccurate. It is always accurate - something is a blemish if and only if it is perceived to be one. --Tango (talk) 03:32, 6 July 2009 (UTC)

   Whether someone chooses to view the block record as a blemish is up to him/her. But it factually is not intended as a punishment, and I'm referring to the mistaken perception that it is.
   Under the current system, all blocks (even brief ones stemming from misunderstandings) remain logged, so it can be conveyed to users that their existence needn't be regarded as a big deal.
   Under your proposed system, we would be proclaiming that the block record is a big deal and presenting it as a permanent punishment for users whose blocks exceed whatever arbitrary duration we set. We wouldn't be "showing that we don't even need to know about old blocks since their purpose is over"; we'd be drawing lines in the sand and assigning increasing penalties to those who cross them. —David Levy 03:44, 6 July 2009 (UTC)

   Not proclaiming, recognising. I maintain that what you describe is already the case. That is why this policy is required. --Tango (talk) 03:51, 6 July 2009 (UTC)

   Please see Wikipedia:Blocking policy and report back on what it says about punishment. —David Levy 03:56, 6 July 2009 (UTC)

   I am perfectly familiar with blocking policy, thank you. If you have a response to my points, please make it. If you don't, concede defeat and change your vote (not that it was appropriate to be voting on a proposal at this early stage). --Tango (talk) 04:02, 6 July 2009 (UTC)

   Concede defeat? What? You've asserted that block records are punishments. I'm asking you to explain how that jibes with the blocking policy. —David Levy 04:06, 6 July 2009 (UTC)

   What's the point in cornering the opponent? Blocks are a form of punishment. They should not be intended to be, but they are punishment for the receiving side. NVO (talk) 04:22, 6 July 2009 (UTC)

   1. I don't know what you mean by "cornering the opponent."
   2. I'm referring to a block's intended purpose and the block log's intended significance. To implement the proposed policy would be to issue an official announcement that a block and the record thereof are intended as punishments, with the latter serving as a "permanent black mark" for those exceeding an arbitrary block duration. Instead of downplaying block logs, it would reinforce the notion that they're a big deal. —David Levy 04:37, 6 July 2009 (UTC)

   Whatever their intended purpose is, they are perceived as punishments and have a punitive effect. Denying that will not get as anywhere. --Tango (talk) 15:13, 6 July 2009 (UTC)

   I don't deny that blocks are perceived by some as punishments. The proposed policy would serve to formally validate that perception and amplify the resultant stress. —David Levy 16:16, 6 July 2009 (UTC)

   I disagree, it is the continuance of the fiction that blocks are never punitive that creates stress. Certainly some blocks are not punitive (ie short duration for edit warring - this is protecting the project), but most incivility blocks are punitive (ie do nothing to solve the underlying issues, only probably move the drama form the article talk to a bunch of other forums), and long-term blocks are always punitive (I prefer community bans).--Cerejota (talk) 12:05, 7 July 2009 (UTC)

   We appear to be discussing two different things. I'm opining (and I do so in more detail below) that this proposal would reinforce/amplify the perception that the a block's record is a big deal and something to be ashamed of. —David Levy 21:23, 7 July 2009 (UTC)

4. Mostly Oppose. Blocking is not punishment, it is a means of temporarily preventing behavior that the Wikipedia community has deemed to be harmful to the Encyclopedia. Before being blocked, users are typically (except in extreme cases) notified of those actions they are not entitled to take, citing policy or guidelines, and are given chance(s) to educate themselves, alter their behavior, and continue by making agreeable contributions. Users who fail to do this are blocked, as they demonstrate what appears to be an intent not to cooperate with the community. When blocked, editors are given an opportunity to appeal the block and be unblocked. At the end of the blocking period, they are most always free to make all the same edits as if the block never occurred. In the case of IP edits, there is hardly a black mark because IP's are anonymous. In the case of established users, I see no reason why the block log has to be accessable to everyone, although when users are blocked but no template is placed on their talk page or is deleted by the user, it is helpful to be able to check the block log to see that they are in fact blocked and for how long. --JBC3 (talk) 06:37, 6 July 2009 (UTC)

   I don't understand your opposition. This proposal wouldn't prevent people from checking the block log to see if someone is currently blocked. Spent blocks would have long ago expired. --Tango (talk) 15:15, 6 July 2009 (UTC)

   To clarify, I was trying to suggest that I'm open to the idea of block histories not being visible to everyone, except that it helps to see when a user is currently blocked. My opposition has to do with the fact that editors are given ample opportunity to avoid blocks in the first place, and therefor earn whatever blocks they get. --JBC3 (talk) 22:31, 6 July 2009 (UTC)

   You are contradicting yourself, first stating blocks are not punishment, then stating that people are given opportunity to avoid them. If blocks weren't punishment, both direct and as a scarlet letter, why would people seek to avoid them in the first place?--Cerejota (talk) 12:00, 7 July 2009 (UTC)

   I'm not contradicting myself. The block is not intended as punishment, but as a method of protecting the encyclopedia and reenforcing community acceptable behavior. People seek to avoid blocks not because they are shameful, but because they are inconvenient and restrictive. --JBC3 (talk) 00:29, 8 July 2009 (UTC)

   But a 3 hour block is not significantly inconvenient or restrictive. It's just 3 hours, you can just go for a nice walk and wait it out or write an article in a word processor while you wait. --Tango (talk) 15:14, 8 July 2009 (UTC)

   I didn't say it was "significant". --JBC3 (talk) 02:39, 9 July 2009 (UTC)

5. Oppose. Deciding that "minor convictions a long time ago aren't important and shouldn't count against you" is something that members of the community can decide on an individual, case-by-case basis. Technical means aren't required here; all it takes is looking at the block log and saying "Gee, that was a long time ago; I can cut this guy some slack." See also my (forthcoming) comment below on possible unintended consequences. TenOfAllTrades(talk) 13:43, 6 July 2009 (UTC)
6. Bad idea. As a community, yes, we're bad at dealing with patterns of harmful behavior. But helping problem editors cover their tracks is not useful in any way that I can see. Someone's previous history is relevant and useful information when deciding how to deal with continuing problems. And, let's face it- as a general rule, editors who have been a problem in the past are likely to continue being a problem in the future. No amount of wishful thinking can change this. Friday (talk) 15:50, 6 July 2009 (UTC)
7. Information retention matters. You damage the project's institutional memory, you jeopardise its future. This proposal, like most that seem to crop up these days, is yet another step away from the radical openness responsible for Wikipedia's success.  Skomorokh  16:02, 6 July 2009 (UTC)

   Actually, so far, the only valid opposition I have heard. However, it crumbles on the fact that there are a lot of things that are hidden from view in wikipedia, like oversight pages, checkuser, various mailing lists, arbcom, IRC chats, and these things were in place for the most part back in the good old days. There is something positive about radical openness, but Wikipedia has never been a radically open project: from choosing the GFDL to having most real administrative functions performed off-wiki by a foundation (and before then Jimbo), we have never been fully open. The fact is, that intended or not, this comment comes across as either naive or sophism.--Cerejota (talk) 12:12, 7 July 2009 (UTC)

   With all due respect, that is a strawman argument. I never claimed Wikipedia has ever been "fully open"; I find it difficult to accept that in the history of encyclopaedia's of any note, you would not accept that Wikipedia, in its articles as well as its administration, is radically more open a) to use and b) to contribute to. Your claim that the argument against decreasing openness "...crumbles on the fact that there are a lot of things that are hidden from view in wikipedia" does not seem to add up. If, to speak very generally, one accepts my initial premises that less transparency is not desirable, this initiative proposes to decrease transparency, and the conclusion that this proposal is therefore undesirable, the argument does not rely on the current or former level of openness and thus is not vulnerable to your objection. Regards,  Skomorokh  12:25, 7 July 2009 (UTC)

8. After consdierable thought. I find Skomorokh, above, has already put my view to words. KillerChihuahua^?!?Advice 16:51, 6 July 2009 (UTC)
9. Oppose- I have only skim-read above, so apologies if this has been mentioned- but there is a strong difference between someone being blocked once, six months ago, for something minor, and being blocked once sixth months ago, once four months ago and twice last month. If there have been blocks since, knowledge of previous blocks would be useful. J Milburn (talk) 22:22, 6 July 2009 (UTC)

   Then lets be creative: the hiding of the log only happens if no subsequent offenses occur. And of course, admins would still have full access to the log, allowing them to make informed decisions. The idea is to avoid scarlet leterring, to have a modicum of human decency and compassion, sadly lacking in the current environment of crime and punishment.--Cerejota (talk) 12:21, 7 July 2009 (UTC)

10. Mostly oppose In principle a policy of infinite tolerance is wonderful, but the proper way for a person to overcome any stigma attached to their blocks is to be a non-disruptive and useful contributor. Since we can't afford proper supervisory staff, we have to make do with simple rules like "3RR and you're out". With this proposed policy, a user might be able to break 3RR twice a year (plus other times when no admin follow up occurs). I can conceive of (say) 12 year olds being silly and getting blocked, and who want to put that behind them several years later, so I might support a policy of clearing (say) short blocks after four years if no further blocks have occurred. Also, I once noticed an admin who accidentally blocked the wrong user (the admin gave an apology and immediately removed the block). Please, let us not have a "remove my block record!" process, but it might be useful if there were a way to clear a log of an incorrect block? Johnuniq (talk) 04:23, 8 July 2009 (UTC)

    Or if not clearing an incorrect block, at least having some immediate notation that the block was removed or reduced. --JBC3 (talk) 07:42, 8 July 2009 (UTC)

    I would be opposed to a way of manually removing blocks from the logs. It would cause added drama. At the moment it is common for a long block to be reduced to "time served", which works quite well as a compromise since it is effectively the same as overturning the block but only accuses the blocking admin of a excessive block, not an incorrect block. If there existed a means to remove blocks you would end up with people wheel warring over it. --Tango (talk) 15:20, 8 July 2009 (UTC)

    And you don't think that there would be any drama or wheel-warring regarding whether a level 1 block (logged for six months), level 2 block (logged for one year) or level 3 block (logged permanently) should be issued in a given situation? ("You can't do that! It'll be logged permanently!" / "The block needs to be longer so it remains logged permanently!") —David Levy 15:46, 8 July 2009 (UTC)

    I agree that is an issue, but I think it is one that can be addressed. Some guidelines on block lengths would help, although if anyone has an idea that doesn't involve rule creep that would be better. --Tango (talk) 17:16, 8 July 2009 (UTC)

    I cannot envision an implementation of this idea under which a block record would not essentially become a formal penalty in and of itself. And I cannot envision the existence of a Wikipedia penalty that wouldn't trigger controversy, strife and attempted exploitation. (Do you know of one?) —David Levy 17:52, 8 July 2009 (UTC)

11. Oppose - per Skomorokh. Logs work 2 ways as well. They're "attached" to the admin who made the block as well as the blockee, so this would have the side effect of removing large chunks of the block history of admins. Meaningful history like log entries and page history should be retained publicly unless there is a very good reason (i.e. legal or ethical) to do otherwise. Mr.Z-man 20:49, 8 July 2009 (UTC)
12. Oppose! This is a well-intentioned, but unsound proposal. (1) The title is misleading. This has nothing to do with rehabilitation. Rehabilitation would be helping disruptive editors change their behavior to become constructive editors. This is about mechanistically (i.e., without applicaiton of human judgment) changing history. (2) A proposal like this is unnecessary. Presumably, whenever a Wikipedian's behavioral history in considered, blocks are considered in their context: how long ago, how serious was the misconduct, and, most importantly, what has the editor's behavior been since the block? If policy enforcement is tightened (as many Wikipedians believe it should be, according to the poll at WP:CIVIL) and blocks are handed out with greater liberality, this will also be taken into consideration when an editor's history is examined. (3) Mechanical rules like this can cause more problems than they solve, and can be subject to gaming the system. The duration of blocks on WP is not closely correllated to the seriousness of the offense and never will be, since one admin determines the length of the block (courts imposising sentences should be more consistent because the stakes are much higher and sentencing standards are better defined); therefore basing expungement solely on the length of a particular block leads to unequal treatment. A 6-day block is relatively serious on WP, and the misconduct that led to it should not necessarily be forgotten. Another inequality is that this proposed policy would expunge (or hide) blocks, but not warnings; warnings are ususally less serious then blocks, but they are considered when an editor's record is examined. Finally, the proposed expungement would mask repeat misbehavior, which should be taken into account whenever an editor's history is considered. This proposal would, for example, hide from voters the (hypothetical) fact that an RfA cadidate has been blocked 12 times for incivility and 1 time for vandalism, including 3 6-day blocks, all within the past 15 months, if all of the blocks were more than 12-months old. I could go on ... Finell (Talk) 21:55, 8 July 2009 (UTC)

    Well said! --JBC3 (talk) 02:39, 9 July 2009 (UTC)

13. OpposeNo need to mess with ze logs.Abce2|Free LemonadeOnly 25 cents!(Sign here) 21:58, 8 July 2009 (UTC)
14. Oppose, per Finell. Don't "clean" the logs! Such a manipulation of history is more typical of dictatorial states, and not appropriate for Wikipedia. Gray62 (talk) 12:30, 9 July 2009 (UTC)

Other

This is a fantastic idea in spirit , but it bothers me severely. Blocks are not supposed to be some sort of judicial punishment - and its useful to be able to see blocks on an account from an administrative standpoint because I need that information to research if I'm seeing a feud or an isolated incident. I may explain better when I am more awake.--Tznkai (talk) 06:51, 6 July 2009 (UTC)

If a feud has stopped for 6 months and never got serious enough to result in a 24 hour or longer block, then I don't think you need to know about it. If you really do for some reason, you can ask a checkuser. --Tango (talk) 15:17, 6 July 2009 (UTC)

Checkusers are overworked as it is, is the problem. My point in general is, having a record is useful for non punitive purposes.--Tznkai (talk) 22:45, 6 July 2009 (UTC)

I can see some unintended consequences, and unanswered questions here.

1. If this policy is implemented, will we start to see 24.5-hour or 7.5-day blocks applied for record-keeping (or even punishment) purposes?
2. How would blocks which were applied and then modified or suspended be handled?
   • Would an indef that was lifted after twelve hours remain on the record permanently, or just for six months?
   • If 'net duration of block' counted (rather than 'imposed duration'), would this be a motivation for wheel warring (by admins) or whinging and ruleslawyering (by blockees)? If I think Bob Admin's eight-day block is unfair, can I render it delible by putting in a five-minute unblock-reblock during day four?
3. Does this mean that certain groups (let's say nationalist POV-pushers) will be entitled to a fresh round of edit warring every six months?
4. On the flip side of that, are we going to have occasional overreactions caused by mistaken identity? (Bob Admin misremembers a block issued seven months ago to JoeUser and issues a longer block for the 'second' offense. In reality the original block had been of JohnUser. There's no easy way to check those facts.)
5. This may severely hamper efforts to look at a particular administrator's record. Documenting long-term issues with blocking may be difficult if the record is invisible to all but a tiny subset of users.

I'm afraid that this proposal – while entirely in good faith and with the best of intentions – is aiming to resolve a social issue with a technical solution. Our blocking policy explicitly notes that blocks aren't for "punishment" and that "cool-down" blocks are to be avoided; blocks are only imposed to protect the project. On the other hand, this proposal aims to encourage more frequent blocking for minor offences to discourage bad behaviour — those sound almost like "cool-down" and "punishment" blocks. (Of course, they'd be punishments for the good of the project, so we're stuck in a logical loop.)

For what it's worth, I'm actually in favour of short blocks for deserving behaviour, and I don't think that blocks ought to be as big a deal as they are. But that's an issue that needs to be taken up through revision of the blocking policy and gradual shifting of attitudes among the admin corps; just making records expire won't do the job. TenOfAllTrades(talk) 14:17, 6 July 2009 (UTC)

I hope you don't mind, I've changed your list to a numbered one so I can respond more easily:

1. You have misread the proposal. I said "less than 24 hours" not "24 hours or less". I think that pretty much eliminates the problem since we don't currently make much use of blocks that are only a little less than 24 hours (or 7 days). If people start issuing 24 hours blocks when they would usually issue a 6 hour block then that is something we can deal with, perhaps with clearer guidelines on block lengths (although I am wary of such rules creep).
2. I didn't really think of that... it's a good point. Off the top of my head, I can't think of a solution, but I'll think about it.
3. If each round never gets bad enough for a 24 hour block, then yes. I don't think that is a serious problem. We can handle minor edit wars once every 6 months. If it gets too bad people will spot it and consult a checkuser anyway.
4. That would be against the rules. Spent blocks should be ignored except under extreme circumstances. The admin you describe would (after suitable warnings) be desysopped. In extreme circumstances, a checkuser can be consulted.
5. That is an interesting point. It would be technically possible to make this only apply to the block log of the blockee not the admin (the logs would still be in the database, otherwise checkusers couldn't see them). I'm not proposing to oversight the user's talk page to remove all trace of the block, so the fact that people can find out about it by remembering who the blocking admin was isn't a big problem.

I don't really support the distinction between "preventative" and "punitive" blocks. One of the main purposes of punishment is to act as a deterrent to prevent future offending. I think most blocks are actually intended to be punitive. Blocks of less than a few months don't really do much to prevent disruption, they just defer it. If the blockee doesn't re-offend it is because they don't want to be blocked again. The alternative is that they don't re-offend because they have cooled-down. So either it's a cool-down block or its a punitive block, both of which we claim aren't allowed. We should stop hiding behind what it says in policy and start talking about what actually happens. --Tango (talk) 15:32, 6 July 2009 (UTC)

PS. I agree this is a social problem, but I can't think of a social solution (other than wishful thinking that it will all sort itself out if we just say a few catchy slogans every now and then). If anyone else can, please propose it. --Tango (talk) 15:36, 6 July 2009 (UTC)

Well, let's be honest, then — this proposal isn't really about clearing out the records of blocked individuals, it's about changing the 'culture' of blocking on Wikipedia to encourage the more frequent use of short blocks for the purpose of behaviour modification. I'm pretty sure I actually agree with you on that point, but I don't think that the technical means you've proposed here are the right way to go about it...or at least are not the right place to start.

I strongly suspect that what you're looking to do is modify the blocking policy to permit admins to do openly what they now have to do covertly, or under cover of other policies: block obnoxious users who are acting like twits. I think that's an excellent aim, but I think you're going at it bass-ackwards. Right now, there are very few short blocks that would be cleared by your proposal, so it is of questionable utility. Its chief justification would seem to be that it would encourage new blocks which might be permitted in the future, preemptively solving a problem which does not yet exist.

I think what you need to do first (and you're free to disagree with me on this) is establish the changes to the blocking policy, giving de jure backing for a more liberal application of short blocks. We can revisit this proposal at that time, or possibly later, or perhaps even never — I'm still not convinced that this technical change will make the difference you're hoping for in the adoption of short blocks, and I still wonder if the potential pitfalls I listed above won't outweigh the potential benefit of 'clean slates'. I also fear that this change might encourage admins to be less cautious in applying and in reviewing short blocks. (You know what we'll see on AN/I: "Joe Admin issued me a bad block." "Quit whining. It was only six hours, you're unblocked now, and your record will be clear in six months." Lather, rinse, repeat.) That lack of caution may be particularly unhelpful during a transition period to more liberal blocking, helping to solidify opposition to what would be a sensible change. TenOfAllTrades(talk) 16:06, 6 July 2009 (UTC)

Well said, TenOfAllTrades. Tango, your assertion that "most blocks are actually intended to be punitive" is not backed by the blocking policy. If you wish to change the blocking policy, please propose that. A backdoor proposal pretending that the blocking policy already says what you want it to say is not the right way to go about this. —David Levy 16:16, 6 July 2009 (UTC)

Please read my comments to the end. I just said: "We should stop hiding behind what it says in policy and start talking about what actually happens." You respond by hiding behind what it says in policy... --Tango (talk) 16:55, 6 July 2009 (UTC)

I read every word. I'm not hiding behind the policy; I'm plainly stating that your concerns would be properly addressed by seeking to change it instead of pretending that it doesn't exist in its current form. —David Levy 17:29, 6 July 2009 (UTC)

I will note for the record that the blocking policy all-blocks-are-preventative bromide has always been a bit of a sham. If we felt that blocks couldn't change behaviour, we wouldn't ever issue anything less than a permanent (not just indefinite) block. 3RR blocks are very much both 'punitive' and 'cool-down' in nature. A lot of heat but little light comes from trying to interpret the blocking policy, in one of our classic descriptive versus prescriptive policy conflicts.

The notion that we cannot issue blocks for punishment (with the aim of changing behaviour and improving Wikipedia thereby) has always been hippy-trippy free-software free-love bollocks. At some point, saying "Stop, or I'll say 'Stop!' again!" always gives way to "Stop, or I'll take away your toys and send you to your room for a bit." TenOfAllTrades(talk) 16:38, 6 July 2009 (UTC)

Also for the record, I'm inclined to agree with the above. —David Levy 17:29, 6 July 2009 (UTC)

Maybe I'm just out of touch with modern process, but I still believe that policy pages are descriptive, not prescriptive. You change policy by changing attitudes and then update the policy pages accordingly. This proposal would remove what I believe is the main thing preventing admins using short blocks. It's not policy that prevents stops them (blocking policy actually gives admins a great deal of discretion) it is attitudes. They know the community doesn't like blocks for minor offences so, to avoid drama (an admirable intention), they don't issue them. If we change attitudes, admins would be free to do what needs to be done. --Tango (talk) 16:55, 6 July 2009 (UTC)

The problem is that your proposal doesn't seek to change attitudes; it assumes that your attitude already prevails.

And even if that were so, it's my opinion that the proposed policy would exacerbate the problem that it purportedly seeks to rectify (and I've explained why above). —David Levy 17:29, 6 July 2009 (UTC)

What are you claiming is my attitude? My attitude is that old blocks shouldn't matter, this proposal is based on the assumption that other people have the opposite attitude. By removing old blocks, they won't matter any more, so the attitudes will change. I understand what you are saying, but that is only the case if I am wrong about why people make a big deal out of short blocks. If I am wrong, then I agree that this proposal is not a good idea. If you think I am wrong, then please explain what you think is the real reason. --Tango (talk) 17:52, 6 July 2009 (UTC)

1. Your attitude is that blocks are inherently punitive and should be officially treated as such. I actually see some merit to this, but it does not reflect current consensus. Said consensus can be changed, but not by proposing a policy that assumes that it already has been.

2. The proposed policy would not result in the removal of block records purely on the basis of age; they would be removed on the basis of age and duration. And irrespective of that, the system's creation would formally tell the community that they should perceive all such records (whether temporary or permanent) as shameful blemishes. It would draw additional attention to them (and doubtlessly lead Wikipedia Review or another outside website to maintain an unofficial block log, thereby drawing still more attention to both the blocks themselves and the "attempted cover-up"). —David Levy 18:20, 6 July 2009 (UTC)

1. That's not an attitude, it's a fact. It isn't determined by consensus it is determined by reason.

2. It doesn't tell the community that they are blemishes, it tells the community that they are perceived to be blemishes but shouldn't be. If you have a suggestion for stopping that perception, then please speak up, but I can't think of one. If we can't change the perception we have to accept it and deal with it, not deny it. It is possible that this will backfire and draw attention to these blocks, but if that does happen I would expect it to be short term. People would get used to the new system over the first few months and it would cease to be a big deal. I don't think enough people pay attention to WR to warrant worrying about how they would react to this policy. --Tango (talk) 18:33, 6 July 2009 (UTC)

1. It is not factual that blocks should be officially treated as punitive. It can reasonably be argued that this is their de facto nature, but current policy is quite different. (You're correct that we mustn't hide behind policy, but it's equally inappropriate to pretend that it doesn't exist.)

2. No, it would not tell the community that they shouldn't be perceived as blemishes. It would do precisely the opposite. If a block record shouldn't be perceived as a blemish, there's no reason for it to be expunged. To enact the proposed system would be to formally announce to the community that the record's removal is desirable, so the record, therefore, must be a shameful blemish (whether it remains for six months, a year or permanently).

I don't know the extent to which the pre-existing perception can be countered, but I'm confident that the proposed policy would serve to validate it. Only by not treating the block log as a big deal can we even hope that it will be perceived accordingly.

3. I don't know how many people pay attention to WR, but I do know that handing them ammunition will not reduce that number. —David Levy 18:59, 6 July 2009 (UTC)

If you keep misinterpreting me like this I'm going to have to believe you are doing so intentionally. I didn't say it was fact that they "should be officially treated as punitive" (that's obviously not a factual matter). I said it was fact that they 'are treated as punitive. What policy says is almost completely meaningless. What matters is what actually happens. --Tango (talk) 21:09, 6 July 2009 (UTC)

1. There's no need to accuse me of acting in bad faith. I believe that you've misinterpreted some of what I've written, but I wouldn't accuse you of doing so intentionally.

2. This is the exchange that occurred:

Me: Your attitude is that blocks are inherently punitive and should be officially treated as such...

You: That's not an attitude, it's a fact...

Perhaps there was a misunderstanding along the way (and I apologize for whatever role I played), but it certainly appeared to me that you were referring to my entire statement.

3. My point is that blocks officially are not punitive. Even if the policy is incorrect (and I find the evidence to that effect compelling), that doesn't render it nonexistent. Current consensus is that punitive blocks should officially be prohibited (even if that isn't what actually occurs), and the correct course of action is to seek to change that consensus (thereby changing the policy), not to proceed as though consensus already has changed.

But again, even if that occurs, this proposal still would be ill-advised (no offense intended). —David Levy 21:45, 6 July 2009 (UTC)

I don't care about "officially". Consensus is that we should make punitive blocks, that is obvious - you just have to observe how many such blocks are made without complaint. That everyone says the opposite doesn't make it so. --Tango (talk) 22:20, 6 July 2009 (UTC)

You might not "care about 'officially'," but the community does. It's unfeasible to introduce a policy based on the assumption that blocks are punitive without first modifying the policy that explicitly states that they aren't.

But let's assume that the blocking policy is revised to indicate that punitive blocks are permitted and recommended. That still won't make this proposal a good idea (again, no offense intended).

Under the current system, one can point out that many respected editors (even Jimbo) have been briefly blocked in the past. ("So see, it's no big deal.") The proposed system would accomplish the following:

• It would reinforce the perception that the block records are a big deal. (Otherwise, why would there be any point in expunging them?)
• It would cause whatever block records are visible at any given time to stand out considerably more.
• It would create a sliding scale of shame (six-month "black mark," one-year "black mark," "permanent black mark").

All of this is in addition to the many logistical problems cited elsewhere. —David Levy 22:31, 6 July 2009 (UTC)

Wait, so other than the name, what does the rehab Wikiproject have to do with this?Abce2|Free LemonadeOnly 25 cents!(Sign here) 16:56, 6 July 2009 (UTC)

There's a rehab Wikiproject? I named the proposal after the Rehabilitation of Offenders Act 1974, on which it is based. --Tango (talk) 17:10, 6 July 2009 (UTC)

There's a link at the bottom of the page.Abce2|Free LemonadeOnly 25 cents!(Sign here) 17:53, 6 July 2009 (UTC)

Oh, yes, I didn't realise someone had added that. There is no real connection other than that they both help previous offenders become useful editors (also that isn't really the intention of my proposal, the intention is to prevent admins being afraid to issue blocks). --Tango (talk) 18:38, 6 July 2009 (UTC)

I will accept that blocks are not punishment when:

1. They are not used as evidence in ArbCom
2. During AN/I discussions, they are not brought up as reasons for community bans
3. There is no stigma attached to them
4. There is no hostility and incivility on the part of blocking admins

Arguing against rehabilition only further reinforces the punitive function of blocking.

A 24 hour block given once or twice, I can see not being punishment. A one week block is punishment. The sooner we do away with pretensions otherwise, the sooner a lot of drama will dissolve into thin air. --Cerejota (talk) 20:18, 6 July 2009 (UTC)

Well said. --Tango (talk) 21:11, 6 July 2009 (UTC)

• I'll debate this matter when there's a culture of properly warning and assessing before blocking experience users. Tony (talk) 11:11, 7 July 2009 (UTC)

A non-overturned block is a sign that someone did something block-worthy. Repeated blocks show a pattern of problematic behavior. Just because they can be used as evidence doesn't necessarily mean that they're a punishment. A block log entry is an easy way to refer to a specific incident without having to dig through talk page archives. Mr.Z-man 21:53, 8 July 2009 (UTC)

A non-overturned block is a sign that either they did something block-worthy or that the block has expired by the time it was determined that they hadn't. With long blocks, it is far more likely to be the former than the latter, but with the short blocks that this proposal in concerned with the latter is not unusual. --Tango (talk) 22:01, 8 July 2009 (UTC)

Revising history

Perhaps I'm simply sleep-deprived, but I'm having difficulty understanding this proposal. Are you suggesting we retroactively alter block logs? I'm strongly opposed to rewriting or revising history for appearances' sake. But maybe I'm simply misunderstanding the purpose here. --MZMcBride (talk) 10:00, 7 July 2009 (UTC)

It would be both retroactive and proactive. I wouldn't call it altering logs, just letting them expire when they are no longer relevant. There is certainly no rewriting or revision, just removal, and it would be perfectly transparent. --Tango (talk) 22:21, 7 July 2009 (UTC)

This seems like a pretty bad idea. If there is stigma attached to logs, that's a societal issue, not a software one. There's no reason to go around messing with logs. It's dangerous and any benefits simply aren't worth the costs. I also can't imagine this functionality ever being implemented into the software. --MZMcBride (talk) 23:59, 7 July 2009 (UTC)

And it's my belief that the stigma would be reinforced/amplified. We would go from "A block record is no big deal. See, even Jimbo has some." to "A block record must be a big deal, because it's considered desirable to expunge." and "Oh, you have block records, so you must've done something terrible!". —David Levy 00:14, 8 July 2009 (UTC)

I'm inclined to agree with MZMcBride and David Levy here — while the intent of the proposal (and its proposer) may be to reduce the social effect and importance of a block log, I fear that the effect will almost certainly be the reverse. There will be two classes of editors: the 'good' ones with spotlessly (apparently) empty block logs, and the 'bad' ones, who have unexpired or – even worse – indelible blocks. TenOfAllTrades(talk) 00:35, 8 July 2009 (UTC)

Why can't you imagine it being implemented? If the community wants it, it will get implemented. It's a pretty simple feature, I could probably write it myself. --Tango (talk) 00:42, 8 July 2009 (UTC)