Misplaced Pages

Revision as of 06:41, 4 April 2013

Where does one make appeals

Where does one go, to appeal one's topic-ban and/or Arbcom restriction? GoodDay (talk) 21:15, 20 March 2013 (UTC)

Misplaced Pages:Guide_to_appealing_blocks#Arbitration_enforcement_blocks (not sure if that includes those imposed by arbcom directly, but presumably you can appeal to arbcom). Jimbo also reserves the right to overrule the committee Misplaced Pages:JIMBO#Arbitration_Committee. IRWolfie- (talk) 22:11, 20 March 2013 (UTC)

If the Arbcom restriction was the result of an arbitration case or motion, you can file a request for amendment. --Alexandr Dmitri (talk) 23:24, 20 March 2013 (UTC)

Is there a less complicated way of appealing a topic ban? This form is long, confusing and frustrating. Humanpublic (talk) 00:50, 21 March 2013 (UTC)

• Previous requests for clarification have gone unanswered. It seems that a (topic) ban cannot be reversed or reduced without community discussion or a Ban Appeals Subcommittee (BASC) decision, but there seem to be no guides to the former.
• Arbcom (BASC) is probably reluctant to overturn AN "community" topic bans—it appears that such bans should at least initially be appealed at AN. Arbcom is supposed to be for problems that the community has not been able to solve.
• Admins at AE seem much fairer and more lenient about topic bans than the "community" is about AN topic bans. Recently somebody was given a one-month topic ban at AE, whereas at AN people have routinely been given indefinite "community" topic bans—without due rationale, process (due diligence), or discussion (a bare 24 hours on a holiday weekend)—for very much less disruption in essentially the same topic area. Lots of people who participate in ANI topic ban discussions seem to be just there to participate in a lynching (they don't have any knowledge of the topic or person)—this is the reason that I put "community" in quotes above: I don't think "community" has anything to do with lynching mobs. It takes courage for an admin. to implement a fair decision (and take flack for it), but it doesn't take any courage to participate in a lynching.
• If GoodDay's "mentor" does not support lifting of his topic ban, then perhaps he could look for a new mentor. LittleBen (talk) 02:13, 21 March 2013 (UTC)

I've much to think about in the coming hours. GoodDay (talk) 03:46, 21 March 2013 (UTC)

As I like to say, "ANI: You will never find a more wretched hive of scum and villainy." But I have seen at least once, ANI (or was it AN?) overturn an ill-conceived and ill-thought-out topic-ban given at AE. A Quest For Knowledge (talk) 10:17, 21 March 2013 (UTC)

From what I've read, there's a way to appeal a block for allegedly violating a Arbcom restriction. However, I'm not seeing how one can appeal the Arbcom restriction itself. GoodDay (talk) 14:09, 21 March 2013 (UTC)

You need to make an amendment request. T. Canens (talk) 14:25, 21 March 2013 (UTC)

What is AE? — Preceding unsigned comment added by Humanpublic (talk • contribs) 14:21, 21 March 2013 (UTC)

AE = Arbitration Enforcement. A Quest For Knowledge (talk) 14:24, 21 March 2013 (UTC)

I'm not sure if there are any hard rules about this, but you probably should appeal to the same venue that (or higher) applied to the restrictions. Although, as I pointed out earlier, there's at least one case where an AE applied sanction was over turned by the community. A Quest For Knowledge (talk) 14:30, 21 March 2013 (UTC)

AN topic-banned me yesterday. What would be the point of appealing there? Humanpublic (talk) 14:36, 21 March 2013 (UTC)

Actually it was a community ban. Try respecting the ban for a minimum of 6 month and the return to AN and request an community un-ban. You jumped straight to Arbcom (which I understand was suggested from your request of how appeal) but you did it immediately after the actual decision was made. If you can't accept it then maybe you are to attached and interested in these subjects to see your own bias. I really suggest working with the community as the best route to take at this point.--Amadscientist (talk) 00:35, 26 March 2013 (UTC)

There needs to be a way to appeal the correctness of a community ban. If we have to submit to the ban before appealling it, there isn't actually a way to appeal. Humanpublic (talk) 18:59, 31 March 2013 (UTC)

When do requests get archived? At the moment the vote is 5-0 against. Humanpublic (talk) 19:00, 31 March 2013 (UTC)

Against the 25 March 2013 block of Rich Farmbrough

Fram and Sandstein have again gone too far.

1. The block discussion was closed too soon, without adequate time for response by anyone. 10:29 to 23:04 is just too damned short. The alleged offense simply does not rise to the level of an emergency response.
2. I disagree with the Request for Arbitration, the block, the logic of it, the short timing of it, the "gang of two" mentality, and the lack of actual discussion (replaced by a monologue). Its implicit rejection of the notion of Misplaced Pages as a socially constructed encyclopedia shocks the conscience. Farmbrough's contributions have so completely outweighed his errors and typos, that this blocking action is just wiltingly, shoulder-stoopingly, depressingly beneath us all. This block is a shameful embarrassment to the project. Of course I wouldn't DARE to unclose the "discussion", but if wikilawyering is what you do, wikilawyering is what you'll get back: Search and replace IS NOT EXPLICITLY FORBIDDEN in the precise language of the linked "the decision", as alleged by Sandtein. Those words do not appear in "the decision". Sandstein wants us to believe they do, but they don't. Decisions are standalone and are atomic; they cannot be arbitrarily expanded at will. Sandstein's rationale cannot stand. It's void. He should not be permitted to fabricate extensions to "the decision". Period.
3. While we're at it, let's have an interaction ban between Fram and Sandstein, to break up this mini-cabal against Farmbrough. No team of two editors should ever be allowed to so egregiously, so repeatedly wage war against a productive, civil, enthusiastic (overenthusiastic) editor. IMHO.
4. Smarter would have been to send notice of all of Farmbrough's "mistakes" to my attention. I'd fix them for the rest of my tenure as a Misplaced Pages editor. That's a constructive, non-punitive solution. --Lexein (talk) 09:52, 26 March 2013 (UTC) revised to clarify too-short time before sudden action --Lexein (talk) 10:24, 26 March 2013 (UTC)

An interaction ban between me and Sandstein? I have hardly ever interacted with him. I don't choose who does AE enforcement and who doesn't. Your request doesn't make any sense. Fram (talk) 09:58, 26 March 2013 (UTC)

(edit conflict) Well, I've seen prose by the two of you, in time sequence. Gee, where? Oh yeah, the other complaints against Farmbrough, and those blocks of Farmbrough. My request will make sense to people who aren't you, apparently, so it's ok. --Lexein (talk) 10:24, 26 March 2013 (UTC)

Fram does have a point here; if you file an AE request about anything then it is very likely Sandstein will be involved. ErikHaugen (talk | contribs) 16:08, 26 March 2013 (UTC)

As for the rest of your comments: "to make only completely manual edits (i.e. by selecting the button and typing changes into the editing window);" is the relevant part of the sanction. So it seems to me as if "search and replace" is explicitly forbidden after all. Fram (talk) 10:00, 26 March 2013 (UTC)

As for your false equivalence, search is ctrl-F, and is legal. Search again is F3 (on PCs), and is legal. Copy (ctrl-c) of say, ' ' ', is legal. Paste (ctrl-V) of say, ' ', is legal. Anything I can type in the edit window is legal. None of these are automatic. So it seems your conclusion that the errors could only have been created offline can be demonstrated false, by a series of human errors while searching in the edit window, ' ' ' again and pasting ' '. You would punish me for coffee jitters, wouldn't you? Your concocted re-imagining of circumstance, without actual literal proof, for the sole purpose of hounding, blocking, and hoping to get an editor banned for editing normally, is not what we expect or want of Misplaced Pages editors. If you don't mind, please stop. --Lexein (talk) 10:24, 26 March 2013 (UTC)

The conversation above piqued my interest, so I looked into the evidence of violation. I believe that I have the skills to distinguish between automated, manual, and cannot-tell edits, plus I have had no interactions with anyone involved in this.

A human typing changes into the editing window does not accidentally change into multiple times without ever accidentally changing into . No conceivable sequence of cuts and pastes would result in that pattern or errors (not to mention that pasting is not typing, and thus is not allowed) Clearly some sort of automated editing was used, and the result pasted into the edit window. --Guy Macon (talk) 11:00, 26 March 2013 (UTC)

Yes, but on that basis, copying and pasting a URL into the edit window (or the cite tool) as part of creating a reference would also be disallowed, and I'm sure no-one is going to claim that's an automated edit. This falls into a grey area, IMO. Black Kite (talk) 11:11, 26 March 2013 (UTC)

For me personally, cut-and-paste is indeed not part of automation (not only the example you give, but e.g. also swapping two sections or some such should still be allowed); what is disallowed is everything repetitive that isn't done one at a time, but appears to be done with the aid of some tool, which has the result that all errors are repeated multiple times as well (which was the basis of the ArbCom case and restriction). Whether this is in one page or over multiple pages is to me not relevant here (although obviously I'ld rather have errors in one place than all over the place). Fram (talk) 14:52, 26 March 2013 (UTC)

I don't think it requires too much imagination to see how the single quotes could get messed up with entirely manual edits. I make almost exactly that kind of mistake myself. In this case, there was a rare angled quote right inside the double single quote for italics. Rich was trying to convert triple-single quotes to double-single quotes (bold to italics) so it's easy to see how he might have found a few tripled single quotes and replaced them. I am not convinced this was done using search and replace. ErikHaugen (talk | contribs) 16:14, 26 March 2013 (UTC)

Please look at version 1 and version 2 (diff) and search for the following phrases (without the square brackets) in each version:

Notice the word to the left of each search phrase? None of these errors are consistent with your "there was a rare angled quote right inside the double single quote for italics" theory. Nor are they consistent with manual editing. No reasonable human removes a quotation mark standing alone with no nearby bold or italic wikimarkup to confuse the issue and misses the matching quotation mark on the other side of the word.

Now look at the text after and before in each version. A big chunk of the article is gone,replaced by a big red cite error. The same mistake was made on the word Sunday inside the cite error. How does a human making manual edits miss the fact that entire sections are missing and replaced with a cite error and yet manage -- inside the missing text -- to delete a quotation mark while missing the matching quotation mark on the other side of the word? Those edits were made by a computer program and no human looked at the preview before saving. --Guy Macon (talk) 17:32, 26 March 2013 (UTC)

I would welcome any comments by anyone who repeated the experiment I detailed above and failed to come to the same conclusion that I did. I would especially welcome any theories about how those particular errors came from manual editing. The "Wikimakup bold/italic next to the quotation mark" theory is not supported by the data. --Guy Macon (talk) 00:32, 27 March 2013 (UTC)

Note project canvassing by Lexein: . Fram (talk) 11:12, 26 March 2013 (UTC)

Discussion there started with a different purpose: to see if that project cares to try to retain editors of all sorts. But you knew that already, since you read it. So, no thanks for the ABF. Cheers, --Lexein (talk) 16:35, 27 March 2013 (UTC)

Blocked until March 2014? That's too severe, folks. GoodDay (talk) 11:13, 26 March 2013 (UTC)

Indeed; 1 year blocks should not be used, ever, in cases like this where so little, if any, harm is being done. Ever. ErikHaugen (talk | contribs) 16:14, 26 March 2013 (UTC)

Rich Farmbrough agreed to certain restrictions in order to avoid an indefinite site ban and was told in no uncertain terms that any type of contributions that are even reminiscent of automated editing would result in a site ban He violated that agreement. A one year block is lenient.

As for harm, when you run an automated tool which makes mistakes, you can harm many, many pages. Users of automation tools have a heightened responsibility to the community, and are expected to comply with applicable policies and restrictions; to respond reasonably to questions or concerns about their use of such tools and to respect the community's wishes regarding the use of automation. Rich failed on all counts.

Arbcom found (all findings 9 to 0 or 10 to 0) that Rich Farmbrough repeatedly violated the restrictions imposed by the community on his use of automation, repeatedly violated the letter and the spirit of the bot policy by running high-speed and high-volume tasks without approval, running unapproved bot tasks from a non-bot account with no bot flag and no indication that automation was used to perform them. He kept doing that after agreeing not to as a condition of avoiding a site ban. --Guy Macon (talk) 18:14, 26 March 2013 (UTC)

Has RF been running automated tools with mistakes over many many pages recently? If not, then what is the harm here that this block is preventing? HaugenErik (talk) 23:53, 26 March 2013 (UTC)

The harm is thus: before he was banned, he ran unauthorized and very buggy bots, leaving a huge mess for other people to correct. This was not an occasional thing. It was an ongoing problem, so the community placed restrictions on him. Which he refused to follow, causing even more problems for others to clean up. Given his history, it is unreasonable to ask us to give him another shot at messing up multiple articles with automated tools. It doesn't take a rocket scientist to figure out that if we let him, he will repeat the behavior of the past. Enough is enough. --Guy Macon (talk) 00:23, 27 March 2013 (UTC)

• Unblock and if really necessary come up with a more sensible restriction which frees Rich up to help the pedia. I probably wouldn't have bothered to look into this if it was a short block of a week or so. But I really see no attempt at defusing this before going straight to enforcement, and then no attempt to build consensus before blocking for a whole year. ϢereSpielChequers 17:54, 26 March 2013 (UTC)

Outsider's comment: The edit in question is a textbook example of an edit that was very likely not automated: It did change two bold phrases in the article to italics, and it did also remove several inverted commas, wherever they were within italics formatting--no offline search-and-replace would do that, unless specifically instructed that way like:

Wherever you find
'' followed by ' OR '' followed by ‘ OR ’ followed by '' OR ' followed by ''
Replace with ''

This would be an extremely stupid thing to do, as there is no use case for such an instruction. On the contrary, had he been using automation this error would not have occurred (Note that there are 3 different characters to consider). I think I can make an educated guess what happened:

1. RF edited Mohan Deep to remove the two bold statements he saw.
2. Without previewing again, he spotted the occurrences of ''‘ and ’'' in the source code, and mistook them for '''
3. He removed one of the ' and saved-visually there is nothing wrong with comma-separated film titles in italics, the missing inverted commas do not hurt anyone.

The other edit is probably automated in some way but do you really want to block for a year for a single edit that clearly improved the 'pedia? --Pgallert (talk) 17:58, 26 March 2013 (UTC)

Your theory is inconsistent with the edit you cited, which contains multiple errors removing left quotation marks that were nowhere near any bold or italic wikimarkup. See my analysis above for details. --Guy Macon (talk) 18:22, 26 March 2013 (UTC)

I also agree this was a bad decision. Rich was only unblocked for a couple days. Fram watched every edit until he say him do something that he could somewhat justify and then submitted the Arbitration enforcement knowing that Sandstein wouldn't question it and would invoke a year long block. A block he wanted to do previously I might add but only reduced because of outrage from the community that forced him to shorten it. There are so many problems with this its hard to start but here are a couple:

1. The AE was opened and with only comments from the one that has been trying to block Rich for years (Fram) and Sandstein a year long block was implemented.
2. The user never had a chance to comment. At minimum AE's should be open for a length of time to allow the accused to respond.
3. AE should not be a one man show as it has been
4. Fram is clearly involved and has been asked repeatedly to stop hounding Rich and following his every edit. No matter how good of an editor you are, the policies of Misplaced Pages are so open to interpretation as is Rich's Arbitration finding that nearly anything can be justified as blockable.
5. Rich didn't agree he was forced to agree. There is a huge difference. Kumioko — Preceding unsigned comment added by 138.162.0.42 (talk) 18:46, 26 March 2013 (UTC)
6. It also seems rather suspicious that this comes on the heals of Rich asking several users why I was blocked. Maybe unrelated but there are an aweful lot of people who want to shut me up because they don't like what I have to say. Maybe its not related but its a rather suspicious bit of timeing at the very least. Kumioko 138.162.0.46 (talk) 18:54, 26 March 2013 (UTC)

There are plenty more but I will leave that since Fram, Sandstein or one of their suppoters will likely delete my comments anyway. Kumioko 138.162.0.41 (talk) 18:45, 26 March 2013 (UTC)

Is this Encyclopaedia now ruled by the likes of Sandstein and Fram? - Don't bother to answer, it's rhetoric. It seems rather sad that good, long term, and productive editors are now blocked in this Draconian fashion as though they are some anon who wanders in off the street shouting obscenities every five minutes. I'm quite sure there is a better and more respectful solution than this.  Giano  19:32, 26 March 2013 (UTC)

Experienced editors should be held to a higher standard, not a lower one. --Guy Macon (talk) 19:43, 26 March 2013 (UTC)

@Giano and Guy Macon. Vested contributors are a poor analogy, if a new editor had arrived and done the mainspace edits that Rich Farmborough did in his brief unblock then would they even have had a warning? If they'd been blocked for this it would have been seen as an obvious bad block. There is a rejected proposal that we should exempt vested contributors from some of the rules that the rest of us are meant to follow, but that doesn't really apply here. More pertinently there is an assumption that experienced editors should be judged by the rate of their mistakes, if you ignored the 99.5% least contentious edits that I've made you might get a pretty poor impression of me. Ignore the 99.5% least contentious edits that Rich Farmborough's has done and there would be over 5,000 still to consider. ϢereSpielChequers 20:08, 26 March 2013 (UTC)

Guy Macon, can you then also search for in both versions end explain how a computer could have managed to remove the ’ after Life while replacing the ’ in It’s? --Pgallert (talk) 19:52, 26 March 2013 (UTC)

No, he decided to remove all the studid ’s and ‘s (except in It’s, where he replaced it with ') while only looking at the edit window, and missed a few, two of them in text that had a preceding <ref>. Looks quite human to me.--Pgallert (talk) 20:04, 26 March 2013 (UTC)

I just did the search. There are multiple occurrences of the phrase, but there is no problem with getting a search-and-replace to find the right quotes and not the apostrophes. Just search for a space after the character.

Why would a human so often miss the quotation marks on one side of a word and not the other? Why are the mistakes not evenly distributed between left and right? If indeed he was manually editing, he would have done what we all have done: "nuke the left quote, nuke the right quote, nuke the left quote, nuke the...OK where is that stupid right quote? Ah! there it is! nuke the right quote, nuke the left quote, nuke the right quote, nuke the... Huh? Another right quote? Must have missed one..." and so on. The only way a human nukes a bunch of left quotes in a row is by using an automated search and replace of some sort. Otherwise hitting the delete key over a bunch of left quotes in a row would have stuck out like a sore thumb. That's a basic part of editing any text: when you add or delete a quotation mark you immediately add or delete the matching quotation mark. And a basic part of editing Misplaced Pages is hitting the preview button and paying attention when an entire section is missing and there is a cite error that is bold and red.

Also, please remember that the restriction he was placed under specifically said "any edits that reasonably appear to be automated shall be assumed to be so" and that what he agreed to was to avoid "any type of contributions that are even reminiscent of automated editing" with the clarification "If it looks like automation, it will be treated as automation—and therefore as a breach". Combine that with the fact that Rich has used automation when using automation was banned not once, but several times before. It was his responsibility to make sure that his edits did not look like automated edits. --Guy Macon (talk) 23:05, 26 March 2013 (UTC)

or better still, not to use any form of automation at all. Leaky Caldron 23:09, 26 March 2013 (UTC)

I wanted to add my voice to those here that think a one year block is an extremely bad policy decision. As I mentioned on Sandsteins talk page, 1 one year block (which BTW is the time limit on the admin tools I believe) should be reserved for things like Copyright violators and Vandals, not editors doing some frivolous edits. Its not bad enough that Rich's automation ban is costing the pedia tens of thousands of edits a month. But we have to make matters worse by imposing a 1 year ban because a vindictive editor who has tried to get Rich banned for years, opens an AE thread, Sandstein closes it 13 hours later with no other comments but his and the one who opened it and then imposes a one year block. And somehow we are ok with this? This doesn't at all seem like a problem to anyone else? Really? I find that rather hard to believe. Especially when we don't know if he was using automation, in fact, it appears the edits were done manually. So we blocked him for a year because an editor who has been gunning for him for years suggested that Rich might have done something to violate his Arbcom determination. Which itself was poorly written and unnecessary. We should all be really proud in this community we have all volunteered our time to participate in. KumiokoCleanStart (talk) 01:23, 27 March 2013 (UTC)

Guy Macon, where exactly are you quoting from when you say "any type of contributions that are even reminiscent of automated editing"? Dsp13 (talk) 03:16, 28 March 2013 (UTC)

I am quoting from Rich Farmbrough's talk page. Here is the conversation, with the direct quote in bold.

(SirFozzie) "I know AGK has been conversing with you via email. We're trying to fashion terms for something short of a long term site ban. If the Committee can come to a consensus to extend you this kind of offer in lieu of a full site ban, would you accept this?"

(Rich Farmbrough) "Yes, indeed I had already said as much."

(SirFozzie) :Ok, good. I'll report to the Committee, and you should be hearing from us, one way or the other."

(AGK) "I have re-voted to oppose the site-ban motions, in light of your agreement to move away from any type of contributions that are even reminiscent of automated editing. I would re-emphasise point C) of Fozzie's e-mail, which I interpret to mean 'If it looks like automation, it will be treated as automation—and therefore as a breach'." (emphasis added)

And, just for completeness, here is where Rich Farmbrough (weeks later) deleted the section without disputing the claim that he had agreed "to move away from any type of contributions that are even reminiscent of automated editing" as a condition of not being immediately banned: --Guy Macon (talk) 05:53, 28 March 2013 (UTC)

Thanks. So it was in a side-discussion, rather than the main Arb page. One of the really unfortunate features of this case seems to me the way in which the scope of 'automated' expanded so precipitously in the rulings and subsequent enforcement: (1) from bot activity (through reading, rather than editing, WP using tools) to cut and paste, (2) from activity of a certain sort to appearance (or even more widely, here, reminiscence) of that sort of activity. The interaction between (1) and (2) has been wild. I can't see that Rich F could reasonably have understood himself to have been agreeing in his exchange with Sir Fozzie to not make edits which look like cut and paste. Dsp13 (talk) 09:59, 28 March 2013 (UTC)

Except it doesn't look like cut and paste. Go to my original analysis (searching on "Please look at" finds it), perform the same steps I performed, and you will see a series of edits that cannot be explained by cut and paste errors, but which can easily be explained by cutting and pasting into a text editor, manipulating the text with a series of automatic boolean or regex search and replace operations (along with some manual editing), then cutting and pasting it back into the edit window and saving without previewing. The edits show a classic automation error: first you automatically search and replace all of the left quotes, then the right, only to discover that the second operation caught some apostrophes, so you undo and try other things like searching for a space after the right quote. This misses the apostrophes (never a space after them) but misses the right quotes that don't have a space after them. And the final nail in the coffin is this: automation makes the same exact mistake on every instance of whatever triggers the bug, whereas humans miss a few.

Now combine this with the following three facts: First, Rich Farmbrough has been shown to have run automation when banned from doing so in the past. Second, arbcom was ready to permanently ban him, but was searching for a set of restrictions that would keep him working but not using automation. Third, he freely agreed to the restrictions as worded whether or not we think he should have. BTW, has Rich Farmbrough himself ever claimed that the edit I analyzed was the result a cut and paste error?

He used automation. In defiance of his ban. Again. That's right, again - he has done this before. Yes, I do realize that some here think that the edit I analyzed can be explained by cut and paste errors, but there are good technical reasons to conclude that they are wrong. All you have to do to prove this to yourself is to load the "before" from the diff into your sandbox and then try to turn it into the "after" from the diff using just cut and paste and just making the kind of errors humans make. (If you try this, please keep a log so others can see exactly what you did). --Guy Macon (talk) 12:29, 28 March 2013 (UTC)

Thanks for your full answer. I'm sorry (esp. if it caused to you to waste your time): I wasn't clear. I should have included search and replace along with cut and paste as included in the functionality I'd think of as 'manual' (generic, everyday text editing of the sort any WP newbie might be familiar with) rather than 'automation' (bot scripts, wp-specific tools or whatever.) You think of things differently.

It doesn't seem surprising to me that different people have different gut feelings about this: the manual / automatic distinction doesn't actually seem to me one it's possible to draw very consistently in an online editing environment. I don't know whether or not you agree with that. Dsp13 (talk) 13:43, 28 March 2013 (UTC)

For me, the difference between cut and paste is that every paste is done manually, at a place of your choosing, under direct supervision. Find and replace is changing many instances with one click, without actually noticing where you change things and whether you get unexpected results. The chance of having multiple similar errors, which don't get noticed, is much higher with find and replace. Because there is no real, manual/visual supervision, becaues you let "the machine" find the places where your change will be made without the need for you to even look at what is being done, I consider such "find and replace" as being automated, and cut-and-paste as not being automated. (I know that you can do a supervised find-and-replace, where every instance of the replace needs a manual "OK", but that doesn't seem to be what happened here). Fram (talk) 14:02, 28 March 2013 (UTC)

Sure, being able to look at what you are doing, ceteris paribus, reduces the risk of errors. Where the text involved is short, and short in relation to its editing box (screen size helps!), cut-and-paste allows this. (The most common mistakes I find I make with cut-and-paste are mistakes in selecting the exact start / end of the text to copy, and pasting it twice by double-clicking. Pasting between different applications sometimes also causes me character encoding / white space problems.) The risks involved in search-and-replace for me similarly depend upon a range of factors: how structured the text is, how well I know it, etc. I don't think that one is always more risky than the other. Unfortunately even typing isn't (for me) risk-free: my inability to touch-type well means that every key is not always struck "at a place of choosing, under direct supervision": I often have to choose between seeing what I have written on the screen and seeing what I am typing with my hands.

So the promiscuous variety of my own tendencies to error make me doubt there's a fundamental distinction lurking here. And certainly using any of these everyday editing processes feels a million miles away from kicking off bots or specialized semi-automated processes which impact large numbers of pages. I've reread the page before and after the edit for which you reported Rich Farmbrough. It was a page needing quite a lot of copy-editing, and he did some but certainly not all of what was needed. He made some mistakes: the first of which - & the most consequential, since in the lede - was actually the typo 'cntroversial'. Copy-editing pages that need it is hard, and every bit helps: he left the page overall in a better condition than when he found it. Does he need especial praise for his edit? No. Does he need condemnation? No. Dsp13 (talk) 15:30, 28 March 2013 (UTC)

'Tis tragic. RF's currently the second most prolific editor on English Misplaced Pages. GoodDay (talk) 01:53, 27 March 2013 (UTC)

He wouldn't have been so prolific if he had followed the rules about not operating bots from his account instead of a bot account. --Guy Macon (talk) 05:53, 28 March 2013 (UTC)

• I agree with Guy Macon's assessment. Rich has once again littered articles with unthinking, automated errors, and the usual suspects are here pleading for leniency. I'm sorry, but no. The entire reason why we are here is leniency, and bluntly, a one year block is lenient. Rich has repeatedly shown that there is no restriction he will follow and that there is no possibility of him editing in a fashion that does not generate numerous mistakes that are left for others to clean up. Blocking is the only option left to protect the project from these damaging and time-wasting errors. Resolute 01:54, 27 March 2013 (UTC)

No one is stating Rich didn't make errors, the fact that he and his bots did more than 5 million edits, more than a million himself, attests to that fact. You can't do that many without making some mistakes and at that volume even 1% is a lot of errors. What should be considered is this. What message are we sending to editors? Don't edit too much because of you make mistakes, particularly a noticable percentage you'll be blocked; create bots at your peril; we allow editors to follow other editors every edit and submit them for punishment at the mere possibility of a violation of a vaguely written forced Arbcom determination? Because those are the messages being sent intended or otherwise. There is a growing perception that its better to just do an occassional edit and not get visible than to do to many and get blocked. Not a great thing. I would also note that if you have some experience at programming bots, programming in Lua or have the ability of answering the technical questions that Rich is asked to answer, for the benefit of the pedia, then it might be worthwhile to step up. Because whether you or others agree or want it, others need and want his help to make this project better. KumiokoCleanStart (talk) 02:03, 27 March 2013 (UTC)

Your entire argument is disingenuous, as usual. Everyone makes mistakes. Not everyone has a multi-year history of making the same mistakes over and over again, even after they are pointed out. And making them so often they were banned from using automated editing. And still making the same mistakes. Short blocks haven't worked. Edit restrictions haven't worked. Both have been tried many times. There's nothing else left. What kind of message do you want to give the community, Kumioko? That it is okay to be incompetent if you edit many, many articles? Resolute 02:16, 27 March 2013 (UTC)

Not to go to far off tangent but just out of curiousity what part of what I said was disingenuous? That he and his bots did aboout 5 million combined edits? That he made mistakes? That people still seek out his knowledge and experience? All of these are verifiable. I'm sorry if I am irritating you because I disagree with you and some of the others here. I simply do not have the zero defect mentality that has rooted itself in this case. I do not agree with the underhanded tactics that have been used against Rich. I do not agree with the extremely vaguely written Arbcom decision and I really don't like being called disengenuous because I do not follow your train of thought. If I said something disengenuous, I invite you to point it out so that I may clarify it with links. This may be surprising since I am not an admin elitist but I have actually been around long enough to be able to interpret the policies and am able to identify when they are being used abusively. And that has occurred on multiple occasions regarding Rich's case.KumiokoCleanStart (talk) 02:23, 27 March 2013 (UTC)

Your entire bit about "what message are we sending to editors" is the disingenuous part. I am well aware of the quantity of Rich's edits, many of which lacked quality, especially when he was running unauthorized automated tools that did useless things like change capitalization inside templates. You are dazzled by a number. I am interested in behaviour. And we have long since proven that neither restrictions nor short blocks work on this editor. Resolute 02:38, 27 March 2013 (UTC)

Well I guess we'll just have to agree to disagree then and I am not dazzled by a number. What does impress me is that after all that has happened Rich still wants to edit and make this place better. Yet here we are quibbling about how to justify a 1 year block over a few rather meaningless edits. KumiokoCleanStart (talk) 02:49, 27 March 2013 (UTC)

Rich Farmbrough's bots are buggy. Really buggy. Other bot operators manage to create bots with far fewer bugs, and they fix the bugs when they are found. Rich Farmbrough leaves the messes his buggy bots make for others to clean up. The other bot operators clean up any messes they make. Rich Farmbrough refuses to follow the rules for bot operators. Other bot operators follow them. You throw out phrases like "a few rather meaningless edits", but you cannot tell us how many are bad, what percentage are bad, or how bad they are -- you have never looked and are just guessing that it is "a few". You are making claims when you have no idea whether they are true. --Guy Macon (talk) 05:07, 27 March 2013 (UTC)

Its fair to say that some of Rich's code was buggy but part of the reason he was making more mistakes was because he was performing edits to actual articles. There are fairly few bots these days that perform edits to aricles. Part of this is due to the utterly inconsistent way that articles are created. There is no standards whatsoever which makes programming a bot to perform many tasks extremely difficult. Most of them create stats (like article assessments or updating the numbers for RFA's, close AFD's and perform other mundane tasks that have little impact to articles). That Rich was willing and able to create bots and bot tasks that would do actual edits and improvements to articles is a big part of why there are more errors and why many bot operators these days are unwilling to create new bots or bot tasks. Requests for bots used to be common. Now they are much less frequent because people are afraid of getting blocked for trying to help and make improvements. You are also wrong that I haven't looked. I am very familiar with the edits Rich did, I am very familiar with the errors and I even helpded to fix some. But most were blown out of proportion by a couple of editors who care more about ensuring minor edits don't get done than that improvements are being made. There willing to throw out the baby with the bathwater. Do I know specifically how many errors no, was it a lot, certainly just by volume of edits. But many of the "errors" that are being mentioned aren't errors, they are a differece of perspective of the need to do certain changes being called errors. KumiokoCleanStart (talk) 12:46, 28 March 2013 (UTC)

Just looking at pages on my watchlist, in the mainspace, where the last edit was by a bot, shows that there are still quite a few bots around in the mainspace; Addbot, Chartbot, Yobot, Emausbot,CommonsDelinker, AnomieBot, ‎Lowercase sigmabot, Xqbot, Cydebot, KLBot2, RussBot, .... Looking at recent changes also gives AvicBot, Citation bot, Cyberbot II and Avocatobot. There probably are a bunch more, these are just the ones I could easily find. Were there so much more mainspace bots active before the Rich Farmbrough bot blocks and ArbCom case? Fram (talk) 13:21, 28 March 2013 (UTC)

Certainly there are some left. Yobot your right, sometimes Anomiebot (but it mostly does minor things like dated Maintenance templates), most of the rest of those don't do things that really affect the articles content. The change, add or remove links, add signatures, archive talk pages and things of that nature. But nothing substantial. So yes you are right there are still some, but that number is dwindling and the number of people willing to do them is dwindling. Even the existing operators are leary of taking on new tasks. Yes there were a lot more, Rich did more than 25, 000 on an average month and Smackbot/HelpfulkPixieBot did more than that. So that's at least 50, 000 per month not done. Even if 10% were in error that's still 45, 000 not done. Some tasks have been taken over by other bots, many have not. KumiokoCleanStart (talk) 15:10, 28 March 2013 (UTC)

Kumioko, all the bots I listed above are editing articles in the mainspace, not "adding signatures, archive talk pages, ..." The rest of your post is mainly a rehash of old discussions of why stopping Rich Farmbrough from using bots and automation was bad. That's a separate discussion from the one on this AE enforcement. Let's stick to the latter please instead of having the same old discussion again at a time and place where it won't change anything anyway. Fram (talk) 15:32, 28 March 2013 (UTC)

• Surely nobody is claiming that a bot was used to make these changes. Or that the fixing of multiple broken references by repeated manual search and replace has damaged WP so seriously as to justify a one-year block. There needs to be a vote—for example, something like:
• (1) Whether repeated manual search and replace qualifies as "automated edits", and
• (2) Whether the fixing of multiple broken references by repeated manual search and replace for consistency (but leaving orphaned quote marks) has damaged WP so seriously as to justify a one-year block. LittleBen (talk) 06:28, 27 March 2013 (UTC)

• It appears that Farmbrough was blocked for edits to an article that was not in main space.  Here is the location where the edit was made.

Here is the only diff brought forward to AE in the section Diffs of edits that violate this sanction or remedy, and an explanation how these edits violate it:

• 2013-03-22T23:10:25‎ (Italics for book titles, not bold. Misplaced Pages is not a reference for Misplaced Pages MInor fixes)

Here is where the article was moved to mainspace:

• 2013-03-23T11:05:41‎ (Titodutta moved page Misplaced Pages talk:Articles for creation/Mohan Deep to Mohan Deep: Created via Articles for Creation (you can help!) (AFCH))

Unscintillating (talk) 03:33, 27 March 2013 (UTC)

And this is relevant how exactly? Articles at AfC are meant to be brought to the mainspace anyway, and his restrictions are not only for mainspace edits anyway, but for all his edits. Fram (talk) 07:54, 27 March 2013 (UTC)

The relevance is that there was no proof he was using automation, just speculation. In fact, the evidence shows he didn't even use find and replace because a couple of the potential changes on the page were missed. So that leaves some of us wondering how this was justified based on the evidence presented. I looked at his edits and although I wouldn't call all of them necessary I couldn't see anything that indicated they were automated. KumiokoCleanStart (talk) 10:24, 27 March 2013 (UTC)

Kumioko, your reply has nothing to do with the remark by Unscintillating and my response to him, so could you please post it somewhere else? If you have some questions for me or others, fine, go ahead, but don't insert them where they have no connection to what proceeded at all. It is disruptive. Fram (talk) 11:31, 27 March 2013 (UTC)

Just because you don't like the message Fram doesn't make it true. Unscintillating gave examples and you asked for the relavence. I gave you one. KumiokoCleanStart (talk) 12:46, 28 March 2013 (UTC)

• Note: I've undone AGK's archiving of the continued discussion below. NE Ent moved it here to talk, archive templates and all, and the combination was quite unfortunate. Restoring AGK's archive comment from the Misplaced Pages:Arbitration/Requests/Enforcement page: "I'm ending this dialogue. If you believe this enforcement action is unsupported by evidence or was procedurally flawed, you need to have Rich bring an appeal. Unsubstantiated criticism of the enforcing administrator is inappropriate and disruptive." AGK 11:08, 27 March 2013 (UTC) Bishonen | talk 12:21, 27 March 2013 (UTC).

This block is disputed by a powerless, regular, everyday Wikipedian, as having been closed too soon, without adequate time for response by anyone. 10:29 to 23:04 is just too short. The alleged offense simply does not rise to the level of an emergency response. See Talk. --Lexein (talk) 10:20, 26 March 2013 (UTC)

Point of order: arbitration enforcement actions are taken by a single administrator, so that administrator is not required to wait for discussion and approval of his proposed decision to take place before he can act—especially in unambiguous or simply enforcement requests. It might also be simpler if you take up your objections with Sandstein directly, and you might have more success if you argue against the actual decision rather than against the time he took to make the decision. Hope this helps, AGK 10:40, 26 March 2013 (UTC)

Ok. Thanks for that info. However, your encouragement aside, I'm seeing no evidence anywhere that communicating directly with Sandstein would work, or help. I would appreciate a link to some block decision Sandstein made being reversed by Sandstein solely on the basis of discussion; starting to hold my breath while I search now. I feel it's necessary to reign in and indeed slow-the-hell-down some of these so-called "simple" actions, based as they are on flawed foundational logic which goes directly against the Five Pillars. I'm of the opinion that wide consensus should rule here, rather than an individual administrator. Arbcom is needed, and AE, but not against Farmbrough, for this arguably manual edit. --Lexein (talk) 16:40, 26 March 2013 (UTC)

I think part of the problem is that 99% of all Arbitration enforcement actions are taken by that same admin. No one else seems to get involved and this particular admin always seems to lean towards the severe end of the spectrum. Its like watching a hanging for petty theft here every day. Murder, Guilty, hang' em! Stole a dollar? Hang him too! This process is just as flawed as the Arbitration process is but in different ways. Personally I think Sandstein got duped into the ban by Fram because Fram knew that Sandstein wouldn't really review the matter. Rich had an arbitration case on file and Fram found a couple of edits that could be argued were automated in some fashion so that's enough for Sandstein to justify it. Sandstein known for being a strict enforcer, a pitbull for the Arbitration committee, but not very meticulous in his research of the evidence. KumiokoCleanStart (talk) 10:39, 27 March 2013 (UTC)

I think we should have left the discussion open for a modicum of time. There was no rush to block, it could have been done in a way that avoids this appearance of it being 'summary judgement'. I happen to have no problem with the specific outcome, but it wasn't necessarily so obvious to preclude some more politeness. --Joy (talk) 12:38, 27 March 2013 (UTC)

I am curious as to what extended discussion would have accomplished. Basically, we have two camps here: One side notes that Rich has violated his editing restriction in a fashion that resulted in the very same pattern of errors that has been a constant and long-running problem with him. The other says no action should be taken because he has millions of edits. Policy vs. emotion. Emotion often wins at ANI if you have enough friends. Fortunately, AE is structured a bit more logically. Resolute 13:17, 27 March 2013 (UTC)

Except that's not an accurate reading of the situation at all. A more careful reading of the discussion would reveal more than a few editors who are saying no action (or less action) should be taken without even mentioning Rich's edit count. They are instead saying that either they don't believe Rich has violated his edit restriction, or - as in my case - that it is very borderline or a grey area. Or, they are saying that they agree that a minor infraction has taken place but that a year's block is too severe. Black Kite (talk) 14:06, 27 March 2013 (UTC)

I agree, Black Kite, and I agree with Joy that it would be better to let these discussions run for a bit longer. I've also let Sandstein know that I feel this way, but Sandstein disagrees and as AGK noted, there isn't any kind of requirement for group discussion or consensus. See Misplaced Pages:Arbitration/Requests/Case/Rich_Farmbrough#Enforcement_by_block; for example. ErikHaugen (talk | contribs) 17:05, 27 March 2013 (UTC)

I already said what it would have accomplished. The quality of being judicious and polite is a worthy goal in and of itself, especially in this kind of a serious process. It's not impossible that we would have just seen appeals to emotion in the extra time, but so be it, it wouldn't have hurt us a lot, certainly not as much as this kind of acrimony. --Joy (talk) 15:12, 27 March 2013 (UTC)

@ Resolute, So it seems then that the one with the most admins wins! KumiokoCleanStart (talk) 15:24, 27 March 2013 (UTC)

Not open long enough

It is my position that the AE request for Rich was not open long enough to allow the community to determine consensus for an action being taken. Specifically, the request was not open long enough to obtain comments from multiple uninvoled administarators. It has been pointed out that this is not required before taking action. But I believe it would be beneficial to the project if muliple uninvolved admins were allowed enough time to comment before any action is taken on any non-urgent AE request. Several people have mentioned this as being an issue and I believe allowing enough time for multiple comments by uninvolved admins would be helpful for a healthy project. I would like to know if others agree. Thanks. 64.40.54.134 (talk) 16:00, 27 March 2013 (UTC) my contribs for those unwilling to WP:AGF.

• In a word, no. Leaky Caldron 16:09, 27 March 2013 (UTC)
• Yes, the block was too severe, was submitted by an involved Admin/editor with a vendetta and there is obviously a number of people in the community who disagree, largely due to a poorly worded Arbcom decision that doesn't clearly define what automation is. Find and replace is not automation any more than using 4 tildes for the signature is. KumiokoCleanStart (talk) 16:27, 27 March 2013 (UTC)
• Others have stated that discussion is not the norm at AE. Fooled me: there's a space right there called "Statement by (username)". I don't like being lied to, when the evidence for room for discussion is right there in my face. --Lexein (talk) 16:35, 27 March 2013 (UTC)
• AE doesn't work that way. They may require discussion if it is unclear what action the original Arbitration requires, but otherwise any admin is empowered to take enforcement actions. AE is not ANI. Looie496 (talk) 16:39, 27 March 2013 (UTC)

If discussion is not necessary or is discouraged then why have the option? Why even allow the accused the opportunity to comment? That logic doesn't even seem to compute. Aside from that, look at Blocking. Multiple places in this policy indicate that extremely long blocks like 1 year or indefinate should be used by extreme exception. Not as the norm or for insignificant reasons such as this. KumiokoCleanStart (talk) 17:57, 27 March 2013 (UTC)

Discretionary sanctions are not community based. It is up to the individual discretion of an admin. There is no requirement for multiple admins to comment. There is not even a requirement that the action even be requested at AE from what I am aware of. All that is required is an official warning. AE is a place for people to propose sanctions be placed, and helps uninvolved admins get different opinions when they are reaching their decision, but an admin can do their own independent research without asking for the opinions of involved editors. The decision can of course be appealed. IRWolfie- (talk) 18:04, 27 March 2013 (UTC)

I'm not saying it's ideal, I'm just saying this is the way it is. IRWolfie- (talk) 18:05, 27 March 2013 (UTC)

Understood, now is as good of a time to start refining the process as any. If the process favors the accuser over the accused without allowing the accused an opportunity to comment or prove their innocence then the function of AE is broken. It's mean to enforce Arbitration decisions but it shouldn't do so blindly or in a vacuum. Disctretionary sanctions also doesn't give the Arbitration enforcer unlimited power to indefinately ban a member from the community nor does it allow them to do whatever they want. Its merely designed to give them some latitude in making determinations without being bogged down with red tape. It seems to me, that based on this decision of a 1 year block for nothing more than the suggestion that he might have violated his sanction, that the process need some review and fine tuning. Perhaps we should require more than circumstantial evidence to levy a block longer than say three months. I mean if the edit summary said it was done with AWB, twinkle or whatever that might be different. But in this case there is nothing more than 1 involved admin trying to block someone they have been hounding for years. KumiokoCleanStart (talk) 19:24, 27 March 2013 (UTC)

• Yes, quite correct. There is no rush. A longer discussion may well end up imposing the same sanction, but that was not enough time for enough uninvolved admins to participate. Black Kite (talk) 18:58, 27 March 2013 (UTC)
• I agree that the discussion was closed hastily. The issue was clearly non-urgent, and there seems (to say the least) disagreement about whether search and replace, cut and paste are 'manual', 'automatic' and so forth. Dsp13 (talk) 02:32, 28 March 2013 (UTC)

• This all looks fine to me. The sanctions imposed by ArbCom were pretty clear cut, and Rich appears to have breaching them. He chose to not post a statement in the AE thread countering Fram's analysis, and Sandstein's reasoning is sound. Nick-D (talk) 10:36, 28 March 2013 (UTC)

Proposal for voting on

• According to the block notice, this block can be amended or overturned, "following a clear, substantial, and active consensus of uninvolved editors at a community discussion noticeboard (such as WP:AN or WP:ANI). If consensus in such discussions is hard to judge or unclear, the parties should submit a request for clarification on the proper page".

Proposal for voting on:  Blocks are supposed to be preventive—to prevent disruption or damage to Misplaced Pages—not punitive. Overlooking eight orphaned single quotes has not caused serious disruption or damage to Misplaced Pages, so a one-year block is excessive, disproportionate punishment, and should be reconsidered—surely a block of no more than one month is more appropriate. (Support or Oppose, please). LittleBen (talk) 15:15, 28 March 2013 (UTC)

• Support. Dsp13 (talk) 15:57, 28 March 2013 (UTC)
  • per Sandstein and Kumioko, I've withdrawn my vote here - though nothing either of them has said has changed my mind about the merits of the case. (In Sandstein's retelling of the Arbcom ruling, the language has accreted new epicyclic flourishes: now it's not just that search and replace is to be regarded as included in automation, but that RF must (per impossibile) demonstrate convincingly that he understands the original decision to have implied this; it's apparently not Sandstein's opinion, but the factual basis for his opinion, that RF's recent edits 'clearly' appear as automated; and, mostly oddly, only "typing characters into the edit box one by one" is allowed. (Is using the Shift key in combination with another keypress allowed? Two clicks on a mouse to highlight a word before typing over it? Three clicks to highlight a line?) Dsp13 (talk) 18:19, 28 March 2013 (UTC)
• This proposal is made in the wrong forum. This is a talk page, not a community noticeboard such as WP:AN or WP:ANI. Any consensus resulting from this discussion is therefore not a sufficient basis for overturning an arbitration enforcement action. Additionally, as long as the blocked user himself does not signal his opposition to the block by appealing it, there is no basis on which to even begin discussing its reduction.

  On the merits, as the blocking administrator, I oppose reducing the block, and I will not lift it as long as Rich Farmbrough does not very convincingly communicate that he understands that he is forbidden by decision of the Arbitration Committee from making any automated edit whatsoever, including by search and replace, and that he will scrupulously observe this restriction until it is lifted by the Committee.

  This proposal proceeds from a mistaken assumption by putting the duration of the block in relation to the damage (if any) done by the edits at issue. These two factors are unrelated. I would also have had to block Rich Farmbrough if his edits had been clearly beneficial. The reason for the block is not any harm done by his edits, but the fact that they violate an editing restriction by the Arbitration Committee, which prohibits all automated editing whatsoever, and explicitly allows only typing characters into the edit box one by one. As WP:BAN makes clear, "bans apply to all editing, good or bad". Therefore, undoing an enforcement action on the basis of consensus for the argument that the edits at issue were not harmful would violate the Arbitration Committee's binding restriction, and would expose whoever reduced the block to the risk of sanctions for doing so.

  The only thing that an appeal discussion can review is whether the edits were in fact automated and therefore violated the restriction. Because Rich Farmbrough did not contest that he used automation to make the edits at issue, and because the remedy specifies that "any edits that reasonably appear to be automated" (which the edits at issue clearly do) "shall be assumed to be so", I am of the opinion that no convincing argument can be made that Rich Farmbrough did not violate his editing restriction.  Sandstein  16:52, 28 March 2013 (UTC)

Although I hesitate in opening up yet another forum with a Rich F discussion I actually agree with Sandstein that this is not the best place for the discussion. This venue is not really the best place to get to a meaningful and unbiased consensus on the merits of this block and/or whether what he dis should or should not be considered automation.

I do think that last half of the paragraph is complete bullshit though. Just because a blocked user doesn't somehow find a way to argue the block doesn't mean they agree with it. Just because the user didn't immediately rush to reply to the comments at AE does not mean they agree with it. Just because the user did not argue what they did was or wasn't automation does not mean it is. All of these arguments are complete nonsense and an utter and complete lack of good faith. In cases like this where there is some reasonable doubt, we should be assuming good faith, not assuming the user was doing something to violate a policy. That's just the wrong attitude to have and a pretty piss poor justification for a one year block. KumiokoCleanStart (talk) 17:14, 28 March 2013 (UTC)

I would have to disagree with your statement "The only thing that an appeal discussion can review is whether the edits were in fact automated and therefore violated the restriction." A community discussion can also determine that, although the restriction was indeed violated, a one-year block is inappropriate, and should be reduced to a shorter duration. Otherwise, that would allow a single administrator to unilaterally impose any block duration desired without the approval of either the community or Arbcom. -- King of ♥ ♦ ♣ ♠ 18:41, 28 March 2013 (UTC)

Yes, you're of course right that the block duration can also be the subject of an appeal discussion. (Although the AE procedures do in fact authorize admins to impose blocks of up to a year without anybody's prior approval. The question is rather who may review such blocks.)  Sandstein  19:02, 28 March 2013 (UTC)

Protected

I've just protected this for 3 hours due to edit warring over the inclusion of a report. I hoped by doing so there would be no need for anyone to be blocked. I realize this is out of the norm, so any AE regulars are free to revert me if they wish. Mark Arsten (talk) 22:41, 28 March 2013 (UTC)