Wikipedia:Arbitration/Requests/Case/Magioladitis/Proposed decision

Proposed motions
Arbitrators may place proposed motions affecting the case in this section for voting. Typical motions might be to close or dismiss a case without a full decision (a reason should normally be given). Suggestions by the parties or other non-arbitrators for motions or other requests should be placed on the /Workshop page for consideration and discussion.

''Motions require an absolute majority of all active, unrecused arbitrators (same as the final decision). See Arbitration Committee/Procedures.''

Template
1)

{text of proposed motion}


 * Support:


 * Oppose:


 * Abstain:


 * Comments:

Proposed temporary injunctions
A temporary injunction is a directive from the Arbitration Committee that parties to the case, or other editors notified of the injunction, do or refrain from doing something while the case is pending. It can also be used to impose temporary sanctions (such as discretionary sanctions) or restrictions on an article or topic. Suggestions by the parties or other non-arbitrators for motions or other requests should be placed on the /Workshop page for consideration and discussion.

Four net "support" votes needed to pass (each "oppose" vote subtracts a "support") 24 hours from the first vote is normally the fastest an injunction will be imposed, unless there are at least four votes to implement immediately. See Arbitration Committee/Procedures.

Template
1)

{text of proposed orders}


 * Support:


 * Oppose:


 * Abstain:


 * Comments:

=Proposed final decision=

Purpose of Wikipedia
1) The purpose of Wikipedia is to create a high-quality, free-content encyclopedia in an atmosphere of camaraderie and mutual respect among contributors. Contributors whose actions are detrimental to that goal may be asked to refrain from them, even when these actions are undertaken in good faith; and good faith actions, where disruptive, may still be sanctioned.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:28, 10 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:00, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:13, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 22:56, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * Doug Weller talk 16:04, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:17, 11 February 2017 (UTC)
 * Mkdw  talk 22:54, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:29, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:43, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:53, 23 February 2017 (UTC)
 * Drmies (talk) 04:51, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Automated and semi-automated editing
2) Fully automated bot editing and semi-automated editing scripts perform an important and valuable function on Wikipedia. To facilitate the regulation and coordination of such editing, the community has a long-established bot policy and a Bot Approvals Group responsible for reviewing potential bot operators' requests for bot approval "from a technical and quality-control perspective".


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:28, 10 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:00, 10 February 2017 (UTC)
 * Indeed they have. kelapstick(bainuu) 10:14, 10 February 2017 (UTC)
 * (Although as I understand it BAG is involved with bots, not with scripts.) Newyorkbrad (talk) 22:57, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * Doug Weller talk 16:04, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:18, 11 February 2017 (UTC)
 * Mkdw  talk 22:54, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:29, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:43, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:53, 23 February 2017 (UTC)
 * Drmies (talk) 04:52, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Bot policy
3) According to the bot policy, approved bots should:


 * be harmless
 * be useful
 * not consume resources unnecessarily
 * perform only tasks for which there is consensus
 * carefully adhere to relevant policies and guidelines
 * use informative messages, appropriately worded, in any edit summaries or messages left for users.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:28, 10 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:00, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:14, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 22:57, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * Doug Weller talk 16:04, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:18, 11 February 2017 (UTC)
 * Mkdw  talk 22:55, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:29, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:43, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:53, 23 February 2017 (UTC)
 * Drmies (talk) 04:52, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

"Cosmetic" or inconsequential edits
4) According to the bot policy, "Cosmetic changes (such as many of the AWB general fixes) should be applied only when there is a substantive change to make at the same time." According to the AWB Rules of Use, AWB users are instructed not to "make insignificant or inconsequential edits", defined as "An edit that has no noticeable effect on the rendered page is generally considered an insignificant edit. If in doubt, or if other editors object to edits on the basis of this rule, seek consensus at an appropriate venue before making further similar edits."


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:28, 10 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:00, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:15, 10 February 2017 (UTC)
 * Although there can be disagreement over specific application of the general principle, such as where an edit has little or no effect on the seen page but improves the page's readability for screenreaders. Newyorkbrad (talk) 22:58, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * Doug Weller talk 16:04, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:20, 11 February 2017 (UTC)
 * Mkdw  talk 22:56, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:29, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:43, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:53, 23 February 2017 (UTC)
 * Drmies (talk) 04:53, 2 March 2017 (UTC)
 * With what appears to be the obvious accessibility exception. Keilana (talk) 04:59, 9 March 2017 (UTC)
 * Oppose:


 * Abstain:


 * Comments:
 * Re NYB above: as far as I understand, accessibility fixes (like those required for screenreader use) are as a rule not considered cosmetic - but there is a lot of room for ambiguity and misunderstanding over exactly what changes need to be made to correct these issues (consider how often people get WP:LISTGAP wrong in long talk-page threads). There are also issues that only have a visual effect in older browsers, or that "look" fine but bork up the HTML, etc. Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * I agree with OR's interpretation about accessibility fixes.  DGG ( talk ) 22:20, 11 February 2017 (UTC)

Collegiality
5) Wikipedia is a project founded on the principles of collaboration and consensus. Even when an editor holds a reasonable belief that an edit or set of edits has consensus, it is collegial to pause when presented with reasonable objections, to take critical feedback into consideration, and to make reasonable efforts to avoid repeatedly making the same mistake. This behavior is particularly important when editing at high volume, whether in an automated or semi-automated fashion. Likewise, in a large collaborative project it is inevitable that some types of edits that irritate or inconvenience some editors will nevertheless gain consensus; it is collegial to accept this inevitability and avoid repeatedly making the same objections.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:28, 10 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:00, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:15, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 22:59, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:14, 11 February 2017 (UTC)
 * Doug Weller talk 16:04, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:20, 11 February 2017 (UTC)
 * Mkdw  talk 22:57, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:29, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:47, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:53, 23 February 2017 (UTC)
 * Drmies (talk) 04:54, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Locus of dispute
1) This case centers on the conduct of and his bot, particularly in relation to allegations of violations of the bot policy and the AutoWikiBrowser rules of use and associated expectations of accountability.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * Yes, although if Wikipedia worked on precedents, I could see this being used as one. kelapstick(bainuu) 10:17, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 22:59, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Doug Weller talk 16:18, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:21, 11 February 2017 (UTC)
 * Mkdw  talk 22:58, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:48, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 04:54, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Magioladitis is an experienced editor and bot operator
2) has been an active Wikipedia editor since 2006, and an administrator since 2008. He has operated a bot,, since 2008. He joined the Bot Approvals Group in June 2014 and stepped down in December 2016 following an unsuccessful reconfirmation request. He is one of the developers of the AutoWikiBrowser (AWB) software, which is widely used for both semi-automated and fully automated editing on Wikipedia. Magioladitis uses AWB in fully automated mode on the Yobot account and in semi-automated mode on his main account.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:19, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:00, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:21, 11 February 2017 (UTC)
 * Mkdw  talk 22:59, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:48, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 04:55, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

There is uncertainty about the scope of Yobot's BRFAs
3) was originally authorized in 2008 for a narrowly defined task. It has since acquired a large number of diverse tasks. The most controversial has been task 16, which covers WP:CHECKWIKI fixes and whose exact scope has been disputed (e.g., , ). Related questions have been raised about other bots with similarly scoped BRFAs.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * (Changed "BRFA" to "BRFAs" in header.) Newyorkbrad (talk) 23:02, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * kelapstick(bainuu) 06:26, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:22, 11 February 2017 (UTC)
 * Mkdw  talk 22:59, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:48, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:01, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Magioladitis has received feedback about Yobot's edits
4) Since Yobot's first task was approved in 2008, a number of complaints have been made in various fora about the bot's edits (see case request). Many of these issues have reflected objections to edits that had little or no effect on the rendered page (so-called "cosmetic" edits), such as bypassing template redirects or regularizing wikicode syntax. Magioladitis' main account was blocked in 2010 due to a series of edits bypassing template redirects and unblocked following a review at AN. His main account has since been blocked four more times for similar reasons, most recently in December 2016.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:19, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:05, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:22, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:01, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:48, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:01, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Yobot has been blocked and unblocked numerous times
5) has been blocked 19 times beginning in 2009. In seven of those cases, Magioladitis himself unblocked the bot account, usually with the explanation that the problem causing the unwanted bot edits had been resolved. Some, but not all, of these blocks were related to cosmetic edits.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:20, 10 February 2017 (UTC)
 * The apparent length of the block log arises partly from having a single bot account performing 16 different tasks, rather than 16 bot accounts each performing one task. But it's still a heck of a lot of blocks. Newyorkbrad (talk) 23:07, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 * (with NYB's caveat)  DGG ( talk ) 22:23, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:05, 11 February 2017 (UTC)
 * Agreed with NYB's and Mkdw's comments, though this is still relevant. GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:48, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:02, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:
 * Every editor and/or bot will presumably have their own caveats based upon their activity and editing type. No two editors or bots are identical and I don't think this finding precludes this and in fact goes out of its way to clarify this point. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:05, 11 February 2017 (UTC)

Magioladitis has performed cosmetic and bot-like edits using his main account
6) In addition to his work as a bot operator, Magioladitis uses AWB as a semi-automated tool on his main account. On several occasions, he has made "cosmetic" edits using his main account and has made series of edits at bot-like speed (e.g., ). He was briefly subject to a community restriction prohibiting him from making semi-automated edits on his main account, imposed in January 2016 as an unblock condition and removed four days later following a re-block of his account. The matter was subsequently discussed at AN.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * kelapstick(bainuu) 10:21, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:08, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:23, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:06, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 13:54, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:04, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

Yobot's authorizations have been revoked
7) Following a request to modify Yobot's bot authorization, Yobot's authorization to perform bot edits was revoked on February 1. The Yobot account has been unblocked, and Magioladitis has begun re-filing BRFA requests to allow new reviews of the tasks he wishes to continue.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:10, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * kelapstick(bainuu) 06:26, 11 February 2017 (UTC)
 * Doug Weller talk 16:06, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:24, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:07, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 13:54, 20 February 2017 (UTC)
 * -- Amanda  (aka DQ) 22:57, 23 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:04, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

"General fixes" and the Checkwiki project
8) Two distinct systematic efforts exist to manage a variety of minor errors, formatting problems, accessibility issues, wikicode syntax irregularities, and other inconsistencies across Wikipedia. AWB supports a set of community-curated "general fixes", documented at AutoWikiBrowser/General fixes, which can be applied alone or in conjunction with other AWB tasks in either semi- or fully automated mode. Separately, WikiProject Check Wikipedia (Checkwiki) maintains a numbered list of errors that project members aim to correct throughout the project. Yobot is one of seven bots working on a subset of checkwiki-defined errors. Some project members, including Magioladitis, also perform semi-automated error correction using their main accounts. Both of these systematic efforts encompass large and diverse lists of errors and are primarily curated by small groups of contributors with technical experience and interest.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:11, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * kelapstick(bainuu) 06:25, 11 February 2017 (UTC)
 *  DGG ( talk ) 22:24, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:08, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * Cas Liber (talk · contribs) 13:55, 20 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Drmies (talk) 05:06, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:

"Cosmetic" edits
9) Both AWB's general fixes and Checkwiki's error lists include items broadly agreed to be cosmetic and inconsequential. However, as the case proceeded, it became clear that there is no widespread shared understanding of the exact nature of a "cosmetic" or inconsequential edit, and that edge cases are frequently misunderstood, disputed, or ambiguous (e.g., , , ).


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * Newyorkbrad (talk) 23:16, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * kelapstick(bainuu) 06:25, 11 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:10, 11 February 2017 (UTC)
 * GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 *  DGG ( talk ) 07:18, 19 February 2017 (UTC)
 * Cas Liber (talk · contribs) 13:55, 20 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 21:53, 24 February 2017 (UTC)
 * Slightly reluctant support as a agree with Amanda, but it's accurate. Doug Weller  talk 21:39, 25 February 2017 (UTC)
 * With Doug, I waver here. Drmies (talk) 05:09, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:
 * While I agree there is confusion to the general idea, it's clear to me that Magioladitis was perfectly aware of what should and shouldn't have been done. (Ramaksoud2000's evidence) -- Amanda  (aka DQ) 23:02, 23 February 2017 (UTC)


 * Comments:
 * Following discussion on the workshop, it occurs to me that the word "cosmetic" as used in the policy is misleading. One would think that a "cosmetic edit" is one that affects only the appearance of the output page, yet we define a "cosmetic edit" paradoxically as one that does not affect the appearance of the output page, but only the markup page. Perhaps introducing a new term such as "non-visible edit" would mitigate the confusion, or "non-output-affecting" edit if we want to take the auditory aspects into account as well. As for the substance of the policy, isn't the operational definition of a "cosmetic edit" really "an edit that isn't useful enough to be worth making if it's the only change on the page"? Newyorkbrad (talk) 23:16, 10 February 2017 (UTC)
 * Agree that terminology could be clarified, though the suggested alternatives pose their own issues. Re the policy substance, yes that's my interpretation of it too, and the policy would benefit from this kind of simpler wording. -- Euryalus (talk) 23:26, 10 February 2017 (UTC)
 * I think the intended meaning is the "cosmetics" of the wikitext - things like replacing a wiki_link_with_underscores with the more natural-looking wiki link with underscores or simplifying links piped to themselves . However, the meaning seems to have been generalized to "any edit whose effect is smaller than a nebulously defined 'smallest acceptable edit unit'". It's hard to replace except with a new jargony term, I suspect - for example, "non-output-affecting" still isn't quite right; a common screenreader-friendly edit is to make WP:LISTGAP fixes, which don't affect what the page looks like to most people, but do affect the page's HTML (ie, the output). Opabinia regalis (talk) 5:37 am, Today (UTC+0)
 * I agree with OR that this is the intended meaning; it seems the only consistently applicable meaning to the extent that I think we should say so.  DGG ( talk ) 18:07, 13 February 2017 (UTC)

Problems with cosmetic edits
10) Cosmetic or inconsequential edits can be problematic because they clog watchlists and page histories, incur time costs for performing and reviewing trivial tasks, and may reduce the likelihood of detecting vandalism and damaging edits when performed using bot-flagged accounts. A phabricator task (T11790) has been open since 2007 seeking a change in watchlist behavior that would mitigate some of these problems.


 * Support:
 * Euryalus (talk) 05:54, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:30, 10 February 2017 (UTC)
 * As I mentioned on the talk page, I included this in order to document that the concerns are substantive and have practical effects, rather than just being complaints about policy violations. (If you violate a policy but nothing bad happens, did it make a sound? ;) It's not intended as an endorsement of those concerns - in fact, I think there were good points made on the workshop about the fact that most possibly-damaging edits "hidden" from watchlists this way will still likely be reviewed by some other means, e.g. Huggle users, so the costs of the watchlist issue aren't as high as you might think at first glance. Opabinia regalis (talk) 05:37, 11 February 2017 (UTC)
 * Is it always problematic? No. Can it be? Yes. Can this problem be overcome by more effort by editors? Yes, but that's not necessarily the debate being proposed by this fact. It's my nature to more heavily scrutinize minor and even "cosmetic" editors by editors because of the human element involved. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 23:17, 11 February 2017 (UTC)
 * Reaffirming my support with the revised changes. I think it's better breaking it into two. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 05:36, 25 February 2017 (UTC)
 * With emphasis on "can be", I suppose. There are ways to mitigate this, but I would agree that these are all concerns users have raised regarding this kind of edit. GorillaWarfare (talk) 23:42, 18 February 2017 (UTC)
 * With the understanding, shared by some of my colleagues above and below, that "can be" does not equate to "are always". Perhaps "can be" would better be read as, or changed to, "are treated by policy as" and with a link to the policy section. Newyorkbrad (talk) 20:48, 19 February 2017 (UTC)
 * Links added. IMO "are treated by policy as..." is subtly different from "can be" as a statement of fact, because neither the bot policy section nor the AWB rules of use explicitly indicate the underlying practical rationale. I suppose we could add links here to people's statements specifically about the topic. IMO this background is relevant to document in part because it's not intuitive if you don't use your watchlist in ways that expose the problem. Opabinia regalis (talk) 05:49, 20 February 2017 (UTC)
 * broadly speaking...though I also broadly agree with NYB and kelapstick below. Cas Liber (talk · contribs) 11:05, 22 February 2017 (UTC)
 * on balance DGG ( talk ) 09:21, 25 February 2017 (UTC)
 * Echoing GW's "can be". Yes, it can. Yes, editors should look farther than their nose is long (as the Dutch would say--i.e., not just look at the last edit), but it seems clear to me that such edits can make life a bit more complicated. Drmies (talk) 05:14, 2 March 2017 (UTC)
 * Per GW and OR. Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:
 * This is a cost-benefit analysis. Do bot edits clog watchlist, sure. Do they incur a time cost for reviewing, yes. Do they reduce likelihood of detecting vandalism, only if editors don't check the pre-bot history. Fortunately Bot edits are specifically flagged. The likelihood is only reduced if people who have pages watchlisted looking for vandalism only go by the last edit. Any subsequent edit to a page after vandalism has the same effect, either by bot or human editor (which is why we have vandalism that sits for months to years). The benefit that we have with human editors is that generally their user or user talk pages are redlinked. Additional due diligence for all editors for those who watch recent changes, or revert solely based on watchlists, to quickly check page history when they see something suspect would be better than preventing bot edits which some editors consider inconsequential. --kelapstick(bainuu) 10:33, 10 February 2017 (UTC)
 * While this is factually correct, I strongly oppose this. It completely misrepresents people's ability to solve the issue. A simple checking of the box "Expand watchlist to show all changes, not just the most recent" in Preferences -> Watchlist -> Advanced options would give them what they need and also help in patroling the rest of the edits they don't even look at. So to selectively claim that not showing the most recent non-bot edit is an issue, but then not even look at the rest of the potentially vandalized diffs by not checking that box is hypocritical. Even better, there is an option where they are collapsed for the user so it isn't a wall of text! -- Amanda  (aka DQ) 23:18, 23 February 2017 (UTC)
 * Per Amanda. If anything should be done the default should be changed, as it makes earlier vandalism easier to miss. I've had it checked for years. Doug Weller  talk 15:00, 24 February 2017 (UTC)
 * Abstain:


 * Comments:
 * Agree with Kelapstick that this is a cost-benefit assessment, support is related to the phrasing that it "can" be problematic (rather than "is") Remedies would necessarily be proportionate to the size of the problem they address. -- Euryalus (talk) 23:19, 10 February 2017 (UTC)
 * Responding to Kelapstick, I had never really thought about the issue before the case, but I think the problem is this&mdash;just to make sure I am understanding correctly, I'll use myself as an example. I have a few thousand articles on my watchlist. If I skim my watchlist and I see the most recent edit on an article is suspicious, I can check it out, and if it's vandalism I'll revert it, and if it's good-faith but poorly-written I can rewrite it. But I have my watchlist set to skip bot edits, so if the most recent edit on an article is a bot edit, I won't see the bot edit&mdash;and I won't see the most recent non-bot edit either. And if I had my watchlist set to include bot edits, I would see the bot edit, but I still wouldn't see the most recent non-bot edit. The intuitive fix would be to add a watchlist option of "list the most recent edit that's not by a bot"&mdash;but that wouldn't quite work either, because if the second-last edit was a vandal edit, and the last edit was Cluebot reverting the vandal, then "the most recent non-bot edit" would show as the vandalism, and I'd wind up trying to revert an edit that was already reverted. So the real fix would have to be "list the most recent non-bot edit that hasn't been reverted," and there seems to be some disagreement at the Phabricator page about how that would be defined. The suggestion of "quickly check page history when they see something suspect" makes sense when there is something suspect to see, but if I see that a bot made a routine edit on an article, there's no way to tell whether there was a problem in the history before that point. Newyorkbrad (talk) 23:26, 10 February 2017 (UTC)
 * Yes NYB that is correct. --kelapstick(bainuu) 06:23, 11 February 2017 (UTC)
 * I think the grouping you see in the expanded watchlist is actually the effect of a different setting: Preferences -> Recent changes -> Advanced options -> "Group changes by page in recent changes and watchlist". In any event, the ungrouped, unexpanded display where the problem occurs is the default option. Opabinia regalis (talk) 08:48, 24 February 2017 (UTC)
 * Regardless of where the option is, it absolutely exists. In it's current state, the FoF just validates one side of the argument, and proceeds to blame the issue on an inactioned phabricator task. Do I agree the feature could be actioned on? Ya. But to blame the entire issue on the phabricator task is a biased view. -- Amanda  (aka DQ) 10:02, 24 February 2017 (UTC)
 * Maybe in this FOF the phabricator sentence needs to be removed and proposed only as a remedy? <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 17:47, 24 February 2017 (UTC)
 * It seems like more is being read into this FoF than was intended, but I can't quite put my finger on why. DQ, which side does it validate - the people who don't like cosmetic edits, by spelling out their argument, or the people who don't think they're a problem, by indicating that there's a technical fix for some of those issues? The main reason this FoF is here is to indicate that those concerned about cosmetic edits are pointing to a practical issue, not just one of insufficient respect for authoritah ;)
 * As for ability to address the problem, I don't think the suggested settings really apply to the specific use case of a large watchlist that is monitored frequently and primarily used for anti-vandalism, as described in Materialscientist's preliminary statement. Having to click a dropdown for every page to see the pre-bot edits is not a viable way of doing the activities he describes. Anyway, this is meant to be a descriptive statement about the concerns, not a judgment about the best way to address them.
 * In any case, I'm hesitant to propose new stuff at this late stage, but see FoF 11 below. Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)

Relevant technical proposals
11) A phabricator task (T11790) related to the appearance of bot edits on watchlists has been open since 2007.


 * Support:
 * Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)
 *  DGG ( talk ) 21:27, 25 February 2017 (UTC)
 * Newyorkbrad (talk) 15:35, 26 February 2017 (UTC)
 * Without judgement on the effectiveness or alternate means available to the community. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:39, 27 February 2017 (UTC)
 * GorillaWarfare (talk) 01:44, 1 March 2017 (UTC)
 * Drmies (talk) 05:16, 2 March 2017 (UTC)
 * True as a general statement. -- Euryalus (talk) 19:58, 5 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)
 * Kirill Lokshin (talk) 22:54, 12 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:
 * In view of the comments on the phab links in FoF 10 and remedy 2, I'm splitting this out as a standalone FoF and striking the references to those links elsewhere. I do think it is useful to link these items (particularly the first one) somewhere in the decision, for documentation purposes. Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)

Proposed remedies
Note: All remedies that refer to a period of time, for example to a ban of X months or a revert parole of Y months, are to run concurrently unless otherwise stated.

Community encouraged to review common fixes
1) The community is encouraged to carefully review the lists of items in AWB's "general fixes" and the Checkwiki project's list of errors to determine whether these items are truly uncontroversial maintenance changes. A suggested approach would be classifying existing fixes as cosmetic or non-cosmetic and thereby identifying fixes that should be ineligible to be applied alone. The groups who currently invest their efforts in maintaining these lists are encouraged to improve their change management practices by soliciting broader community input into the value of adding proposed new items to the lists, and specifically to make their proposals accessible to members of the community who are not bot operators or whose interests are non-technical.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * Euryalus (talk) 00:25, 11 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * The community may also wish to consider whether to recommend that the Phabricator task request described elsewhere on this page, which would mitigate some of the adverse effects of "cosmetic edits," should be given a greater priority. Newyorkbrad (talk) 23:29, 11 February 2017 (UTC)
 * kelapstick(bainuu) 09:38, 13 February 2017 (UTC)
 *  DGG ( talk ) 18:08, 13 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:15, 22 February 2017 (UTC)
 * Sure, noting the words "encouraged" and "suggested". Ks0stm  (T•C•G•E) 21:54, 24 February 2017 (UTC)
 * As a suggestion, I think any consideration the community is willing to endeavour would be welcome. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 22:12, 27 February 2017 (UTC)
 * GorillaWarfare (talk) 01:44, 1 March 2017 (UTC)
 * Doug Weller talk 13:53, 1 March 2017 (UTC)
 * Drmies (talk) 05:17, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:
 * There are a few things I can't wrap my head around right now on this, and I think it's me, not the wording. I may revisit this, but if there is seemingly no issue, i'm not going to hold it up. -- Amanda  (aka DQ) 11:22, 24 February 2017 (UTC)


 * Comments:

Community encouraged to review policy on cosmetic edits
2) The community is encouraged to hold an RfC to clarify the nature of "cosmetic" edits and to reevaluate community consensus about the utility and scope of restrictions on such edits. Technical feedback may be provided at T11790 or T127173. The committee notes that an RfC on this topic is currently under development.


 * Support:
 * Euryalus (talk) 07:32, 9 February 2017 (UTC)
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * I think clear definition on this is necessary, as it appears to be a bone of contention. --kelapstick(bainuu) 06:28, 11 February 2017 (UTC)
 * Same comment as on remedy 1. Newyorkbrad (talk) 23:30, 11 February 2017 (UTC)
 *  DGG ( talk ) 18:27, 14 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:16, 22 February 2017 (UTC)
 * With the revised changes. Thank you OR. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 05:34, 25 February 2017 (UTC)
 * -- Amanda  (aka DQ) 20:36, 25 February 2017 (UTC)
 * Doug Weller talk 21:44, 25 February 2017 (UTC)
 * GorillaWarfare (talk) 01:46, 1 March 2017 (UTC)
 * Ks0stm (T•C•G•E) 03:25, 1 March 2017 (UTC)
 * Drmies (talk) 05:17, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:
 * On the basis of the phabricator ticket inclusion per my strong oppose above. -- Amanda  (aka DQ) 23:48, 23 February 2017 (UTC)
 * # Ditto, plus Headbomb's comment here on including the phabricator tickets. Doug Weller  talk 15:02, 24 February 2017 (UTC)


 * Abstain:


 * Comments:
 * I take Headbomb's point on the talk page that the purpose of linking to phabricator here isn't all that clear. (FWIW, the point was not to encourage people to go use phab to talk about local policy, which would be dumb, but rather that a decision on the proposed technical changes might affect the direction of the recommended RfC.) I've struck the phab sentence and proposed a new FoF 11 that simply links the relevant tasks. Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)

Developers encouraged to improve AWB interface
3) While the Arbitration Committee has no direct authority over the volunteer developers of open-source tools, we encourage the AWB developers to carefully consider feedback gathered in this case in order to use technical means to avoid problematic edits more effectively.


 * Support:
 * Euryalus (talk) 06:46, 9 February 2017 (UTC) Moved to oppose. See reasoning there. -- Euryalus (talk) 04:38, 1 March 2017 (UTC)
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * kelapstick(bainuu) 06:28, 11 February 2017 (UTC)
 * I have no personal knowledge to what extent a "technical" fix here is feasible, but the possibility certainly warrants evaluation. Newyorkbrad (talk) 23:32, 11 February 2017 (UTC)
 * I don't know if this will help, but I don't see any downside to it. Doug Weller  talk 15:34, 12 February 2017 (UTC)
 *  DGG ( talk ) 18:08, 13 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:16, 22 February 2017 (UTC)


 * Oppose:
 * I will be the first to admit I don't know everything about AWB, even being a bot developer. While I can understand there are specific proposals below, there are many options within AWB, and I don't think we understand them all as a group to be making specific recommendations, and would need some evidence from the uninvolved developers directly that this feature request is not already handled in the settings. I see many options like: Do not apply MOS fixes, Several skip options, and Even more skip options. Also I see there is a way to do customized genfixes. Overall as a developer I would hate to receive a "fix all the issues with this checkbox" bug report, which I feel is what this comes down to. -- Amanda  (aka DQ) 11:03, 24 February 2017 (UTC)
 * This issue has not been caused by any fault in AWB, which already has settings to skip genfixes and other specific types of edits. The important thing is that the community decide what is and is not a cosmetic edit; until then, AWB developers can't do much to mitigate this issue. GorillaWarfare (talk) 01:48, 1 March 2017 (UTC)
 * This seemed a harmless enough request, which is why I was an early support. However Deltaquad and GW make good points, from a position of more detailed knowledge on the topic. On which basis, while sympathetic to the intention of the remedy am shifting to oppose per their arguments above. I note in passing that the case does not in any way hang on this remedy. -- Euryalus (talk) 04:38, 1 March 2017 (UTC)
 * Per my colleagues - AWB is a fine tool, the problem is the behavior under review here, which I do not believe would be prevented by any technical solution. However, I have approximately negative technical knowledge, so I defer to my colleagues on that judgment. Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Abstain:
 * I don't feel well enough versed in the AWB interface to feel like I can adequately judge this remedy's merit. Ks0stm  (T•C•G•E) 21:57, 24 February 2017 (UTC)
 * Ks0stm beat me to it but I feel the same way. I'm sure they're going to review this case with much interest but I'm sure they'll take away what they can with or without a remedy. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 05:41, 25 February 2017 (UTC)
 * I'm sure that at least most of them have noticed this case and I'm pretty clueless about AWB, so I also don't feel I'm in a position to support this. 21:50, 25 February 2017 (UTC)
 * Drmies (talk) 05:19, 2 March 2017 (UTC)


 * Comments:
 * Not sure what we are asking to be done. I see comments on the talk page and at about the AWB option "Skip if genfixes only" which evidently would solve a number of the current issues. Striking my support right now.  Doug Weller  talk 19:11, 13 February 2017 (UTC)
 * What I had in mind was classifying the existing genfixes - including the checkwiki errors - to identify those that are pretty much always going to be "cosmetic-only", and giving those a separate checkbox. That way cosmetic fixes could be enabled only for AWB runs where you know a priori that every edit will be a "substantive" one. The problem with the "skip if genfixes only" option is that some of the genfix items are non-cosmetic and unambiguously useful even in isolation, while others are trivial linter-type changes.
 * Of course, the whole point of having a category of "general fixes" at all is that these issues will slowly get resolved over time as AWB operators move about the wiki doing other tasks. For any individual set of edits, it's easy to say that the genfixes were trivial and can easily be skipped, but adopting that as a general practice would reduce the effectiveness of the tool. Opabinia regalis (talk) 21:48, 13 February 2017 (UTC)
 * BTW, going to ping the AWB devs and  as the checkwiki project coordinator in case they have input on whether this is feasible/useful/desirable. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)
 * I'm leaning to oppose as well. I think without an FOF that AWB has a wider issue that a remedy of this nature is not necessary. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 17:24, 24 February 2017 (UTC)
 * To my mind the point is to encourage turning the outcomes of remedies 1 and 3 (if any) into something usable in practice, to whatever extent is feasible. Reciprocally, the constraint that any definition of "cosmetic" would ideally be technically identifiable at the time of the edit might help nail things down in those discussions. Opabinia regalis (talk) 09:54, 26 February 2017 (UTC)

Community encouraged to develop policy on bot unblocking
4) The community is encouraged to hold an RfC to clarify the currently-ambiguous policy surrounding the unblocking of one's own bot account, and the blocking and unblocking of problematic bots more generally. The committee notes that an RfC on this topic is currently under development.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * In general, I think that your garden variety admin (myself included) may not be qualified to determine if changes to a bot account address the issues, however a clear "yes or no" policy regarding the subject is warranted. if the answer is "no", speaking on my own advice, for the most part I would willing to assume good faith on the part of the operator and unblock the bot if they said the issues were addressed, but best practice would be for the operator to not unblock their own bot. --kelapstick(bainuu) 10:51, 10 February 2017 (UTC)
 * Limited support, even though I wrote the proposal :) I think Euryalus' interpretation (which I take it is also shared by Ramaksoud and Rob, among others) is an overreading of the policy. Bot accounts are "alternative accounts", yes, but a fundamentally different kind than the one you use on your phone or whatever; bot blocks are frequently in response to technical rather than "social" or behavioral issues, and the owner of the bot is best placed to judge whether the technical problem is solved. That this has come up twice in the space of five years suggests to me that we actually don't have any great need for a formal policy on the subject. Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * I don't think it's clear whether the general policy on self-unblocking is applicable to bots blocked for temporary malfunction.  DGG ( talk ) 18:10, 13 February 2017 (UTC)


 * Oppose:
 * Unnecessary. Existing policy at WP:BLOCK and WP:BOTPOL make clear that the edits of a bot are considered to be, by extension, the edits of the editor responsible for the bot. WP:BLOCK also makes clear that Unblocking will almost never be acceptable ...to unblock one's own account. The blocking policy does not contain an exemption for the alternate accounts of admins, whether they be bot owners or no. Barring an appropriate application of IAR, you may not unblock your own account, or your own alternate account, regardless of whether you feel you've addressed the issue that led to the block. I do agree with Kelapstick that there will occasions where a blocking admin is unable to determine if a technical change has addressed the reason why a bot was blocked; however these can be via the standard process for unblocks: with a combination of good faith, some balanced judgement and a watch-and-see attitude. This would be akin to the circumstance of any other editor (including any non-admin bot owner) who appears disruptive, is blocked and then promises that the problem will not recur. -- Euryalus (talk) 00:23, 11 February 2017 (UTC)
 * I do think the policy is unclear, and so my first inclination was to support this remedy, because who wants unclear policies? Moreover, in my prior term on the Committee, I supported several instances of this type of remedy, and at least some of the time the ensuing RfC was helpful in eliminating a source of dispute. But if "ArbCom says the community should discuss X" will typically lead to the community's discussing X, we should save it for the most important discussions that ought to happen, and I'm not convinced that this is one of those, especially given that we're already requesting in this discussion that the community have two other discussions. In other words, I'm certainly not against the community having this discussion&mdash;and if they do someone can feel free to copy my comments below if I don't get there first&mdash;but I don't think it's a top priority. Newyorkbrad (talk) 00:48, 15 February 2017 (UTC)
 * Like Brad I'm not opposed to the community having the discussion if they wish, but this is a one-time issue, and wasn't even an issue in the Betacommand case. I don't see this as a community-wide problem, but a individual problem with someone overstepping their bounds. -- Amanda  (aka DQ) 09:07, 22 February 2017 (UTC)
 * Covered well enough in current policy Cas Liber (talk · contribs) 11:18, 22 February 2017 (UTC)
 * I'm inclined to agree with my colleagues that this instance seems isolated and can be addressed in an other resolution. As DQ said, the community certainly has the ability regardless of our recommendation to hold an RFC on the practice but it's something I think they can decide themselves. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 18:58, 23 February 2017 (UTC)
 * Per Mkdw. Ks0stm  (T•C•G•E) 21:59, 24 February 2017 (UTC)
 * Doug Weller talk 11:29, 25 February 2017 (UTC)
 * Per Mkdw. We're already asking the community to begin discussions on other matters, and I don't think the policy is terribly unclear here. GorillaWarfare (talk) 01:49, 1 March 2017 (UTC)
 * Per Euryalus, Mkdw. Drmies (talk) 05:21, 2 March 2017 (UTC)
 * Per Mkdw and DQ. Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Abstain:


 * Comments:
 * In response to Euryalus, if what is stated in policy isn't being done in practice, than perhaps it's time to review the policy and either modify it, or reinforce it. If someone were to ask me if one could unblock their own bot, I would say "no". Having said that, it appears that in practice, the answer is "yes" (or "sometimes"). --kelapstick(bainuu) 06:31, 11 February 2017 (UTC)
 * I agree that policy on this issue is less than crystal-clear; that became obvious in the arbitrator voting in the Rich Farmbrough case five years ago, and it hasn't been clarified since. But I'm not sure that a bright-line rule is what is warranted here. Let's step back for a moment and ask why an admin is not allowed to unblock himself or herself: it's because a block is a sanction imposed for user misconduct, and no one can be relied upon to judge fairly (or be seen to judge fairly) whether he or she has engaged in misconduct, or made a sufficiently credible promise to stop the misconduct, or whether his or her own misconduct warranted a block and for how long. By contrast, a bot block (or "emergency stop" as it's sometimes called) may reflect an unanticipated outcome in applying coding to an unprovided-for case, and not reflect any form of user conduct issue at all. In those situations, the person who is running the bot, who is often the person who coded it, may be the editor best situated to determine whether a coding fix has resolved the unwanted performance issue, and I think he or she could make that decision neutrally in most cases. In such case, "code fixed, so unblocking my bot" might be uncontroversial and I gather it often is. By contrast, if the bot is blocked for "making the same unwanted edits for the sixth time" or "these are improper edits even though the operator disagrees," a self-unblock would be inappropriate. Whether these nuances can be picked up in a policy, or whether they are too complicated and a bright-line rule would be better, are questions for the policy-writers, I suppose. Newyorkbrad (talk) 18:55, 11 February 2017 (UTC)


 * Thanks for the replies. I contend that the existing prohibition on unblocking ones own account is already a bright line. There is no exemption in blocking policy for bot owners who happen to also be admins, to unblock their alternate accounts. There's also no need for such an exemption: it is not especially onerous to advise the blocking admin that the reason for the block has been addressed and give them a chance to assess that in appropriate good faith. Alternatively the admin bot owner (and note, we are only talking about admin bot-owners; other bot-owners already have to follow the standard unblock process) can raise the bot block at AN or ANI if they feel the blocking admin is unresponsive, acting in bad faith or simply doesn't understand that a technical fix has solved the issue. True, following this process is slower than just unblocking oneself and carrying on. But a) it's not that much slower; b) it ensures that there is oversight of the claim that the issue is resolved, and c) it helps preserve the concept that standard editing policies are universally applied.


 * Of course its open to the community to introduce an RFC on this or any issue. My reason for opposing this motion is that it gives sanction to the idea that the existing policy is unclear, rather than unevenly enforced. -- Euryalus (talk) 22:34, 11 February 2017 (UTC)
 * IMO the chain of reasoning "You can't unblock yourself --> Therefore you can't unblock your own alternate accounts --> Bots are alternate accounts --> Therefore you can't unblock your own bot" is one of those things that looks superficially convincing, but is stringing together bits of different policies developed semi-independently to reach a conclusion that wasn't obviously intended and isn't obviously of practical value. The implication is there well enough, but it's the policy equivalent of OR. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)

Bot approvals group encouraged to carefully review BRFA scope
5) The Bot Approvals Group is encouraged to carefully review the proposed scope of any new bot request for approval to ensure that the scope and tasks are clearly defined and will resist scope creep.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * Mild support. Recent BAG discussions suggest this is happening anyway. Euryalus (talk) 22:43, 11 February 2017 (UTC)
 * Subject to my comment below. Newyorkbrad (talk) 23:33, 11 February 2017 (UTC)
 * And suggest scope creep if NYB would prefer a less militaristic phrase with a less developed article. --kelapstick(bainuu) 09:38, 13 February 2017 (UTC)
 * with kelapstick's suggested change .  DGG ( talk ) 18:11, 13 February 2017 (UTC)
 * Done. Opabinia regalis (talk) 21:17, 13 February 2017 (UTC)
 * Cas Liber (talk · contribs) 11:20, 22 February 2017 (UTC)
 * Title isn't completely clear, maybe distracting too, but the remedy says what it needs to say. -- Amanda  (aka DQ) 11:24, 24 February 2017 (UTC)
 * I think BAG is already aware of the situation but having this formalized and empowering a community process would be a positive step forward. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 17:15, 24 February 2017 (UTC)
 * Sure. Ks0stm  (T•C•G•E) 21:59, 24 February 2017 (UTC)
 * Support, agreeing with Mkdw. Doug Weller  talk 11:30, 25 February 2017 (UTC)
 * Can't hurt, though I agree with those above who think the BAG is already aware. GorillaWarfare (talk) 01:53, 1 March 2017 (UTC)
 * Drmies (talk) 05:22, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:
 * Our "battlefield conduct" wording has become entrenched [sic] in wiki-usage although I've never liked it, and I'm not sure I really want to introduce another military metaphor into our decisions (that is said with great respect to the MILHIST contingent, of course&mdash;or maybe this is my revenge for the effort to stamp out legalisms). More substantively, we want bot approvals to be reasonably well-defined, but do we really want to require that every modest tweak of an existing, non-controversial task or reinterpretation of the scope of a task requires a new formal approval process? We know from the disputes that arise in enforcing our decisions that even the seemingly clearest remedy can give rise to issues of scope and interpretation, and I've pointed out elsewhere that this is unavoidable even in theory. If that is true here it is true of other aspects of the project as well: There are limits to the degree of precision that can be expected of virtually any wiki process. Our suggestion here is probably best read with this in mind. Newyorkbrad (talk) 23:27, 11 February 2017 (UTC)
 * Changed to "scope creep" as suggested by Kelapstick.
 * On the substance, I hope "resistant to scope creep" is a reasonably flexible criterion that BAG would find useful. It would be perfectly reasonable to, for example, be pickier about scope for bots that will edit in mainspace than for those that just do internal maintenance tasks. (The example I'm thinking of is the creation of WP:TFD/O awhile back under the existing authorization for TfD maintenance tasks. Opabinia regalis (talk) 21:17, 13 February 2017 (UTC)

Magioladitis reminded to refrain from cosmetic-only edits
6) Magioladitis has begun the process of resubmitting BRFAs for the set of Yobot's previous bot tasks that he wishes to continue. He is reminded to use caution when performing future edits, either automated under the Yobot account or semi-automated under his main account.  He is warned not to make cosmetic-only edits, either automated under a bot account or semi-automated under his main account.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * kelapstick(bainuu) 09:34, 13 February 2017 (UTC)
 *  DGG ( talk ) 18:12, 13 February 2017 (UTC)
 * I could live with a somewhat stronger verb. In the last sentence, "so long as applicable policy prohibits such edits" is implied (we need not make it express as I gather the likelihood of a sea-change in the policy in the near term is remote). Newyorkbrad (talk) 00:42, 15 February 2017 (UTC)
 * Cas Liber (talk · contribs) 12:39, 22 February 2017 (UTC)
 * <S>Conditonal support provide the remedy title is changed to "Magioladitis warned to refrain from cosmetic-only edits" so as to match the wording of the remedy. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 17:22, 24 February 2017 (UTC)
 * On amended wording above. I share Doug Weller's view that this is a very mild remedy, and am happy to consider further wording alternatives. I note however the routine lack of clarity about what is a cosmetic-only edit, and the varied opinions about how much they actually matter. -- Euryalus (talk) 22:21, 14 February 2017 (UTC)


 * Oppose:
 * Policy says "Cosmetic changes (such as many of the AWB general fixes) should be applied only when there is a substantive change to make at the same time." Magioladitis needs to follow policy and shouldn't need a reminder. The second sentence about caution is a separate issue. Doug Weller  talk 17:54, 14 February 2017 (UTC)
 * Sorry, probably not a second issue if it's part of the cosmetic-only issue. But this should still be an instruction, not simply a reminder. Doug Weller  talk 19:23, 14 February 2017 (UTC)
 * I very much have to agree with Doug on "shouldn't need a reminder". -- Amanda  (aka DQ) 11:28, 24 February 2017 (UTC)
 * Prefer 7.1 over this. Ks0stm  (T•C•G•E) 22:01, 24 February 2017 (UTC)
 * I've supported 7.1 one instead. I think the stronger language is important. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:36, 27 February 2017 (UTC)
 * Too weak. "So-and-so is reminded/warned to follow policy" remedies are useless. GorillaWarfare (talk) 01:55, 1 March 2017 (UTC)
 * We shouldn't do just reminders here--per Doug. Drmies (talk) 05:26, 2 March 2017 (UTC)
 * Per Doug and GW. Keilana (talk) 04:59, 9 March 2017 (UTC)
 * No need for this given support for 7.1. -- Euryalus (talk) 23:43, 14 March 2017 (UTC)


 * Abstain:


 * Comments:
 * As suggested on the talk page, I fixed this to read "cosmetic- only edits", since no one disputes that "cosmetic" edits in combination with more substantive changes are permitted. Opabinia regalis (talk) 22:46, 10 February 2017 (UTC)
 * I'm okay with the change. Kirill Lokshin (talk) 23:36, 10 February 2017 (UTC)
 * The wording doesn't entirely reference the header. How about a new second sentence. "...is reminded not to make cosmetic-only edits, either automated under the Yobot account or semi-automated under his main account." -- Euryalus (talk) 01:57, 14 February 2017 (UTC)
 * Sounds good to me. Opabinia regalis (talk) 21:47, 14 February 2017 (UTC)
 * Thanks, per BRD have made the change. Pinging everyone else who already voted to check this is ok (and feel free to revert if not)., , , . -- Euryalus (talk) 22:17, 14 February 2017 (UTC)
 * Added a stronger verb ("warned" instead of "reminded") per above. Views welcome. -- Euryalus (talk) 01:35, 15 February 2017 (UTC)
 * Sounds fine, though the more "strong verbs" get added, the more significance attaches to the question Ramaksoud asked on the talk page about responding to possible violations. The more this case goes on, the more I'm convinced that (to risk another military metaphor) this is a pick-your-battles issue. Errors that are really worth escalating over are those that harm the reader's experience or actively damage community processes. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)
 * No concerns on my part. --kelapstick(bainuu) 09:53, 15 February 2017 (UTC)

Magioladitis restricted from making general fixes or checkwiki edits
7) Magioladitis is temporarily restricted from making any edit that solely applies AWB's "general fixes" or that addresses a checkwiki error until such time as the community discussions recommended in the preceding sections conclude, and is required to use the "skip if genfixes only" option when using AWB except as explicitly authorized by an approved bot request for approval.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * As modified - use "skip if genfixes only" unless specifically authorized to do otherwise. Opabinia regalis (talk) 08:18, 24 February 2017 (UTC) See comments on 7.1. That one's better than this one, at least in that there would be no way to tell for sure whether this restriction is being adhered to. (Incidentally this is the same type of problem that affected the "no using automation" remedy in the Rich Farmbrough case.) Opabinia regalis (talk) 08:35, 26 February 2017 (UTC)
 * Second choice, prefer 7.1. -- Euryalus (talk) 23:09, 26 February 2017 (UTC)
 * Second choice, prefer 7.1, and see my comments there. Newyorkbrad (talk) 03:35, 27 February 2017 (UTC)


 * Oppose:
 * Was supporting, but there are two issues here. First, there is no fixed definition of "temporary", and second, since we cannot mandate that these discussions take place, it is an indefinite restriction until a theoretical discussion occurs. Which it may not. --kelapstick(bainuu) 09:44, 24 February 2017 (UTC)
 * While this is a good start, it's not got the entire issue at hand, and the use of temporary is too vague. That and these community discussions can not be forced upon the community. -- Amanda  (aka DQ) 10:24, 24 February 2017 (UTC)
 * I'm inclined to agree that this might be difficult to fulfill and enforce. The remedies provided elsewhere might do enough to resolve the issue. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 17:20, 24 February 2017 (UTC)
 * Prefer 7.1. Ks0stm  (T•C•G•E) 22:01, 24 February 2017 (UTC)
 * Opposing in favor of 7.1. GorillaWarfare (talk) 01:56, 1 March 2017 (UTC)
 * Drmies (talk) 05:25, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)
 * Doug Weller talk 17:02, 9 March 2017 (UTC)


 * Abstain:
 * Am keen to wrap up my remaining votes on this PD, so putting this here for now in absence of consensus on a reword. I like OR's "skip if genfixes only" suggestion below and would move to support if this was added to the remedy. -- Euryalus (talk) 13:47, 17 February 2017 (UTC) Moving to support per amendment. -- Euryalus (talk) 08:20, 24 February 2017 (UTC)
 * Sorry, I completely missed this comment. See suggested changes above. Opabinia regalis (talk) 08:18, 24 February 2017 (UTC)


 * Comments:
 * Hmm, I just realized this has the same problem as the preceding remedy - what we want to restrict is making these edits in isolation. Do you object to changing the first sentence to Magioladitis is temporarily restricted from making any edit that solely applies AWB's "general fixes", or that solely addresses a checkwiki error,...? Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * Actually, on rereading I may be adding rather than reducing confusion, since many of those edits shouldn't be applied in isolation anyway. It's Friday, maybe I'll have a better idea tomorrow. Opabinia regalis (talk) 06:22, 11 February 2017 (UTC)
 * Awaiting further clarification here before voting. Newyorkbrad (talk) 00:42, 15 February 2017 (UTC)
 * I don't seem to have gotten any smarter over the weekend ;) One suggestion that's come up is to require the "skip if genfixes only" option to be used. (I think that would be more useful if the software/interface changes I suggested above were implemented, though - it's easy to "skip if cosmetic-only".) Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)

Magioladitis restricted
7.1) Magioladitis is restricted from making any semi-automated edits which do not affect the rendered visual output of a page. This restriction does not apply to edits which address issues related to accessibility guidelines. Further, Magioladitis may seek consensus to perform a specific type of semi-automated edit that would normally fall under this restriction at the administrators' noticeboard. Any uninvolved administrator may close such a discussion with consensus to perform a specific type of semi-automated edit. All discussions should be logged on the case page, regardless of outcome.


 * Support:
 * This is not to define cosmetic edits by any means, but to work with a restriction that will prevent the issue at hand. -- Amanda  (aka DQ) 10:24, 24 February 2017 (UTC)
 * This seems more likely to be effective at preventing recurring issues. Ks0stm  (T•C•G•E) 22:03, 24 February 2017 (UTC)
 * Kirill Lokshin (talk) 02:12, 25 February 2017 (UTC)
 * First choice over remedy 7 above. -- Euryalus (talk) 23:10, 26 February 2017 (UTC)
 * Regarding the discussion in the comments, I interpret "which do not affect the rendered output" to mean "which do not affect the rendered output in any reasonably common browser or skin" (or whatever the relevant parameters would be). Newyorkbrad (talk) 03:37, 27 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:35, 27 February 2017 (UTC)
 * GorillaWarfare (talk) 01:56, 1 March 2017 (UTC)
 * Doug Weller talk 13:59, 1 March 2017 (UTC)
 * Drmies (talk) 23:46, 7 March 2017 (UTC)
 * Provided that Opabinia's solution from below is implemented. Keilana (talk) 04:59, 9 March 2017 (UTC)
 * Cas Liber (talk · contribs) 23:58, 12 March 2017 (UTC)


 * Oppose:
 * Sorry, this isn't doing it for me. I changed my mind on 7.0 above because it fell into the same trap the Rich Farmbrough decision did by trying to mandate a behavior that was unverifiable. This one takes the logical step of trying to regulate the output instead of the input, but falls back into the Farmbrough anti-precedent by defining the output requirements like porn - you know, we'll know it when we see it. But we won't, because the evidence page and various talk-page threads about this case are chockablock full of people not knowing if the edit they're looking at "affects the rendered visual output" or not. (Beans alert: can Magioladitis, under this restriction, replace the probably undesirable а (U+0430) with the probably correct a (U+0061) in running English text?) Opabinia regalis (talk) 06:48, 3 March 2017 (UTC)


 * Abstain:


 * Comments:
 * We're getting closer, but I don't think this quite does the trick either. Magioladitis seems to make (at least) two types of semi-automated edits that commonly get objections. One is running through a list of pages that all appear to need the same fix, and applying genfixes to a few on the list that turn out not to need the "substantive" edit he was trying to make - which then only get trivial/cosmetic changes, often with a confusing edit summary. A recent example is here. Another is making edits to correct very small defects in the wikitext that are interpreted by observers as cosmetic edits because the defect is only visible with certain browsers/OSes/screen resolutions/window sizes/whatever, as in this thread (There are also cases of apparent non-cosmetic edits that turn out to have no effect, like this one.)
 * Remedy 7 is aimed mostly at the first problem, but has the disadvantage that it's not really possible to know what AWB options someone used; you can only infer from the output. Remedy 7.1 has the advantage of focusing on the output, but at the cost of reintroducing at least some of the gray areas covered in this case about what the "rendered visual output" really is. If it changes the output HTML, then there is probably some browser out there that will render it differently. It's also possible to change the "output" only for users of certain preferences, like those who show hidden categories. I find this exchange on the workshop between Rob and Whatamidoing to be illustrative of this issue - even experienced and knowledgeable editors differ on what types of edits qualify as making a visual change or being a cosmetic edit. Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)
 * Running through a list of pages can use a separate remedy if you feel it's needed, it's really a completely different issue. As I read it, that wasn't included in it. We should be taking the strictest of interpretations, that a change, if it occurs for one person, is still valid, hence what I was getting at with the accessibility guidelines. So I do see this remedy as doing it's job. Are we missing a remedy though for clear edit summaries? -- Amanda  (aka DQ) 20:29, 25 February 2017 (UTC)
 * I don't mean that lists of pages or accidentally screwy edit summaries need to be specifically restricted. They're just symptoms. But I think they illustrate how the process is going awry. Thinking about this more, I think the problem is that "rendered visual output" isn't a well-defined category. If the HTML is different, a scenario can probably be constructed in which the display will be different, even if most users under most conditions would see no change. "Edits to wikitext that do not affect the HTML that is served to a logged-out user" is at least a defined set, though probably narrower than what some think of as "cosmetic", and sometimes unintuitive (as in the thread on my talk linked above). Opabinia regalis (talk) 09:27, 26 February 2017 (UTC)

Magioladitis reminded to avoid duplication of undesirable editing patterns
8) Magioladitis is reminded that performing the same or similar series of edits in an automated fashion using a bot and in a semi-automated fashion on his main account is acceptable only as long as as long as no objections have been raised in either case  the edits are not contentious. Should Yobot be stopped or blocked for a series of edits, Magioladitis may not perform the same pattern of edits via semi-automated tools from his main account where this might reasonably be perceived as evading the block. In this circumstance, Magioladitis (like any other editor) should await discussion and consensus as to whether or not the edits are permissible and useful, and resume making such edits through any account only if and when the consensus is favorable.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * As best practice. --kelapstick(bainuu) 10:55, 10 February 2017 (UTC)
 * Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * to prevent gaming.  DGG ( talk ) 20:34, 14 February 2017 (UTC)
 * Euryalus (talk) 13:56, 17 February 2017 (UTC)
 * Support as reworded, and added the last sentence, which I believe summarizes the point. (Any arbitrator who disagrees with the added sentence is free to remove it and this vote will still stand.) Newyorkbrad (talk) 20:46, 19 February 2017 (UTC)
 * Doug Weller talk 16:23, 20 February 2017 (UTC)
 * Cas Liber (talk · contribs) 12:40, 22 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:00, 23 February 2017 (UTC)
 * -- Amanda  (aka DQ) 11:30, 24 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 22:03, 24 February 2017 (UTC)
 * I'm not thrilled with the "where this might reasonably be perceived as evading the block" wording, which I think is quite vague, but I support this nonetheless. GorillaWarfare (talk) 01:58, 1 March 2017 (UTC)
 * Drmies (talk) 05:24, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:


 * Abstain:


 * Comments:
 * "Inadvisable" is too mild. If your alternate (bot) account is blocked following a series of undesirable edits, you are flatly not permitted to then go make those same or similar edits from your main account while the issue is unresolved. -- Euryalus (talk) 00:07, 11 February 2017 (UTC)
 * Eh, it depends on the exact reason for the block. For example, if the bot malfunctioned by making a type of edit on the wrong set of pages, it wouldn't be crazy to think you could make the edits manually on the correct pages while sorting out the bot issue, assuming the edits themselves were reasonable. Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 * True enough, though in that case the issue has been resolved (or not revisited), because the problem was not the edits but the malfunction. Not opposed to the motion per se, just exploring whether the word "inadvisable" might be misinterpreted as allowing more licence than intended. That said, I can't immediately think of a better way of putting it without the motion becoming needlessly detailed. Will add an actual vote a bit later today. -- Euryalus (talk) 22:52, 11 February 2017 (UTC)
 * Could we add "and if there are complaints he should stop until the issue is resolved" or something like this? Although it is just a reminder, so I'm not sure if this will make a difference. Doug Weller  talk 15:40, 12 February 2017 (UTC)
 * How about a new last sentence: "Should Yobot be stopped or blocked for a series of edits it is not permitted to make the same pattern of edits from another account, unless doing so entirely addresses the reason for the block being imposed." Not great wording, but hopefully conveys the sense of what I mean. -- Euryalus (talk) 22:40, 14 February 2017 (UTC)
 * Hmmm. Nailing this down is going to be hard - on the one hand, plausible counterexamples are easy to dream up, but on the other, it would be easy to argue that some set of edits doesn't entirely address the issue and therefore partial fixes aren't permitted. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)
 * I strongly agree with the substance of the proposed remedy, but "as long as no objections have been raised in either case" seems too strong&mdash;we can't mean to give every individual user on the project a veto power over Magioladitis's edits. When this is wordsmithed to address the other comments, perhaps this can be adjusted as well. Thanks, Newyorkbrad (talk) 00:51, 15 February 2017 (UTC)
 * "...only if the edits are not contentious"? Which arguably means the same thing, but people generally take "contentious" to mean multiple people involved in the contention. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)
 * Reworded the remedy to incorporate the "not contentious" wording and reword the last sentence. Pinging those who've voted in case they disagree, and as always everyone feel free to revert if I haven't captured the spirit of the thing. -- Euryalus (talk) 13:56, 17 February 2017 (UTC)
 * "discouraged" in the last sentence is too weak. I suggest "should not perform..."  DGG ( talk ) 01:51, 18 February 2017 (UTC)
 * Amended to "may not," views welcome. -- Euryalus (talk) 02:14, 18 February 2017 (UTC)
 * This is my fault for not noticing the change till just now, but I'm not fond of the change to "might reasonably be perceived as evading the block". That has the same problem as the "unblock your own bot" issue - it implicitly canonicalizes this interpretation of "block evasion". As far as I'm aware this has never specifically been considered as a policy issue, and stringing together bits of policy wording not developed in coordination to derive novel implications is policy synthesis we don't really need in order to substantiate this remedy. I actually like the original suggestion above better, loopholes notwithstanding. Opabinia regalis (talk) 22:20, 21 February 2017 (UTC)
 * No strong opinions either way as yet, but is this intended to be an elaboration on the theme, or a substantive change? I read the new sentence as requiring a positive consensus in favor (by some definition), which is different than simply requiring that the edits not be "contentious". Opabinia regalis (talk) 05:49, 20 February 2017 (UTC)
 * I wasn't intending any substantive change, but You make a good point by asking what should happen in a "no consensus" situation. What do you think should happen?&mdash;should the edits be allowed or not?&mdash;or should we just avoid the question? (This reminds me of the situation that arises occasionally where an unblock request is discussed on AN or ANI and opinion is closely divided. Does that mean the editor should stay blocked, because there is no consensus to unblock, or that the editor should be unblocked, because "unblocked" is the default state and there's no consensus for the block? But I digress.) Newyorkbrad (talk) 18:27, 20 February 2017 (UTC)
 * Heh, my original question was dumber than that - more a practical consideration of the difference between "consensus in favor" (implying a discussion with a positive result, possibly with a formal close) and mere "lack of contention" (a squishier concept, but more suited to things like talk-page threads and other minor matters) as the criterion for moving forward with a set of edits. Come to think of it, that issue could be somewhat fixed by a change to "discussion or consensus". I think that's not quite the same as what to do when a formalized discussion of some type has occurred and the result is ambiguous. My personal preference in that case is that "no consensus" means "restore the status quo ante", but I can't remember ever convincing someone of that in practice :) Opabinia regalis (talk) 22:20, 21 February 2017 (UTC)

Magioladitis requested to seek broad input for new tasks
9) With the exception of existing Yobot tasks currently being re-reviewed, Magioladitis is requested to seek broad input into new bot tasks, such as by posting a link to the BRFA on one of the Village Pumps, or any other noticeboard deemed relevant by BAG.


 * Support:
 * Kirill Lokshin (talk) 00:31, 10 February 2017 (UTC)
 * I'm open to persuasion, but I don't see this as an onerous requirement, and it's probably a good idea in any case for a large-scale new bot task. OK, maybe the village pump instead of AN. Opabinia regalis (talk) 06:19, 11 February 2017 (UTC)
 *  DGG ( talk ) 18:13, 13 February 2017 (UTC)
 * Support in principle subject to finalization of the verb and the venue. Newyorkbrad (talk) 20:47, 19 February 2017 (UTC)
 * I am as yet undecided on whether this should be a requirement, but at the very least it doesn't hurt to make it a request. Ks0stm  (T•C•G•E) 08:29, 3 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:
 * This should be a requirement. New tasks should either achieve consensus or no opposition for a reasonable period such as 72 hours. Doug Weller  talk 17:42, 14 February 2017 (UTC)
 * Agree "request" is a bit mild for this remedy to have value, but am also not a supporter of a compulsory referral. If there's greater scrutiny at BAG (which there is at least right now), then there's no need for referral to another board as well. Also, I suspect that a village pump or AN discussion on any specific Yobot task would only attract a) the same people as might otherwise comment at BAG, and b) others who sought to broaden the discussion into a general debate about BOTPOL. Interesting though that might be as a policy debate, it would drag out any future Yobot approvals while unrelated matters were discussed. Happy to discuss further but for now an (at least temporary) oppose. -- Euryalus (talk) 22:46, 14 February 2017 (UTC)
 * I now agree with Doug that this should be a requirement. Using any weaker language might well result in reoccurrence of the same pattern that led to this case.  DGG ( talk ) 01:54, 18 February 2017 (UTC)
 * GorillaWarfare (talk) 01:59, 1 March 2017 (UTC)
 * too nebulous and able to be gamed Cas Liber (talk · contribs) 03:25, 3 March 2017 (UTC)
 * BAG should be able to handle this. Drmies (talk) 02:21, 14 March 2017 (UTC)
 * Abstain:


 * Comments
 * AN does not strike me as the right forum to be burdened with this type of announcement, which might confuse editors who aren't familiar with this case and would wonder "why is this being announced here?" The relevant Village Pump might be better ... but isn't there a BRFA new/pending requests page that any interested editors could watchlist, which would make them aware of any pending requests from Magioladitis or anyone else? Failing that, how about having Magioladitis cross-reference any pending requests on his user-talkpage? Newyorkbrad (talk) 00:55, 15 February 2017 (UTC)
 * The thinking was to get input from a broader cross-section of the community than would normally be watching for new BRFAs, which are mostly technical and boring. However, it does seem vulnerable to filibustering. Talk page is a good idea. Opabinia regalis (talk) 09:02, 15 February 2017 (UTC)
 * Headbomb has suggested "With the exception of existing Yobot tasks currently being re-reviewed, Magioladitis is requested to seek broad input into new bot tasks, such as by posting a link to the BRFA on the Village Pumps, or any other noticeboard deemed relevant by BAG." I agree with him that AN is not appropriate. Doug Weller  talk 21:40, 25 February 2017 (UTC)
 * This sounds like a good solution. However, since the purpose is avoidance of doubt, and we don't want all the village pumps to get the post, is just VPT sufficient? Opabinia regalis (talk) 09:37, 26 February 2017 (UTC)
 * Also, as pointed out on the talk page, the title is probably better written as "broad input" to match the text. Opabinia regalis (talk) 10:20, 26 February 2017 (UTC)
 * Rewording to change title and revise "Village Pumps" to "one of the Village Pumps" - this would usually be VPt but I don't think we should prescribe. This still leaves the request vs require issue however. Doug Weller  talk 13:27, 26 February 2017 (UTC)
 * Per Headbomb, leave it to BAG. Opabinia regalis (talk) 19:13, 26 February 2017 (UTC)
 * OR, any thoughts about making this mandatory as opposed to a request? <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 19:34, 27 February 2017 (UTC)
 * Sorry, I'm working on a real-life deadline and my brain is fried. The above was meant to answer that question - leave it to BAG to decide which BRFAs should be required to be advertised (which I think they sometimes ask for anyway, as Headbomb pointed out). Opabinia regalis (talk) 05:45, 1 March 2017 (UTC)

Magioladitis required to seek broad input for new tasks
9.1) With the exception of existing Yobot tasks currently being re-reviewed, Magioladitis is required to seek broad input into new bot tasks, such as by posting a link to the BRFA on one of the Village Pumps, or any other noticeboard deemed relevant by BAG.


 * Support:
 * given the background yes Cas Liber (talk · contribs) 03:24, 3 March 2017 (UTC)


 * Oppose:
 * As I said above, and as Headbomb said on the talk page, deciding whether an RfC is needed for a particular BRFA is BAG's job. For those tasks that are uncontroversial enough that no explicit consensus is needed, there's no benefit to broader notice. In fact it's the opposite, and likely to lead to notification fatigue. Opabinia regalis (talk) 05:40, 3 March 2017 (UTC)
 * per my comments on remedy 9 above. However it's helpful that this remedy was posted as an option, so thanks for that. -- Euryalus (talk) 06:11, 3 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)
 * I am persuaded. Cas Liber (talk · contribs) 23:56, 12 March 2017 (UTC)
 * See above (in other words, what Opabinia regalis says). Drmies (talk) 02:22, 14 March 2017 (UTC)
 * And me. Doug Weller  talk 16:02, 14 March 2017 (UTC)
 * Also  DGG ( talk ) 03:04, 22 March 2017 (UTC)
 * Abstain:


 * Comments:
 * There seems to be some level of support above for this being required as opposed to requested. Ks0stm  (T•C•G•E) 07:58, 2 March 2017 (UTC)

Magioladitis restricted from unblocking own bot
10) Magioladitis is restricted from unblocking their own bot when it has been blocked by another administrator. After discussion with the blocking administrator and/or on the bot owners' noticeboard, the blocking administrator or an uninvolved administrator may unblock the bot.


 * Support:
 * A clear issue that is part of the bigger problem is Magioladitis unblocking their own bot. Community norms are that you speak, consult and resolve the issues with the blocking administrator before unblocking. It's been unblocked out of process several times by Magioladitis and they have failed to address community concerns. This is an easy support. -- Amanda  (aka DQ) 09:25, 22 February 2017 (UTC)
 * Euryalus (talk) 12:18, 22 February 2017 (UTC)
 * Cas Liber (talk · contribs) 12:41, 22 February 2017 (UTC)
 * I was thinking of suggesting something like this myself, either informally as advice to Magioladitis, or more formally as part of the decision as here. As stated above, I'm not certain that policy forbids, or should forbid, an administrator from unblocking his or her own bot, where the block was simply an "unanticipated malfunction stop" and the issue has been resolved. But Magioladitis's unblocks of Yobot have been controversial, and if he were to unblock Yobot himself again, it would be even more controversial. More generally, where there has been a long string of blocks and unblocks then it is more desirable to have another admin do the unblocking. Please note that I copyedited the remedy by adding the words in parentheses, subject to DeltaQuad's approval. Newyorkbrad (talk) 15:43, 22 February 2017 (UTC)
 * Doug Weller talk 16:27, 22 February 2017 (UTC)
 * Kirill Lokshin (talk) 18:59, 22 February 2017 (UTC)
 *  DGG ( talk ) 04:03, 23 February 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 18:55, 23 February 2017 (UTC)
 * Limited support, in that I disagree that a debatable-but-not-obviously-against-policy act that last occurred in June 2016 and prior to that in June 2014 is an ongoing problem that warrants a remedy. However, it's clear that if Magioladitis does this again there will be a shitstorm, and it's consistent with the rest of my views on this case that behavioral expectations should be clearly stated. I also think it's important to emphasize that a) this does not have any implications for bot-unblocking in general, and b) it is best practice to use a stop mechanism in a non-emergency if a bot provides one. On a more practical note, how about just "with the blocking admin and/or BOTN". Current practice for blocks in general allows for reversing a block by broader consensus even if the admin is available. Opabinia regalis (talk) 08:13, 24 February 2017 (UTC)
 * As I noted on the talkpage, I agree with your "important to emphasize" comments. I also have no objection to your final proposal to clear up venues, wasn't my specific wording. -- Amanda  (aka DQ) 11:34, 24 February 2017 (UTC)
 * See above; is that copyedit OK? Opabinia regalis (talk) 01:07, 25 February 2017 (UTC)
 * Ya. I made a few changes, grammar still isn't perfect, but...meh. -- Amanda  (aka DQ) 19:28, 25 February 2017 (UTC)
 * kelapstick(bainuu) 08:18, 24 February 2017 (UTC)
 * Ks0stm (T•C•G•E) 22:04, 24 February 2017 (UTC)
 * Much-needed. Yobot is a very helpful bot, but I can think of no circumstance that would require Yobot be unblocked so urgently that Magioladitis could not request another administrator review and unblock the bot for him. GorillaWarfare (talk) 02:01, 1 March 2017 (UTC)
 * I think this is obvious. Drmies (talk) 05:28, 2 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)


 * Oppose:
 * Abstain:
 * Abstain:


 * Comments:
 * I see that Thryduulf notes that some bot owners have blocked their own bots and that we should allow Magioladitis to unblock his bot if he's the one who's blocked it. I agree. Doug Weller  talk 19:24, 22 February 2017 (UTC)
 * Addressed by inserting "when it has been blocked by another administrator" in the first sentence of the remedy, subject to approval of DeltaQuad and others. Personally I don't think this was likely to be an issue, but the clarity can't hurt. Newyorkbrad (talk) 22:03, 22 February 2017 (UTC)
 * Ok with the first, should we maybe specify the venue to AN/ANI? I'm not sure where else is appropriate. -- Amanda  (aka DQ) 07:45, 23 February 2017 (UTC)
 * On further thinking, and thanks to Xoasflux, I've made it BOTN as they will have the proper technical understanding to deal with the block. -- Amanda  (aka DQ) 22:30, 23 February 2017 (UTC)

Modified enforcement

 * 1) In the event of problems with the editing of any bot run by Magioladitis, administrators are encouraged to use the stop feature if provided. Blocking the bot account should preferably be reserved for urgent situations, failures or misuse of the stop feature, or edits that would be obviously unacceptable from any user account.
 * 2) Blocks of Magioladitis' main account as enforcement of restriction 7.x should be of a duration of no more than two weeks. After four such blocks, the matter should be raised at WP:ARCA.
 * 3) Blocks under any other provision of this decision should follow the standard enforcement provisions, with an initial block of up to one month, and then with blocks increasing in duration to a maximum of one year.


 * Support:
 * Opabinia regalis (talk) 21:40, 7 March 2017 (UTC)
 * To be clear, and to forestall the future ARCA request on the subject, Euryalus' comments below capture the intention of this provision. Opabinia regalis (talk) 22:46, 16 March 2017 (UTC)
 * Keilana (talk) 04:59, 9 March 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 15:00, 9 March 2017 (UTC)
 * Newyorkbrad (talk) 02:41, 12 March 2017 (UTC)
 * Per the discussion below - happy for additional wording to be added if required. It should be very clear that misuse of the stop feature (ie. Magioladitis restarting the bot without fixing the apparent issue) should be considered a "failure of the stop feature" and justification for further action like blocking the bot. -- Euryalus (talk) 22:57, 13 March 2017 (UTC)
 * Doug Weller talk 14:49, 12 March 2017 (UTC)
 *  DGG ( talk ) 18:05, 12 March 2017 (UTC)
 * Kirill Lokshin (talk) 22:54, 12 March 2017 (UTC)
 * Cas Liber (talk · contribs) 23:55, 12 March 2017 (UTC)
 * Lukewarm support. Drmies (talk) 02:24, 14 March 2017 (UTC)
 * Very much emphasizing what Euryalus said. Ks0stm  (T•C•G•E) 05:39, 16 March 2017 (UTC)


 * Oppose:
 * Too much ambiguity leaves this with many holes to circumvent. -- Amanda  (aka DQ) 06:29, 13 March 2017 (UTC) Looks like it'll be changed. --  Amanda  (aka DQ) 04:09, 20 March 2017 (UTC)


 * Abstain:


 * Comments:
 * The purpose here is to provide a defined and relatively predictable enforcement mechanism to avoid distracting arguments about the lengths of blocks in response to edits that violate the restrictions but are not otherwise inherently disruptive. For the avoidance of doubt, I originally floated an idea along these lines on the mailing list last week, before the current AE kerfuffle. Opabinia regalis (talk) 21:40, 7 March 2017 (UTC)
 * I wasn't going to be stopping any further in on this case, and was doing a cursory glance. Looking at this remedy, it seems to completely undercut the point of Remedy 10 as it points administrators to not block the bot, but use the stop button instead, which Magioladitis has reverted liberally without fixing the issue at hand and will insist and yell at admins that it needs to be used instead of block button as we've seen before. This will only increase the tension in the area. Your proposal also does not include any remedy for things Magioladitis has said they have "fixed" but continue to occur many times after, which is why we are here in the first place. If we pass this remedy, we undercut the case. -- Amanda  (aka DQ) 06:00, 13 March 2017 (UTC)
 * Question: If the bot gets stopped and Magioladitis restarts it without fixing the problem, is that not a "failure of the stop function" which should be addressed by blocking the bot? -- Euryalus (talk) 06:11, 13 March 2017 (UTC)
 * That's the million dollar question in the ambiguity of this. The way I as a developer read it is that failure of the stop function means the bot does not stop operating when that stop function is set to request the bot to stop. Normal bots get blocked for that regardless of arbcom remedies. If Magio then reenables it, it's still working fine, and hasn't failed to stop, it's just not ceasing the edits that occurred in the first place. Also I disagree with the number of blocks it takes to get back to us given the recommendation for people to use the stop button instead of the block button. -- Amanda  (aka DQ) 06:29, 13 March 2017 (UTC)
 * In my original suggestion on the mailing list, there was another sentence If the stop feature is used, Magioladitis is required to address the issue with the person who used it or with uninvolved administrators before restarting the affected task. I dropped it because it's not really about "enforcement" and is implied by point 1 and the existing remedies, but stating it explicitly sounds like it would solve your problem? Opabinia regalis (talk) 07:44, 13 March 2017 (UTC)
 * Partially. The big thing i'm looking for is another line in point one just saying that the bot failing to be fixed can be a reason to be blocked. -- Amanda  (aka DQ) 17:14, 14 March 2017 (UTC)
 * "Failures or misuse"? Opabinia regalis (talk) 00:41, 15 March 2017 (UTC)
 * I can live with that. -- Amanda  (aka DQ) 04:09, 20 March 2017 (UTC)
 * I've gone ahead and added "or misuse". Doug Weller  talk 11:23, 20 March 2017 (UTC)

General

 * I'm still in the process in reviewing. I think a resolution on this case is still missing a key piece, possibly in the form of a remedy not yet proposed, such as addressing the issue around semi/fully automated editing from the main account and unblocking their bot account in the form of a future restriction. <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 05:59, 19 February 2017 (UTC)
 * On the first, isn't that what remedy 8 is (in whatever final form), or do you think something different is needed? On the second, I'm still not comfortable requiring someone to follow, or sanctioning them for failing to follow, one of multiple possible interpretations of existing policy. Canonicalizing one of several interpretations turns into effectively backdoor policymaking. In any event, given that Magioladitis last unblocked Yobot in June 2016, and the last time before that in June 2014, I don't think a specific restriction is necessary. Opabinia regalis (talk) 05:49, 20 February 2017 (UTC)

Implementation notes
''Clerks and Arbitrators should use this section to clarify their understanding of the final decision—at a minimum, a list of items that have passed. Additionally, a list of which remedies are conditional on others (for instance a ban that should only be implemented if a mentorship should fail), and so on. Arbitrators should not pass the motion to close the case until they are satisfied with the implementation notes.''

These notes were last updated by Ks0stm  (T•C•G•E) 05:42, 16 March 2017 (UTC); the last edit to this page was on  by User:.


 * Notes

Vote
Important: Please ask the case clerk to author the implementation notes before initiating a motion to close, so that the final decision is clear.

''Four net "support" votes (each "oppose" vote subtracts a "support") or an absolute majority are needed to close the case. The Clerks will close the case 24 hours after the fourth net support vote has been casted, unless an absolute majority of arbitrators vote to fast-track the close.''


 * Support
 * Move to close. We've done what we can here and hopefully outlined a path forward. Newyorkbrad (talk) 21:39, 11 March 2017 (UTC)
 * Not normally a fan of close motions before all remedies have passed or failed, but progress has become slow and I think we need to move it along. Also the War of the Pacific case, which is much simpler than this one - can everyone still considering remedies in either case please have another read through when convenient so we can get them closed off and the remedies implemented. -- Euryalus (talk) 03:39, 12 March 2017 (UTC)
 * Cas Liber (talk · contribs) 23:59, 12 March 2017 (UTC)
 * <span style="color:black;text-shadow: 4px 4px 15px white, -4px -4px 15px white">Mkdw  <span style="color: #0B0080;text-shadow: 4px 4px 15px white, -4px -4px 15px white">talk 15:47, 13 March 2017 (UTC)
 * It's over two months since this case started, five weeks since the PD was posted, and almost a week since the first vote to close. I appreciate the desire to keep polishing and to preemptively clarify any possible points of future ambiguity, but it's time to close this and let everyone move forward. I do exactly this all the time writing real-world stuff and usually require several drinks before finally submitting the stupid thing. Will drinks help? I've got a well-stocked bar... ;) Opabinia regalis (talk) 23:07, 16 March 2017 (UTC)
 * Only remedies 9.x are yet to be decided, and they're decidedly going nowhere. Time to close this and hit up Opabinia's bar. Ks0stm  (T•C•G•E) 23:28, 16 March 2017 (UTC)
 * Drmies (talk) 16:14, 17 March 2017 (UTC)
 *  DGG ( talk ) 03:02, 22 March 2017 (UTC)
 * Oppose

Doug Weller talk 14:51, 12 March 2017 (UTC)
 * Comments