Wikipedia:Arbitration/Requests/Case/Magioladitis 2/Evidence
Case clerk: Miniapolis (Talk) Drafting arbitrator: Callanecc (Talk)
Wikipedia Arbitration |
---|
![]() |
|
Track related changes |
![]() | The Evidence phase for this case is closed.
Any further edits made to this page may be reverted by an arbitrator or arbitration clerk without discussion. If you need to edit or modify this page, please go here and create an edit request. |
Any editor may add evidence to this page, irrespective of whether they are involved in the dispute. You must submit evidence in your own section. Editors who change other users' evidence may be blocked without warning; if you have a concern with or objection to another user's evidence, contact the committee by e-mail or on the talk page. The standard limits for all evidence submissions are: 1000 words and 100 diffs for users who are parties to this case; or about 500 words and 50 diffs for other users. Detailed but succinct submissions are more useful to the committee. This page is not designed for the submission of general reflections on the arbitration process, Wikipedia in general, or other irrelevant and broad issues; and if you submit such content to this page, please expect it to be ignored. General discussion of the case may be opened on the talk page. You must focus on the issues that are important to the dispute and submit diffs which illustrate the nature of the dispute or will be useful to the committee in its deliberations.
You must use the prescribed format in your evidence. Evidence should include a link to the actual page diff in question, or to a short page section; links to the page itself are inadequate. Never link to a page history, an editor's contributions, or a log for all actions of an editor (as those change over time), although a link to a log for a specific article or a specific block log is acceptable. Please make sure any page section links are permanent, and read the simple diff and link guide if you are not sure how to create a page diff.
The Arbitration Committee expects you to make rebuttals of other evidence submissions in your own section, and for such rebuttals to explain how or why the evidence in question is incorrect; do not engage in tit-for-tat on this page. Arbitrators may analyze evidence and other assertions at /Workshop, which is open for comment by parties, Arbitrators, and others. After arriving at proposed principles, findings of fact, or remedies, Arbitrators vote at /Proposed decision. Only Arbitrators (and Clerks, when clarification on votes is needed) may edit the proposed decision page.
Evidence presented by Beetstra
'Magioladitis 1' was strongly dependent on COSMETICBOT
Per Wikipedia:Arbitration/Requests/Case/Magioladitis#Remedies:
- 1:
... cosmetic or non-cosmetic ...
- 2:
... nature of "cosmetic" edits ...
- 7.1:
.. do not affect the rendered visual output of a page ...
The ArbCom requested review on common fixes was heavily dominated by bot operators, with limited community participation
The ArbCom has encouraged the community to carefully review the lists of items in AWB's "general fixes" and the Checkwiki project's list of errors ...
. In this discussion the lists of items/errors was reviewed. 16 out of 20 edits were by bot operators.
The ArbCom requested RfC reviewing the policy on cosmetic edits was heavily dominated by bot operators, with little community participation
The ArbCom encouraged the community to hold an RfC to ... reevaluate community consensus about the utility and scope of restrictions on [cosmetic] edits
. The 'community wide' discussion was initiated and closed by a bot operator. Participation was by 10 bot operators (49 edits, 9 support) and 11 non-bot operators (26 edits, 3 support and 3 oppose).
The ArbCom requested technical means to avoid problematic edits more effectively were not implemented
On 22 March 2017 the developers were encouraged to carefully consider feedback gathered in this case in order to use technical means to avoid problematic edits more effectively.
Last version of AWB is from 3 January 2017.
The request for the first topic ban about WP:COSMETICBOT was brought by a bot operator
""Proposal to topic ban Magioladitis from COSMETICBOT-related discussions" was initiated by User:Headbomb.
The outcome of the first topic ban was heavily influenced by bot operators
The first topic ban discussion was initiated by User:Headbomb and closed by User:Cyberpower678. 7 of the 16 support !votes for the topic ban were cast by bot operators.
The block for the alleged violation of the topic ban was made by a bot operator
- block performed by User:Kingpin13
The thread that resulted in the second topic ban was brought by a bot operator
The ANI thead 'User:Magioladitis high speed editing' was initiated by User:Xaosflux, where it was asserted that Magioladitis has been continuously editing at a high speed
. It was shown that Magioladitis was editing at 13 edits per minute (1 edit in 4.6 seconds).
For the second topic ban, Magioladitis was not editing at an unreasonable edit rate
An editing speed of 13 edits per minute is not unreasonable Naraht (19), Headbomb (13), Ser Amantio di Nicolao (17), BU Rob13 (22), BU Rob13 (19), Beetstra (13), Hmains (11), BD2412 (8), Lepricavark (8), Neve-selbert (22), Koavf (58), I dream of horses (7), Joeykai (8), Hasteur (8) , Frietjes (8), Skr15081997 (8), Rjwilmsi (11), Red Director (7, unassisted), Ronhjones (7, unassisted), WOSlinker (15, unassisted)
Claims about high speed editing flooding watchlists are arbitrary
Despite complaints (e.g. diff, diff, diff, diff, diff), the community has never defined 'high speed editing'
For the second topic ban, no errors in ISBN replacement by Magioladitis were shown
User:Magioladitis high speed editing.
At the start of the second topic ban discussion Magioladitis was not restricted from semi-automated or assisted editing from their main account
- Magioladitis restricted (
Magioladitis is restricted from making any semi-automated edits which do not affect the rendered visual output of a page. ...
) - Magioladitis restricted ..
- .. topic ban Magioladitis ..
The request for this case was brought by a bot operator
- diff by User:BU Rob13, indicating that
the community has frankly lost patience with Magioladitis entirely
.
There is a heavy involvement of bot operators (BAG members) in the election of BAG members
- User:BU Rob13 - revid - 8 out of 13 !votes.
- USer:Headbomb - revid - 7 out of 19 !votes.
- User:Cyberpower678 - revid - 7 out of 9 !votes.
The forelast finished RfC on changes in the bot policy
Evidence presented by Glrx
Current word length: 500; diff count: 1.
...developing...
Magioladitis requested permission to do Magic link/ISBN bot edits
Magioladitis knew that magic links were going to disappear.
- In some pages this will increase the number of transcluded templates by a lot. Magioladitis 9 October 2016.
- Yeah, there's no rush here, as far as I know. The task description calls for "deprecat[ing] magic links on Wikimedia wikis (e.g. Wikipedia), providing alternatives for this functionality and tools to aid the migration." We don't need to act hastily and I don't think anyone is suggesting that we do so. MZMcBride 10 October 2016
- NicoV 26 November 2016:
- Some people started to replace the magic links by some other syntax (templates) on some wikis, and a strange case was also found : it seems that magic links is broken when there are several consecutive filling characters in the text.
- For example, this version of Ahmad Shah Durrani has a broken magic link at the end of the Bibliography. I experimented a bit and found that the magic doesn't work when there are several consecutive filling characters in the ISBN (2 whitespace, 1 whitespace and a dash...). None of the following work for example : ISBN 978- 1-4907 - 1441-7 ; ISBN 978- 1-4907-1441-7 ; ISBN 978 14907 14417 ; ISBN 97814907 14417
- I think we should fix that before removing the magic links because the articles using this broken syntax are probably not listed in the tracking categories, so the bot won't go over them...
On 3 December 2016, Magioladitis proposed a 'bot task to change the links at Wikipedia:Bots/Requests for approval/Yobot 27: "Replace ISBN magiclinks with ISBN template". Also cites to Phab:T148274.
Xaosflux denied Magioladitis's request on 3 February 2017: "Task denied due to lack of established community consensus for a job of this size."
Magioladitis immediately opened discussion
On 3 February 2017, Magioladitis opened Village Pump discussion
https://en.wikipedia.org/w/index.php?oldid=772743896#Future_of_magic_links
conclusion was replace magic links with templates:
- There is consensus to replace the magic links with corresponding templates, and to do this replacement via bot. The consensus for the bot extends only to the conversion of magic links, not any other ISBN/ISSN/etc tasks like linking things that currently aren't linked at all. ~ Rob13 Talk 18:14, 19 March 2017 (UTC)
Bot policy disallows cosmetic edits
WP:COSMETICBOT is a policy that states bots should do not do cosmetic edits on their own.
Bots may do cosmetic edits in conjunction with a substantive edit.
The policy states a substantive edit "affect something visible to readers and consumers of Wikipedia". The policy lists 4 example classes of substantive edits.
- Edits that present a difference to the user.
- Edits that improve other interfaces (such as fixing category sort keys); those changes will be visible on other wiki pages.
- Edits that help the administration or maintenance of the wiki. An example is adding a date to a {{citation needed}} template.
- Bad HTML (such as unclosed tags) even if it does not change the appearance of the page.
Genfixes
There are cosmetic edits that are run when substantial edits are discovered.
AWB and Bot tasks can hold off on some edits until they find a substantial fix.
The impact on watchlists is thus minimized.
There will be a big spurt of trivial changes, so it can be difficult to find the substantial change because it is buried by dozens of
Evidence presented by BU Rob13
Current word length: 915; diff count: 9.
Spamming discussions – first topic ban
Magioladitis bludgeoned the process by creating a slew of discussions regarding COSMETICBOT and general fixes in a very short period of time:
The last several discussions came within the space of a few days. They also came after a holistic discussion on COSMETICBOT, which Magioladitis participated in, at Wikipedia_talk:Bot_policy/Archive_26#WP:COSMETICBOT_update. This behavior led to a topic ban.
Pattern of process spam
Magioladitis submitted 25 BRFAs in a three day period in February. See Wikipedia:Bots/Requests for approval/Yobot 28 through Wikipedia:Bots/Requests for approval/Yobot 52. At the time, he was warned about bludgeoning the process [1].
Mass changes from main account – second topic ban
In the previous case, the Committee reminded Magioladitis to only make mass changes from his main account in a semi-automated fashion as long as "the edits are not contentious".
Magioladitis did large series of edits related to replacing magic links with templates from his main accounts while he was awaiting bot approval. This was challenged by numerous editors, including Justlettersandnumbers and Materialscientist, at multiple discussions. The underlying issue was that making these mass changes semi-automatically is a net negative when bots are carrying out the same task, because doing so spams watchlists with notices. In one of those discussions, he promised to cease such editing.
Up until this point, Magioladitis had behaved properly; he made changes, they were challenged, then he stopped. Nothing is inherently wrong with these edits except that they proved contentious, at which point they must stop until consensus is demonstrated. Note that AWB Rules of Use #3 requires this when using AWB, but it's also required by WP:CONSENSUS and WP:DE. The standard of stopping when challenged is very evenly enforced across the board.
Unfortunately, the same edits resumed days later, spawning this discussion and ultimately a second topic ban.
Topic ban violation and block
Magioladitis almost immediately violated his first topic ban. This was discussed on his talk page at User_talk:Magioladitis/Archive_6#Violation_of_your_topic_ban. An administrator, Kingpin13, warned Magioladitis repeatedly that commenting in certain subsections of Wikipedia_talk:Bot_policy#TIDY that were clearly related to how WP:COSMETICBOT applies to certain tasks is a violation of his topic ban against participating "in discussions concerning the impact of WP:COSMETICBOT on other bot operators (such as whether or not bot operators are allowed, or should be required, to perform WP:GENFIXES with their own bots, or theoretical bots which may be developped in the future)". The most egregious violation was [2]. The original post in this sub-section ended with the question "Is this change cosmetic?", so the section was unambiguously about that topic. He did not desist after several warnings and was blocked.
Gaming a topic ban
Shortly before his block, Magioladitis also created an information page providing a new definition of a cosmetic edit. When challenged on this, he replied "Creating an info page is not connected with it" (see here). This is technically not a topic ban violation, as the ban is on discussing COSMETICBOT, not making bold edits related to COSMETICBOT. Still, it is very clearly the intent of the topic ban to remove Magioladitis from the topic area of COSMETICBOT. Creating such an information page with his own personal definition of a cosmetic edit is gaming the ban.
Disruptive edit to policy
Magioladitis attempted to introduce a substantive change to COSMETICBOT to support his preferred version of the policy [3]. He did this with the misleading edit summary "fix", implying it was a simple language change. This was against the consensus from this RfC.
Behavior toward me
In apparent retaliation for starting this discussion, Magioladitis started a discussion attempting to revoke one of my approvals based on a minor bug that he failed to discuss with me beforehand. Almost immediately after I nominated myself for BAG, Magioladitis opened another discussion on the same issue despite zero errors occurring since the previous discussion. This follows a pattern of using process to retaliate against perceived wrongs.
Magioladitis has repeatedly made passive-aggressive (or outright aggressive) statements relating to me directly or generally to "those who don't wish to improve Wikipedia". It is clear who he's referring to in all such comments. I'm not going to dig too much for many examples, but here's a couple: [4] [5] here
Response to Rich
While you've spent a large section detailing how AWB should not be considered an administrative tool, I see no-one in this case advocating that as a major reason desysopping should be considered. WP:ADMINCOND is what editors have been arguing. Besides, the argument that we shouldn't remove the bit because of misuse of one aspect granted by it is contradicted by WP:ROLL, which explicitly states desysopping is possible when rollback is misused.
Diffs demonstrating inaccuracies or misleading claims
- [6] The editor who actually started the discussion on COSMETICBOT does not believe Magioladitis influenced the outcome of the discussion.
- I did not disagree with rewording COSMETICBOT. My first comment on the proposed reword was "Generally, support" [7]. My second comment said "a major improvement" [8].
- In this discussion, I most certainly did not oppose making it possible to hide AWB edits. I just noted it does not solve a problem.
- WP:DISRUPTSIGNS (particularly #1, #4, and #5) clearly forbids forcing through mass changes over the objections of many editors. This is a behavioral guideline.
- The diff conveniently not linked in Magioladitis' section for 20:04, 19 January 2017 [9] doesn't mention anything about "hostility". I've never stated that Magioladitis' main issue is hostility. The main issue is an inability to understand and follow consensus, as I clearly stated in the diff that he selectively quoted from.
- Magioladitis has claimed he fixed errors that bots could not (in his evidence below). He previously stated that he was using the code for Yobot "manually" (sic - I assume he means semi-automatically, since it involves code). [10] If he was fixing errors using the code from Yobot and that code was "working fine", then the edits could have been handled by bot – his bot. He just got impatient with the BRFA and decided to make the edits from his main account.
- Rich asserted that I said botops would abuse a hole in a proposed policy in this diff: [11]. This is false. I said they could. When constructing a policy, one should ensure what's said in the policy is what the community intends.
Deleted comments on Workshop
Here's a diff of some comments from Magioladitis that he deleted. I'm including them because they show the WP:IDONTHEARTHAT mentality that has been discussed exhaustively and a great example of Magioladitis casting aspersions at me. [12]
Evidence presented by Magioladitis
Current word length: 1160; diff count: 19.
BRFAs procedure since February
- Yobot started editing in 2008. 24 BRFAs were approved the period 2008-2017 (9 years) with CHECKWIKI related tasks approved in 2010. [13]. Some contained multiple tasks.
- No task were ever questioned apart from some connected to the CHECKWIKI errors.
- After a discussion instructing to re-submit all tasks, not only those about CHECKWIKI, separately I started doing it.
- A list can be found at User:Yobot/Tasks 2017-. 28 BRFAs submitted since February. Only 5 got approval till now. This is less than 1 per month while there are 100 tasks to be submitted. Some of the submissions were about the non-CHECKWIKI and since none ever questioned these tasks, one would expect these tasks to be speedy re-approved.
- In most cases the procedure is slow, it's not clear how to gain consensus for the task since links to Manual of Style, links to WP:CHECKWIKI, links to previously approved tasks do not seem to work as a reason to approve the task.
Many editors prefer AWB editing from BRFA process
Many bot requests are done by willing editors via AWB editing. Some examples:
- Wikipedia:Bot_requests/Archive_74#Bryan_.283rd_edition.29_parameter_modification "this request can be shut I've done the lot using AWB"
- Wikipedia:Bot_requests/Archive_74#Tagging_run_for_Wikipedia:WikiProject_Women_in_Red
- Wikipedia:Bot_requests/Archive_73#Replace_deprecated_WikiProject_Chinese-language_entertainment_template
- Wikipedia:Bot_requests/Archive_71#Replace_external_links_to_IreAtlas (done by BU Rob13)
- Wikipedia:Bot_requests/Archive_72#WikiProject_Belgrade_to_WikiProject_Serbia.2FBelgrade_task_force read this very interesting cases about fixing ~1300 pages. The BRFA took 20 days.
Per a research more than 10% of English Wikipedia editing is done by AWB (See for instance Marios Magioladitis, Facts about AutoWikiBrowser, 2015, page 5) It's very difficult to conduct a full research on that but checking the WP:Bot requests pages will show that many tasks are done manually. In various cases getting approval via BRFA is slower than just using AWB in semi-automated way.
My BRFA for ISBN fixes led to RfC and finally to the consensus formed
In December 2016 I requested a BRFA to fix ISBN links. Editors have been converting ISBN templates to magic links (Rich, me, et al.) and vice versa. I noticed this inconsistency. After I got informed that MW programmers are deprecating magic links since October 2016 ([14]) I started the BRFA to fix this the way the WMF staff wanted and against my practise so far. The discussion closed 4 months later. A discussion started in Village pump in February 2017 Wikipedia:Village_pump_(proposals)/Archive_138#Future_of_magic_links. It showed community consensus. I started a new BRFA Wikipedia:Bots/Requests for approval/Yobot 54 in March 25, 2017. One month later. It still took 4 months to get approval (22 July 2017).
BRFAs on ISBNs never properly coordinated
- [15], [16], [17] diff of PrimeBot fixing after Magic links BOT (According to the BRFAs page these two bots use exactly the same method)
- [18] [19] diffs of Magioladitis fixing after Magic links BOT
- [20] diffs of things not fixed by any bot.
- [21] diffs of a link that Yobot fixed and added as fix during the test period. No other bot could fix that.
- Between the two run mentioned by others there was a period of 7 days of no ISBN fixing by me. I resumed to fix cases the bots did not fix
An important part of the community endorsed the manual editing
-
ISBN fixes thanks example
-
More...
-
More...
The watchlist system
During the ISBN fixes process I got a message saying: [22] "it's easy to hide bot edits in a watchlist, but I know of no way to hide yours." The editors seem to be looking for a better watchlist system.
Still the watchlist system does not work 100% perfectly. [23] Many pages are overlooked and some obvious vandalism stays around for even months.
Many editors use Watchlists to follow edits. The phenomenon of having many pages in the watchlist is discussed in Wikipedia:Watchlistitis. The site also proved Wikipedia:Recent changes patrol. During that system editors can Hide/Show (un)registered users, their edits, bots, minor edits, Wikidata, etc. Recently "probably good edits" was added. Still the system does not allow to hide edits from certain editors or certain kind of edits.
Checking broken or unexpected syntax also serves in detecting and fixing vandalism. This is one of the WP:CHECKWIKI purposes.
Various discussions to help systematic work on Wikipedia started
I did a huge effort to form consensus and most of all, to form the policies and the guidelines to assist systematic work:
- On BRFA process: Wikipedia_talk:Bot_Approvals_Group#Should_admins_also_be_able_to_close_BRFAs_under_certain_circumstances.3F
- On encouraging combining bot tasks: Wikipedia_talk:Bot_policy/Archive_27#Should_bots_perform_multiple_edits_in_a__page.3F
- On hiding certain non-bot edits: Wikipedia:Village_pump_(technical)/Archive_156#Hide_AWB_edits_in_Special:Contributions.3F.
What is to be done is exactly to form a consensus of doing things in an effective way. If someone reads carefully the complains they will that the complains are not all of the same nature. Most, if not all, of the discussions here are chronologically after the Evidence presented by Hchc2009. This is a proof that I tried not only to adjust my editing style based on the comments but also help in not having repeated discussions on these matters by forming a stable consensus.
Requested permission to do tasks from main or bot account as instructed
- Wikipedia:Administrators'_noticeboard/Archive291#Request_to_remove_invisible_characters_from_pages led that this task is a suitable task for a bot account
- Wikipedia:Administrators'_noticeboard/Archive291#Request_to_replace_p_tags
- Note, by reading the comments in those two discussions, that the community was in fact unprepared to see the various options that were there.
- This is a proof a respect WP:CONSENSUS
These discussions are chronologically after the Evidence presented by Hchc2009.
Bans are unrelated to cosmetic changes editing; still COSMETICBOT discussions determined the results of these discussions
None of the bans had to do with my bot or my main account performing cosmetic editing.
- The first ban was for discussing various aspects of a newly formed policy in the spectrum of correctly implementing it.
- The second ban was done for "high speed editing" on a task that has consensus, I was editing in lower editing rates than in the past and I was fixing bot edits.
The current case itself states "Magioladitis' conduct since the previous case was closed". This means that the conduct before the previous case is not related and something triggered what it seems to be a different behaviour.
- Wikipedia:Bots/Requests for approval/Yobot 47 was denied [24] as a bot task after it has completed trial test and without violating COSMETICBOT by Headbomb.
COSMETICBOT and human editors
The way the current policies and definitions are written is confusing.
Right now there is no definition for cosmetic edits outside the context of bots. WP:COSMETICEDIT redirects to WP:COSMETICBOT which is a part of bot policy. The term "cosmetic" is associated to the speed of editing. This makes things even more confusing.
WP:COSMETICBOT tries to cover human editing too. It's impossible to say if this series of edits or this edit fails under the a bot policy. Same for [25] or [26] (500 edits in 60 minutes), etc.
[27] An uninvolved editor stared a discussion about the redirect with edit summary "need a definition independent of bots".
COSMETIBOT gives examples of non-cosmetic edits after a discussion I initiated in January
I started a discussion at Wikipedia_talk:Bot_policy/Archive_26#Defining_cosmetic_changes in December 2016. In multiple occasions any try to question the way the policy was written had reactions. See for example this change from 14 January 2017. I questioned the mentions to an old python script and AWB.
The discussion I started revealed that the ideas I added are not opposed by the community. Anomie created a draft in 21 January 2017. Final decision on the new wording take alive at 18 May 2017. It contains no mention to AWB, pybot and gives examples of edits that are not considered cosmetic even if they do not change the visual outcome exactly as I was proposing for six (6) months.
The first oppose was by BU Rob13 4 minutes after the initial discussion started. [28] (See also: Rob's behaviour...).
"Spamming discussions"
The first 3 posts mentioned by Rob happened in January-February 2017 before the fisrt ArbCom
- The first led to a draft rewrite based on my complaints that the COSMETICBOT was outdated and lack basic examples
- The second was a small proposal that was interrupted due to the ArbCom discussions
- The third discussion about the secondary tasks revealed something amazing: The majority agrees with that but this was never implemented as a strategy.
- Arbcom 1 closed with remedy Community encouraged to review policy on cosmetic edits. (22 March 2017)
Rob's Behavior toward me and other bot operators
- Rob registered to Wikipedia 2 years ago 18:36, 7 June 2015 (2 years ago)
- Rob said for me be removed from BAG member "This has been going on for years, and a bot operator that doesn't comply with the bot policy should obviously not be a BAG member" (06:39, 27 December 2016, emphasis is mine). [29]
- Rob comments about me: "The behavior over half a decade is far below what's expected of any editor on the project" (20:04, 19 January 2017, emhasis is mine). [30]
- Rob comments on Bgwhite's page [31] (January 2017) and later on Ladsgroup's page [32] (February 2017). So at the same period 3 bot operators (at least) are contacted and asked to discuss their bot tasks under a hostile environment during an ArbCom that also discusses COSMETICBOT.
- During the ArbCom Rob suggests I am removed from sysop for different reasons from those stated here. [33]: "For failure to maintain the standards compatible with adminship and abuse of administrator tools"
- Rob became BAG member after the first ArbCom. Still says we won't close , I recused from handling any of your bot tasks or CHECKWIKI-related tasks. (11:53, 23 June 2017) but still comments in all CHECKWIKI error related tasks (many diffs can prove this): "I also specifically recused from your bot task" (12:15, 23 June 2017)
- Compare this Wikimania discussion with this one. -- Magioladitis (talk) 06:06, 12 August 2017 (UTC)
Evidence presented by Nick
This evidence solely relates to my closure of the ANI discussion which ran from 30 June 2017 to 7 July 2017.
Magioladitis high speed editing
Xaosflux opened discussion at ANI on 30 June 2017 concerning high speed editing Magioladitis was undertaking from his 'main' account (in this case, their main account refers to the Magioladitis (talk · contribs) account). The initial ANI report contained evidence that the high speed editing that Magioladitis was undertaking was considered disruptive by at least one user (Justlettersandnumbers [34]).
The ANI thread resulted in several experienced users agreeing that the volume/speed of edits being undertaken by Magioladitis using their main account was problematic, with the reasons given primarily relating to the flooding of watchlists, making the monitoring and resulting maintenance of users watchlisted pages difficult or impossible.
Magioladitis, being an administrator, cannot have access to AWB removed in the way a regular editor would. This is due to a decision made by the AWB developers to automatically grant access to AWB for all administrators. (Magioladitis also confirms that in addition to the administrator access to AWB he enjoys here on English Wikipedia, he has global access on projects where he is not an administrator, as one of the developer team).
Two proposals were therefore brought forward. BU Rob13 proposed a block of Magioladitis for a period of one month. This proposal gained very little support, but several respondents suggested banning Magioladitis from using AWB for the period of one month.
Justlettersandnumbers brought forward a second proposal, which was to ban Magioladitis from using AWB for a period of one month. The respondents to this discussion agreed with a ban on AWB use, with differences in opinion relating to the nature and length of the AWB ban, with broadly three options - a one month ban, a three month ban and an indefinite ban.
I concluded two things in closing the discussion - firstly, there was definite consensus that Magioladitis' behaviour using AWB on their 'main' account was disruptive behaviour (and that users were not being unreasonable when they complained about Magioladitis behaviour flooding their watchlists; Magioladitis specifically recognised the issue and discussed the use of a script to hide their edits on watchlists during the discussion). There was, secondly, definite consensus that Magioladitis should be banned from using AWB for a period of time.
The closure was complicated by the competing proposals for an AWB ban, in making the closure there, I tried to accommodate as many of the different opinions as possible, and my closure was to ban Magioladitis from using AWB or any other automated or semi-automated tools on their main account (or any other non bot account) for a period of two months, after which time Magioladitis would have to ask the community whether or not the ban could be lifted. The ban has a clause that definite consensus would be needed to extend (or retain) the ban, otherwise it would 'expire'.
Evidence presented by Rich Farmbrough
Assertion: reason is jettisoned
It is claimed that the only way to ban Magioladitis from using AWB is to de-sysop him. This is nonsense. AWB is software, and the developers could easily hard code a ban. Of course Magioladitis is a developer, so we would have to trust him. We would have to anyway, since he could easily code around a check-page ban. This has come up several times before, and it is absolutely clear that people of a certain calibre are capable of effectively ignoring constraints against productive editing, should they choose.
(And indeed a large segment of the community is happy to wink at such. Others who are rule-bound, prefer drama to the idea that someone is "editing against the rules", even if the drama means damaging the wiki, and the the editing is improving it. In cases where the rules are the editor's own this constitutes unmitigated vandalism (deliberately damaging the encyclopaedia), even when they are the community's rules some argue that the vandalism is merely mitigated.)
Hence it is clear that any call for Magioladitis to be de-sysopped "to deny him acess to AWB" has no basis in reason.
Assertion: Lack of good faith
- Magioladitis suggested that adding or removing a maintenance category should not count as cosmetic (and indeed the policy was amended to include this). BU Rob13 said that bot operators would use this as carte blanche to do as they wish.[35] This shows a complete lack AGF to the entire bot community, and to editors in general. This seems to characterise the attitude of those trying to stop Magioladitis and others from running fix-up tasks.
Apologies
I have not had the time, which would run to many hours, to provide all the evidence I should in this case. I did provide my opinion, as a Wikipedian, which I believe is valid evidence qua the opinion of a reasonably well informed, if not expert, Wikipedian. This has been removed by a clerk, but is still available in the history.
I think this sharpens up the question of why we subject good faith contributors to these proceedings, which consume vast swathes of their time, effectively going counter to WP:NOTCOMPULSORY not to mention the stress that it engenders.
Evidence presented by Headbomb
Current word length: 1058; diff count: 3.
Re: The community did not significantly participate, nor endorse the requested review on common fixes
While "the discussion was heavily dominated by bot operators" is true to an extent, it is certainly not true that the community did not participate, nor that input was not solicited. See in particular
And the last two Bots Newsletters
which were sent to a lot of people, including WP:BOTN and WP:VPT.
That non-technically oriented people didn't want to participate is on them.Headbomb {t · c · p · b} 15:46, 25 July 2017 (UTC)
Re: The community did not significantly participate nor endorse the requested review on the policy on cosmetic edits
This is complete nonsense. We had a well-advertised RFC on WP:COSMETICBOT, which featured in WP:BOTNEWS, was cross posted at WT:BOTPOL/WP:VPT/WP:BOTN and at WP:CENT too. The language was both drafted and supported by everyone in the BAG (at least those that commented), the RFC was open for comment to everyone, and ran for a month, closing with "a clear consensus exists to adopt the proposed language".Headbomb {t · c · p · b} 15:51, 25 July 2017 (UTC)
Re: Carefully review BRFA scope
In response to ARBCOM's non-binding request, the BAG created WP:BAGG, a guide to BAG tasks and duties. Particularly relevant is the guidance "a BRFA with multiple subtasks should demonstrate technical soundness and consensus for all subtasks during the trial..." which is something that had sometimes been overlooked/AGF'd in the past.
We have also significantly updated WP:BOTISSUE, to better guide the community on what to actually do when issues arise. Headbomb {t · c · p · b} 16:01, 25 July 2017 (UTC)
Re: Bot policy disallows cosmetic edits
An important clarification here is that WP:COSMETICBOT disallows cosmetic edits by bots, unless there is consensus for it. This is an important distinction. That currently ISBN 978-3-16-148410-0 and {{ISBN|978-3-16-148410-0}} render the same is irrelevant, as there is consensus to replace the former with the latter because they won't render the same in the future. This is perfectly allowed under WP:COSMETICBOT.
I also note that COSMETICBOT states "human editors may also wish to follow this guidance... especially if making such changes on large scales." with a link to WP:MEATBOT which states "... it is irrelevant whether high-speed or large-scale edits ... are actually being performed by [a bot or human]. No matter the method, the disruptive editing must stop or the user may end up blocked." Headbomb {t · c · p · b} 16:16, 25 July 2017 (UTC)
Re: COSMETICBOT and human editors
Magioladitis is the only bot operator confused by what WP:COSMETICBOT means.
- "Right now there is no definition for cosmetic edits ouside the context of bots." [sic]
- The definition applies regardless of if the edit was made by a bot or not. The policy part (i.e. don't do them on their own, unless you have consensus for it) only applies to bots and bot-like editing. Editors are free to make cosmetic changes on small scales (e.g. [36], part of a larger copy-editing run). I will additionally point out that all bot/bot-like edits are subject to consensus, regardless of whether they are considered cosmetic or not.
- "Cosmetic is related to speed of editing too making things even more confusing." [sic]
- The only relation between editing speed and cosmetic bots is most bots are designed to have high-speed, high-volume edits. This is a general feature of bots, and a recognition that malfunctioning/badly-designed bots can cause great levels of disruption.
- "WP:COSMETICBOT tries to cover human editing too."
- As a policy, it only covers human editing when it reaches WP:MEATBOT levels.
Headbomb {t · c · p · b} 12:41, 28 July 2017 (UTC)
Re: There is no policy based reason that editors should stop on complaints that their high speed editing is 'disruptive' due to 'flooding watchlists'
There absolutely are policy based reasons for that.
- WP:CONSENSUS - All edits are subject to consensus.
- WP:MEATBOT - High-volume/High-speed editing falls under WP:BOTPOL for the purpose of dispute resolution
- WP:BOTREQUIRE - Bots are required to make edits that "[do] not consume resources unnecessarily" and "performs only tasks for which there is consensus"
- WP:COSMETICBOT/WP:BOTFLAG/WP:HIDEBOTS/Help:Watchlist#Options
Bot edits are community-vetted as a) harmless b) useful c) have consensus d) adhering to relevant policies and guidelines. Via the WP:BOTFLAG, they can easily be ignored, both in the recent change feed and in watchlists. And despite being community-vetted and easily ignorable, those are still subject to restrictions because we don't want them to clutter watchlist [bots do sometimes malfunction]/page histories with useless or near-useless stuff. High-speed/high-volume non-bot edits are not community-vetted, cannot easily be ignored, and will also clutter talk page histories, watchlists, and also the recent change feed. They have substantially higher reviewing costs than bot edits, and require consensus like anything else on Wikipedia. If consensus is that an editor's high speed editing is disruptive, they absolutely must stop it per WP:DE. Headbomb {t · c · p · b} 14:04, 29 July 2017 (UTC)
Re: For the second topic ban, Magioladitis was not editing at an high unreasonable edit rate
That Magioladitis' edit rate was "reasonable" and comparable to that of other people doing semi-automated work (or other bots) is irrelevant, other than to show he was editing in a bot-like manner, and that WP:MEATBOT applies. To quote WP:BOTREQUIRE
- "bots doing non-urgent tasks may edit approximately once every ten seconds, while bots doing more urgent tasks may edit approximately once every five seconds."
I'll note here that this section of the bot policy is a bit outdated and was never meant as a strict "THOU SHALL NOT EXCEED" guideline, but it does provide a ballpark/typical figure. The section exist mostly to limit the damage in case of malfunctioning bots, but also to not overwhelm server capabilities during times with high server loads. I'll point here that watchlist flooding was never a concern with "edits-per-minutes" limits. Editing 1500 articles over 1 day (~1 edit per minute) would cause just as much watchlist flooding as 1500 edits over one hour (~25 edits per minute), and both can be considered bot-like editing.
And while it's true that high-speed editing is not by itself disruptive, it certainly can be. It is the disruptiveness and the contentiousness of the edits made that lead to the second topic ban (see #Mass changes from main account – second topic ban). Headbomb {t · c · p · b} 14:15, 29 July 2017 (UTC)
Re: Magioladitis was not restricted from automated editing from their main account
While true that the first case did not restrict him from automated editing in general, it did restrict him from making cosmetic changes, which culminated in a general ban on semi-automated editing.
Re-reading this, if the claim here is that Magioladitis was not restricted from automated editing, rather than semi-automated editing, that's obviously untrue. Everyone is restricting from running automated processes from their own account. See WP:BOTACC/WP:BOTBLOCK in particular. Headbomb {t · c · p · b} 14:22, 29 July 2017 (UTC)
General remarks
If these proceedings are about de-sysoping Magioladitis, then IMO most of the evidence presented here has nothing to do with whether or not Magioladitis is fit to be an admin. I mean, diffs of User:PrimeBOT and User:Magic links bot? Assertions that the community doesn't participate enough in technical areas? Debates on "are people bot"? (And I'll remind everyone that for purpose of dispute resolution, they are.) Arguing that ISBN changes are cosmetic and disallowed/allowed? Absolutely none of those things have anything to do to do with Magioladitis's behaviour / fitness to be an admin.
The evidence should focus on actual things Magio did, or didn't do. The relevant policies here are
And
As well as other applicable restrictions. Headbomb {t · c · p · b} 16:50, 25 July 2017 (UTC)
General remarks (2)
For some reason Dirk Beetstra makes a lot of fuss about that some actions against Magioladitis were done by bot operators. Two things
- Bot operators are part of the community just as much as anyone else. Our voice is no more or less important than anyone else, although as bot operators, we're certainly more qualified to comment on technical and bot issues.
- Every bot operator mentioned in his evidence is actually a member of WP:BAG. That is User:Headbomb, User:Xaosflux, User:BU Rob13, User:Cyberpower678, User:Kingpin13.
- As a reminder, WP:BAG is a group of users, who "oversees most areas and processes dealing with bots on Wikipedia[...] are tasked to approve or deny the various bot tasks submitted [... and] are trusted to understand Wikipedia's bot policy, and to offer sound bot-related advice to bot operators, admins, bureaucrats, and editors alike." Every BAG member is vetted by the community in an open and well-publicized process.
- Cyberpower wasn't a BAG member at the time of his close, but subsequently got elected to be one. Magioladitis used to be one, but lost membership during the first ARBCOM case.
Headbomb {t · c · p · b} 17:15, 3 August 2017 (UTC)
Evidence presented by Hchc2009
WP:IDIDNTHEARTHAT and WP:CONSENSUS
Magioladitis has continued to edit in a manner that suggests "I didn't hear that", with many editors noting his increasing inability to assess consensus.
This first came to a head in December 2016, when Xaosflux and myself highlighted the community concerns over the large scale edits he had been approving. In the resulting renomination process, there was overwhelming opposition to his renomination. Comments in that debate included "He seems not to understand the need for consensus or that repeated complaints mean he must stop the task. Communicating with him is not easy. There are dozens of talk-page complaints about his or Yobot's edits, many showing a failure to take the point. If an editor complains about X and mentions Y in passing, Magioladitis will focus on Y and ignore X. This wastes a lot of time." "He seems completely incapable of discerning what amounts to consensus, and I'm not sure that he cares much about it anyway". He "can't be trusted to care about enwiki community consensus". He is "is unable to follow the policy and reasonably resolve disputes".
There was a similar view expressed by the community in June 2017, during the COSMETICBOT topic ban discussions: "it's necessary at this point so we can stop having basically the same discussion over and over again", "much of his energy is directed towards ensuring the wikitext is perfect, even to the detriment of the good functioning of the project", "it's clear that this is just going to keep repeating itself until a topic ban is enacted". "This is the very definition of WP:IDHT. If this was a non-admin, they would have been indef'd years ago." "Mag's badgering of every oppose only reinforces my view that they cannot self restrict the concerning behavior". A substantial number of editors took part in the discussion with an overwhelming consensus for the result. Magioladitis broke the topic ban soon afterwards in July, leading to his being blocked.
Finally, in June/July 2017, a number of editors complained about Magioladitis' approach to high speed editing. Comments from editors reflected the same pattern as before, including:"I'm perplexed and disappointed at his disregard of the concerns of others", "he's barely listening to anyone, barely following the rules, and annoying many people in the process", "Mag is not interested in what others think", "he simply doesn't get it", "Any other editor would have long since seen blocks for probably the most dramatic IDHT I've seen in my many years of lurking", "If Magioladitis hadn't been an admin, he'd at the very least have been stripped of the AWB permission and almost certainly be community banned from the project by now".
Evidence presented by Bgwhite
Magioladitis made me aware of this case a few weeks back. I hadn't planned on saying anything, however, some of the same reasons and people that drove me from here are evident again. Just remember, anything decided here will be used against others to stop editing via AWB or editing altogether.
- Speed: Editing at 15 edits per minute can be too slow. Editing at 5 edits per minute can be too fast. It all comes down to the experience of the editor and what they are doing. I don't know how Magioladitis was making his ISBN edits with AWB, but if it was a find/replace, 15 edits per minute can be too slow. This kind of edit is the same thing that bots do. It's the same kind of edits Ser Amantio di Nicolao does with AWB running upwards of 50 manual edits a minute. When I was fixing broken brackets, 5 edits a minute would be too fast to sustain. Finding where a bracket fix goes could be simple, but could also be complex. If a bot can run 20 edits per minute, why are manual edits with the same type of task limited to around 5?
- Watchlists: The idea one cannot edit because it may fill up somebody's watchlist drives me crazy. Forbidding edits because it may be an inconvenience to some is ridiculous. Doing 5 or 15 edits a minute is still going to fill up watchlists. Running a bot is still going to fill up watchlists. Very few turn off bot edits on their watchlist. People have said they want bot edits showing up on their watchlists because bots make errors or do something they don't agree with. I got complaints about filling up watchlists on manual and bot edits. Filling up watchlists should never be an excuse to stop legitimate edits.
I just see this case being used like the last case, putting arbitrary limits on legitimate editors and drumming people out. I warned the arbs last time they would come after me and they did. Is Ser Amantio di Nicolao going to be told what took him 10 minutes before will now take two hours because of watchlists and editing too fast? Or is this a case of Magioladitis having a bullseye on his back... not allowed to function the same way others already do with no problem? Bgwhite (talk) 21:33, 13 August 2017 (UTC)
Evidence presented by {your user name}
before using the last evidence template, please make a copy for the next person
{Write your assertion here}
Place argument and diffs which support your assertion; for example, your first assertion might be "So-and-so engages in edit warring", which should be the title of this section. Here you would show specific edits to specific articles which show So-and-so engaging in edit warring.