Wikipedia talk:Requests for arbitration/Betacommand 2/Workshop

Structure of workshop
Just a note, I plan on removing the sections of my "proposals" section that are unused—e.g., remedies, which I do not plan on proposing at this time. I assume this is acceptable. AGK § 17:36, 16 March 2008 (UTC)
 * I did the same thing already.  Lawrence  §  t / e   18:31, 16 March 2008 (UTC)
 * Can I suggest that anyone who makes their own "Proposals by me" section remove any unused templates does this, or at most leave one to copy off? The page is about fifteen screens long already.  Happy‑melon 18:48, 16 March 2008 (UTC)

My principles
See: Requests for arbitration/Betacommand 2/Workshop. Please weigh in, thats probably going to be my only contribution, and what prompted me to leave a statement in the first place. I ask that the AC add these questions to the Proposed Decision, since this disconnect (see all the times Betacommandbot has been blocked and yelled at, for example) of authority over bots and the BAG seems crucial here. This really needs to be defined, because I'm sure we're sick of the nigh daily "I BLOCKED BCBOT" threads and incredible hate they generate. Lawrence §  t / e   18:31, 16 March 2008 (UTC)

How does this work
This is the first ArbCom case I've been at all involved with, so I'm wondering how this page tends to develop, and any conventions that might not be obvious. Are editors, for instance, encouraged, discouraged, or prohibited from editing the wording of other people's proposals? Is there usually a flurry of proposals which are consolidated into a few before acceptance, or do the Arbitrators and clerks cherry-pick ones that look the best from a larger number of (possibly similar) proposed wordings? Happy‑melon 18:56, 16 March 2008 (UTC)
 * You're generally forbidden from editing others' sections, but feel free to post whatever you want otherwise. It's expected that the pages will get long and unwieldy; the long and acrimonious IRC and Mantanmoreland cases both generated around 2000KB of discussion. east. 718 at 19:13, March 16, 2008
 * OK, thanks. And the other question? Happy‑melon 19:51, 16 March 2008 (UTC)
 * The arbitrators, in my experience, do read all this, but often just write their own proposals and remedies, Some arbitrators participate directly on the page (posting proposals and commenting on those by others), others don't. They have their own mailing list and private wiki, and I suspect some concentrate mostly on the evidence. But possibly the discussions here by the community help the arbitrators gauge community feelings. So in effect, say what you want (remembering to keep it calm and considered), but don't expect your carefully worded remedy to be lauded as the perfect solution and used in the final decision. On the other hand, there have been changes to the way things work, so maybe what I just wrote was all wrong. Probably best to wait for an arbitrator to weigh in here. Carcharoth (talk) 21:27, 16 March 2008 (UTC)

Is any one person so valuable they can't be blocked or sanctioned?
That seems to be the implication of a lot of the responses to proposals here ("You can't block or restrict Beta unless you can replace him"). Interesting precedent, there: is anyone above restriction or sanction? What would have to be the circumstances for that? Lawrence §  t / e   19:43, 17 March 2008 (UTC)
 * This is exactly what I was thinking when I read some of these responses: "are some editors so valuable they're exempt from the policies and guidelines other editors are held to?" My answer would be no, of course not, and any remedy that would be appropriate for another editor should be appropriate here as well, regardless of the value of his work. —Locke Cole • t • c 19:46, 17 March 2008 (UTC)
 * I don't believe he's unreplaceable. I think it might be better to sanction him short of taking his bot away, however, and have said exactly that.  A bot on parole, perhaps?  A long accepted principle of arbitration is to avoid imposing sanctions greater than are necessary. --Anticipation of a New Lover's Arrival, The 19:47, 17 March 2008 (UTC)
 * I agree with Tony here. I think something with teeth needs to be done, but taking the bot away seems a bit draconian. Perhaps the enforcement can be structured such that there is a "1 strike you're out for a week" rule, escalating to indef if the problem doesn't go away after a set number of "escalating blocks." One way or the other, something must be done. Bellwether B  C  19:55, 19 March 2008 (UTC)
 * I think the point is that Betacommandbot does a necessary job, and if you were to sanction Betacommand then someone else should do the work of tagging non-free images. The job Betacommandbot does is not at issue here, the conduct of Betacommand (and perhaps the exact operating details of Betacommandbot) is what is at issue here. -- RoninBK T C 21:13, 19 March 2008 (UTC)
 * To what extent is that addressed by having someone else run the code as nfcc10c bot? Gimmetrow 21:33, 19 March 2008 (UTC)
 * That's an entirely plausible scenario, You could run your own bot, (I don't know the copyright status of BCB's code is,) or for that matter incorporate BCB's functionality into the wiki itself, into the post-3/23 image-upload process. -- RoninBK T C 22:15, 19 March 2008 (UTC)
 * BCB is closed source code. Currently no one else is permitted to run it on their own machine. User:Lar has a copy of that portion of the code, but it prohibited from sharing it.  Other users have access to the control to turn on that code for NFCC Bot, but I do not believe they have the code.  Therefore, it would need to be open-source re-written to run under another user or be incorporated into Wikipedia.  MBisanz  talk 22:21, 19 March 2008 (UTC)
 * From what I've gathered in the discussions I've been involved in, the image portion of what BCBot code would be quite simple to just write from the ground up. I don't know about the other portions, though. Bellwether B  C  23:08, 19 March 2008 (UTC)
 * Yes the basic idea is very simple, but due to issues that I have discovered while running the bot it has become very complex building in unique issues that arise. βcommand 23:10, 19 March 2008 (UTC)
 * Such is the case with closed-source, you could theoretically replicate the functionality, but you do not gain the benefit of the troubleshooting history of the initial programm. It's an interesting irony that a program that is flagging non-free items in an open-source project is itself non-free, but that's apparently not a requirement for bots. -- RoninBK T C 23:35, 19 March 2008 (UTC)
 * If I had more faith in the community it would be more open, but given the fact that there are users who game what I say to avoid writing rationales, insult my profession, make countless personal attacks, and what some would call trusted users (the antichrist Archtransit incident) lie to your face, I have lost my trust with the average user. there have even been cases of standing ArbCom members (at the time) un-blocking a user who knowingly made legal threats, (and actually took steps in filing a case) against a long standing, valued member of the community who was then forced to retire. so what makes me willing to blindly trust such a group with anything sensitive? βcommand 23:45, 19 March 2008 (UTC)
 * I'm not trying to disparage your reasons for withholding your rights, they are exactly that, your rights. Forcing you to turn over your code would run contrary to the principles your bot aided in protecting.
 * I would note observationally that the complete depletion of your ability to assume good faith anymore is probably a large portion of why this case is here, -- RoninBK T C 00:41, 20 March 2008 (UTC)


 * Please note that assuming good faith and trust are completely separate. I know there are bad apples out there. I approach most situations with with a neutral point of view, examine the facts and then come to a conclusion. yes I know a majority of the community is here for the betterment of wikipedia. But the flip side of that is you must admit that there is a part of the community that their sole propose is counter to what wikipedia stands for. there is no need to go into detail we all know a spammer/POV pusher/long term vandals/trolls. this comes from long term observation from all sides of the sphere. Wikipedia as a whole also acknowledges that, otherwise we enable anon page creation, wouldnt have captchas on account creation, or IP users who edit external links, AWB would not have a checkpage, and rollback would be given to all users. My position in regard to code is fairly simple. Verify that the user is not going to abuse it. how often is Twinkle abused? at one point there was at least 3 post on ANI at any given time. Just like AWB, you have to prove that you can be trusted with a very powerful tool. βcommand 01:05, 20 March 2008 (UTC)
 * Can you please explain what you found objectionable about Nandesku, when he asked for access? It has been my experience that you deny access to the code based as much upon whether you like or dislike the person as their general trustworthiness. Bellwether B  C  01:10, 20 March 2008 (UTC)
 * there were comments made by Nandesku that gave the impression that if Nandesku received the code he/she would not abide by the restrictions that I have about release of the code. and that the motive behind Nandesku's actions were to get the code released against my wishes. βcommand 01:14, 20 March 2008 (UTC)
 * So you were assuming bad faith of an established editor (and I believe and admin, as well), without any real evidence? Bellwether B  C  01:21, 20 March 2008 (UTC)
 * Given the extremely hostile and negative approach by Nandesku, gave me no reason to trust him/her. like I have stated trust is not equal to AGF. βcommand 01:26, 20 March 2008 (UTC)
 * see also this edit summary. βcommand 01:45, 20 March 2008 (UTC)
 * (To Betacommand) Um, I assume you mean the Archtransit incident, rather than the Antichrist incident, unless there is a recent development that I am missing. :) Newyorkbrad (talk) 00:48, 20 March 2008 (UTC)
 * To BC: The quote above, "If I had more faith in the community" is quite a demonstration of one of the major problems here. A user who no longer has faith in the community needs to either take a break or leave altogether. BC does good work. But his complete lack of AGF is breathtaking. Bellwether B  C  00:50, 20 March 2008 (UTC)


 * Maybe this is a silly question but why can't the task of the bot be split and ran by a number of operators. Obviously, there is so much work done by Betacommandbot that unresponsiveness is bound to happen. (Though I happen to think Betacommand has been less responsive than what the situation dictates, but that's another story.) Why can't an informal committee develop and run Betacommandbot, either by splitting its various tasks or simply working in parallel? I'm sure a number of experienced bot operators would be willing to help in that respect. Responsiveness to complaints would be improved, bugs would be addressed more quickly, tempers would not flare as often. Reading the discussions around here, people generally seem to agree that a) the bot is doing important work and b) Betacommand is not operating it responsibly. Betacommand's comments above with regards to sharing his code seem childish: sure, he has every right to withhold the code but his rationale regarding the trust he has in the rest of the community is precisely the kind of siege mentality that poisons Wikipedia. We're not going to get anywhere with editors who suspect that they are one of the precious few who understand the project, cherish it and know what it needs to develop. Note that the committee approach I'm suggesting here does not require that the code be made public. It simply requires that the code be shared with other experienced bot operators, and there isn't a dramatic shortage of those. I think it's also important to point out the following obvious fact: it so happens that Betacommand built this bot but if he hadn't then someone else would likely have written the equivalent. If Betacommand gets run over by a bus tomorrow (not that I wish him that of course), then his bot is lost forever (unless, you know, if he wrote his testament accordingly). This would not be the end of the world. Someone would have the patience to build some sort of NFCC bot and in a couple of months, that someone would be drowned in complaints. Pascal.Tesson (talk) 20:47, 26 March 2008 (UTC)

How will this be affected by the March 23 deadline?
(Not sure where to bring this up, so I'll post it here.) We're coming upon a major deadline thanks to the Licensing policy resolution, which will likely occur during the duration of this case. I am not clear about how Betacommandbot would operate after March 23, (if allowed to continue without sanction,) and how that could/would/should factor into the proposed remedies here. -- RoninBK T C 21:38, 19 March 2008 (UTC)
 * I believe the Arbs are aware of the deadline through their reading of the evidence, but feel free to add a line on that to evidence.  MBisanz  talk 22:23, 19 March 2008 (UTC)
 * I would like to add here, for the record, that I would absolutely love it if someone else could take the initiative and start a centralised and well-advertised discussion about what the March 23 deadline means and how Wikipedia should respond to it, and what will change (if anything). I've said a few times before that I fear people will take the expiration of this deadline as an excuse to go on some Foundation-mandated deletion "on sight" process, instead of the current system of tagging and allowing time for images to be fixed (where possible - obviously not all images are suitable for fixing). I hope my fears are groundless, but I haven't been able to find the time to start such a centralised discussion, though my previous attempts are given in my statement at the request to open this case. I will actually be away when this 23 March deadline arrives, so I would appreciate if people could make sure any discussion that does ensue doesn't actually result in anything unless there has been a long discussion and clear community consensus. ie. No knee-jerk reaction to the passing of the deadline, and fair warning if any major changes are planned (even those handed down by the Foundation). Carcharoth (talk) 01:11, 20 March 2008 (UTC)
 * There are loads of things that should be written in light of this case such as a guide on how to write rationales that is newcomer-friendly, not just that. x42bn6 Talk Mess  04:57, 20 March 2008 (UTC)
 * Someone might care to ping User talk:Mindspillage since she was on the board when this was passed, seems to know about IP and related legal stuff, and might be able to offer insight into what sort of deadline this is.  MBisanz  talk 08:15, 20 March 2008 (UTC)

List of image bots (link)
People following this arbitration case might be interested in the following: Wikipedia talk:Bots/Approvals group. As you can see, BetacommandBot is not the only bot doing image work. Carcharoth (talk) 01:15, 20 March 2008 (UTC)


 * That's a bit misleading though. There are only 3 bots on that list that are doing NFCC10 work. One is Beta, one hasn't edited (NFCC10 Bot), and the other one seems to be testing (FairUseBot) and is not in actual use.--128.226.198.100 (talk) 03:49, 20 March 2008 (UTC)


 * STBotI actively works with NFCC#10 (a recent approval). - AWeenieMan (talk) 15:08, 20 March 2008 (UTC)


 * This case shouldn't be just about NFCC#10c (note the 'c' - the IP note above said NFCC#10, a common mistake to make). Other bots do NFCC work without generating the opprobrium that BCBot has. And there is more to image work than just NFCC work. Fundamentally, I think this is because there was insufficient planning surrounding the whole BCBot business (and I should have helped more when I became aware of things in the later half of last year). Better planning would have resulted in less aggravation. Carcharoth (talk) 19:36, 24 March 2008 (UTC)

Refactor
Would it be possible for a clerk to refactor the empty sections (except the remaining USER:Z)? I think we could save a few bytes and a lot of real estate that way. &mdash; Arthur Rubin (talk) 16:50, 20 March 2008 (UTC)