Wikibooks:Requests for permissions/Lcarsbot

+Bot
I would like to apply for a bot flag for my bot. It will be used solely to do mass categorisation work on uncategorised pages. It will use the AutoWikiBrowser, and while it will be automated I will be sitting at the PC so if anything goes wrong I will quickly be able to fix it. Lcarsdata 11:03, 17 June 2007 (UTC)


 * -- I had recommended that he get a bot flag for using AWB, so that it won't choke up the RC feeds for our RC patrollers. -- SB_Johnny | talk 11:42, 17 June 2007 (UTC)
 * Note: I provisionally gave him the flag for now, since he's already using the account. If there are any issues, we can always revoke it. Select "show bot edits" on watchlists or RC if you want to keep an eye on it. -- SB_Johnny | talk 11:54, 17 June 2007 (UTC)
 * Whoa, that's really not how it works here and you know that. I'm not going to revoke this bot's rights since I'll wait for the community to decide but this is not how we work around here. A person who wants to run a bot here should establish themselves here a little bit, not just start running it (and why did this user get an express bot flag?). This user seems to also be running a bot under his/her main account as well. I can't stand bots running rampant so I'll make this an . Tell us what you plan on doing before you edit thousands of pages. -within focus  15:10, 17 June 2007 (UTC)
 * Actually, there is precendent... provisional flag was added for a similar reason when we had a bot flagging images. I gave the flag because it was seriously f**-ing up the RC page to have so many rapid edits going on. Anyone can feel free to revoke it, but unless you're going to say "don't organize, please", I'd strongly recommend maintaining the flag.-- SB_Johnny | talk 16:55, 17 June 2007 (UTC)
 * I agree that destroying RC is definitely a bad thing, but the action should not be to make the user a bot but to block the user. Just running your bot across the whole project is beyond being bold to me. This kind of thing needs to be discussed beforehand, especially since two accounts involved in this are doing bot actions. This user should suspend all activity until we can all hear what's actually going to happen. Categorizing, although helpful, shouldn't just get whipped out. I know I haven't been especially active lately, but I haven't heard of any of this. The use of two accounts, both relatively unknown here, is especially troubling. We can do this more professionally. -within focus 17:13, 17 June 2007 (UTC)
 * The "precident" that Johnny is referring to can be found HERE. I granted the bot flag to User:HDBot in response to discussions that had been happening on the staff lounge. I also provided information as to the limits and timeframes of the bot flag, and what work specifically needed to be done. The difference in this case is that there was no prior discussion on this subject, and that there really aren't any definite limits provided: how long will the job take? what precisely is being done ("organizing" or even "categorizing" are very vague)? at what point do we revoke this temporary bot flag? under what conditions do we block the bot from editing? --Whiteknight (Page) (Talk) 17:18, 17 June 2007 (UTC)
 * Guys, Lcarsdata was recruited by the foundation's volunteer coordinator to help "clean up wikibooks". We've been talking endlessly about the need to get this done, but with the exceptions of herby and darkcode (and probably others I'm not aware of), we've all been too busy writing or doing other things. The flag was given provisionally, since he was already using the bot (doing good stuff), and completely choking RC. There are probably several thousand bot-appropriate things to be done yet as well, and since he's willing and able, I'd say let him do it because we've been unable to do so.
 * As far as him being an "unknown user", just consider him to be acting under my supervision during this provisional stage. If his bot causes problems, I will work with him on fixing them (I already did that today)...bots can clean up their own messes just as easily as they can clean up messes other people make. Lcarsdata's trustworthiness and good faith is absolutely beyond question. Bot flags are as easily revoked as given, and if a bot is introduced to the project in order to clear a backlog for us, we shouldn't subject the volunteer-owner of the bot to the third degree. I'm sure he'll be willing to wait for the RFBF to finish before continuing, but IMHO we should encourage people like him as soon as they volunteer, because we desparately need this sort of connection with projects like commons. -- SB_Johnny | talk 22:21, 17 June 2007 (UTC)
 * I'm certainly not saying anything negative about Lcarsdata, I'm mostly making a blanket warning against the use of bots without the proper forethought and supervision. When I use a bot myself, it's typically on fewer then 20 pages at a time, and I'm watching the live RC feed while I do it to make sure things go right. Between the time it takes to set up the bot's task, watch the RC feed, check the edits made by the bot, and possibly make corrections, It typically takes me more time and effort then it would if I had done the work by hand in the first place. Again, i'm not saying anything negative about Lcarsdata, and I'm not even asking that he stop doing the job he was doing. I am, however, saying that bot flags be given out after discussion and community consideration, because they are an issue worth careful consideration. I have removed my oppose vote below, but I would like to see the bot in action again before I cast a vote in favor of the flag being kept. Also, I would like to discuss some kind of moratorium: how long will the bot flag be kept by this user in the absence of affirmative community decision to grant that flag after the fact? If the community does not decide to give the flag (or worse yet, if the community actively decides not to give it) we need to remove that flag in accordance with the process of consensus. I would say in the span of 1 week (by June 24) if the community has not voted in the affirmative, or if it has not progressed significantly towards that end, that the flag should be removed. --Whiteknight (Page) (Talk) 23:22, 17 June 2007 (UTC)
 * Now that I've heard and understand what's actually going on here, I have no problem with the bot going back to work. I like categories. . -within focus 13:44, 18 June 2007 (UTC)


 * - known to me as a good user, won't break anything! -- Herby talk thyme 14:57, 17 June 2007 (UTC)
 * Comment - my vote was not an endorsement of the granting of the flag -- Herby talk thyme 15:53, 17 June 2007 (UTC)
 * Humm!? Then you should remove the Support as that is what is being requested here, change it to a comment... --Panic 19:42, 9 August 2007 (UTC)


 * - In a similar mood as Withinfocus, Unless there is a clear and pressing need for it, I really can't see a reason to grant an express bot flag without getting community input on it first. Since we don't really have a bot policy around here (and the community has been pretty disinterested in creating one), It's really imperative that we get community input on these issues beforehand. Also, since bot's have an unparalleled power to disrupt the project (a human cannot possibly edit so many pages so quickly), it is worthwhile for us to monitor the bot. All the bot flag does it make it more difficult for other users to keep track of the bot's actions, and that means that when a mistake is made fewer eyes will be watching. For future reference, I think that any bot should operate without the flag for a sufficient period for the community to see it's actions and become comfortable with it's operation. --Whiteknight (Page) (Talk) 15:40, 17 June 2007 (UTC)
 * As an example of what I am talking about see this diff. All bots have some kind of bugs, and when you spread a bug over many pages very quickly, it can be a huge mess to clean up. All I am advocating is a little bit of caution when we are allowing people to use tools like this. --Whiteknight (Page) (Talk) 15:45, 17 June 2007 (UTC)
 * It may be worth noting that several of the (especially) older images, though not tagged, were described as being "Public Domain" or "GFDL" or what have you in the image description. See for example this diff. I'm not sure how many of these there are but if the untagged images are to be deleted it must be done with caution. Mattb112885 (talk) 16:02, 17 June 2007 (UTC)
 * Yeah, those will be taken care of (the bot will not be deleting). -- SB_Johnny | talk 16:57, 17 June 2007 (UTC)


 * There is nothing that can be done regarding Matt's point, the images obviously won't be blanket deleted so there should be no problem providing the deleting admin takes care to fully read anything on the image page - this has been done in the past and, as it has been completed now, should not need to be done again for a while. Regarding what the bot is going to do in future: currently there are many, many, pages which are uncategorised and also many other pages which are incorrectly categorised. It is extremely tedious to go through each and every one - especially where there may be hundreds of pages which need categorising. What the bot will do is go through a selection of pages and where they do not already have a category add the category that needs to be added to them. If a page is already categorised then it will not be altered and must be done by hand. I hope this clarifies things. Lcarsdata 16:50, 17 June 2007 (UTC)


 * - I'm not sure why we even need mass categorisation work on uncategorised pages. At least we need to determine what pages ought to be categorized, why, and in what format that categorization ought to take place.  Some unilateral (and without any major community discussion of the topic) categorization efforts have taken place in the past, but to apply a principle of Wikipedia to Wikibooks without any sort of forethought is something that doesn't make sense at all.  The complaints I've had in the past of similar categorization efforts is that they solve symptoms (reducing the number of uncategorized pages) rather than the problem (trying to find content).  As far as untagged images are concerned, that is a seperate issue, but even then, why is the bot flag needed for what is admittedly a one time use operation?  --Rob Horning 22:55, 17 June 2007 (UTC)
 * If volunteers are willing to apply categories to pages, they should be so applied. I can't see a reason not to categorize a page, although I also don't see a strong necessity of demanding pages be categorized either. If this bot operator is willing to categorize pages, so long as that categorization doesnt cause more problems then it fixes, then I see no reason to stop him from doing that. Also, the "we aren't wikipedia" mentality doesnt mean that we should avoid doing anything the "wikipedia way" just for the sake of being different. The fact that categorization is perhaps "over-used" on wikipedia is not an indictment of our use of categories here. While perhaps it might be useful to come up with clear community guidelines on the subject, I think enough pages have been categorized in a pretty standardized way that we could write a policy/guideline to simply codify common practice. You are right however, in that the bigger problem is that of organizing content: the simple application of categories to pages does not make content any more well-organized or any easier to find, although indexing efforts such as the WB:CCO could be re-vamped to help bring some order to pages that are already categorized. I don't want to get into a long discussion here about organization, but the point stands that if a volunteer wants to apply page categories, they should not be dissuaded from it (although the necessity of the bot flag is another issue entirely). --Whiteknight (Page) (Talk) 23:30, 17 June 2007 (UTC)
 * I am not necessarily saying that we should be "anti-Wikipedia", but I am questioning the necessity of having a bot here and asking how this is going be used. I support categorization of uncategorized books, but I have strongly questioned the need to place each and every page of every Wikibook into some sort of category.  And I don't necessarily buy the argument that categorization is necessarily going to help find problem modules either.  With the added issues that we are talking about having this whole process automated, that means in theory a page will be categorized without anybody but the original spammer or clueless new contributor ever even seeing the page.  I am just suggesting that for some tasks people just do it better, and it would be useful to know ahead of time just how this is going to be used.  'Bots can be useful, but like all things related to computers, they also help to make even larger messes when things (or users) go bad.  This is one reason why discussion of what exactly the bot is going to do is relevant.  I'm also suggesting that the current category system has some real flaws and no significant community discussion has occurred to talk about what appears to be a major update of the current categorization system.  I think it would be useful to have that discussion first before a major overhaul involving a bot happens.  --Rob Horning 05:04, 19 June 2007 (UTC)
 * I can think of a couple of good reasons fairly quickly for categorising all pages and not just the title pages of books:
 * Navigation breaks; categories help in finding pages not otherwise linked properly through good navigation
 * Special:Recentchangeslinked/Category:... can be used to monitor (only) changes to all pages of a book when they are all added to a single category
 * Just my AU$0.02, not having a viewpoint on the rest of this discussion. Webaware talk 05:23, 19 June 2007 (UTC)
 * This still gets back to the question of how these pages are going to be categorized. Lumping them into a generic category for the book doesn't really provide any additional information, or help you to find a collection of modules about sorting algorithms as implemented in a dozen different Wikibooks about different computer languages.  The current methodology of categorization doesn't deal with that issue.  And as for the way to monitor modules instead of simply something like Special:Prefixindex that can also list all of the modules of a Wikibook without even using categories, that is simply a matter of adding a plug-in or extension to MediaWiki to do this sort of category search.  This is describing a hack of a hack and not necessarily a hard pressing need for requiring categorization.  Still, it is an application that can use categories.  I'm still asking what the guidelines are going to be on how this categorization is going to be implemented, and where the discussion about this has been.  It shouldn't really be here on this page either, so I'm just saying that this bot request may be premature, not necessarily inapprorpiate.  --Rob Horning 07:52, 19 June 2007 (UTC)
 * Note: Provisional bot status revoked. -- SB_Johnny | talk 23:08, 17 June 2007 (UTC)
 * - I can't see any problem. Catergorization doesn't attract any new visitors in my opinion but many people see it as important.  I also don't see any problem with SBJohnny's actions - if any bot goes crazy then we can simply block them.  Wikibooks is well patrolled by admins and an uncontrolable bot can easily be stopped. Xania [[Image:Flag_of_Italy.svg|15px]]talk 16:11, 18 June 2007 (UTC)


 * ❌ - User failed to regain bot status. -within focus 20:03, 13 September 2007 (UTC)