→Should a 4th CSD for unused templates be added?: closing - opposed |
Winged Blades of Godric (talk | contribs) Tag: 2017 wikitext editor |
||
Line 153: | Line 153: | ||
== Proposal: Expand G5 to include undisclosed paid editing == |
== Proposal: Expand G5 to include undisclosed paid editing == |
||
{{atop|Closing.....[[User:Winged Blades of Godric|<span style="color: red">∯</span><span style="font-family:Verdana"><b style="color:#070">WBG</b></span>]][[User talk:Winged Blades of Godric|<sup><span style="color:#00F">converse</span></sup>]] 19:03, 14 April 2019 (UTC)}} |
|||
===Main proposal=== |
===Main proposal=== |
||
Currently [[WP:G5|G5]] is actionable for articles created by editors violating a block or ban. This is a ''personal'' block or ban: "To qualify, the edit must be a violation of the user's specific block or ban." |
Currently [[WP:G5|G5]] is actionable for articles created by editors violating a block or ban. This is a ''personal'' block or ban: "To qualify, the edit must be a violation of the user's specific block or ban." |
||
Line 346: | Line 347: | ||
*{{replyto|Od Mishehu}} Please could you be specific about the wording you are proposing as there are several different formulations above "this" could refer to (and the exact wording matters at speedy deletion). However, unless your proposal addresses the reasons for opposition that are unrelated to it being part of G5 (which I think is most of them) it is very unlikely to gain consensus. [[User:Thryduulf|Thryduulf]] ([[User talk:Thryduulf|talk]]) 19:11, 24 March 2019 (UTC) |
*{{replyto|Od Mishehu}} Please could you be specific about the wording you are proposing as there are several different formulations above "this" could refer to (and the exact wording matters at speedy deletion). However, unless your proposal addresses the reasons for opposition that are unrelated to it being part of G5 (which I think is most of them) it is very unlikely to gain consensus. [[User:Thryduulf|Thryduulf]] ([[User talk:Thryduulf|talk]]) 19:11, 24 March 2019 (UTC) |
||
*'''Oppose''' per [[WP:CREEP]]. There are too many half-baked proposals now and this is vexatious. No deal is better than a bad deal, eh? [[user:Andrew Davidson|Andrew D.]] ([[user talk:Andrew Davidson|talk]]) 12:03, 25 March 2019 (UTC) |
*'''Oppose''' per [[WP:CREEP]]. There are too many half-baked proposals now and this is vexatious. No deal is better than a bad deal, eh? [[user:Andrew Davidson|Andrew D.]] ([[user talk:Andrew Davidson|talk]]) 12:03, 25 March 2019 (UTC) |
||
{{ab}} |
|||
== Update P2 proposed == |
== Update P2 proposed == |
Revision as of 19:03, 14 April 2019
Should a 4th CSD for unused templates be added?
- The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.
Should a new CSD criteria (T4T5) be added for unused templates that meet the following criteria:
- Template is not used anywhere, I.E. has zero transclusions excluding templates own documentation of course.
- Template is NOT a substitute only template. Should go without saying but templates that are substitute only by definition should never have transclusions, doesn't mean the template can been speedily deleted.
- Template is older than 6 months. No speedily deleting a new template that hasn't been used just yet.
- Template is NOT a sometimes unused/temporary template. An example of this would be {{help me}} which may have 0 transclusions at any given moment.
Please discuss. --Zackmann (Talk to me/What I been doing) 18:41, 21 February 2019 (UTC)
- Support Practice shows that such templates are routinely deleted at TfD without much discussion, CSDing them would save effort. {{3x|p}}ery (talk) 18:46, 21 February 2019 (UTC)
- For the same reason it was unanimously rejected barely a month ago. Please read the "Please read this before proposing new criteria" box at the top of this page and do the due diligence of at least a minimal search before squandering the community's time with a formal rfc. —Cryptic 19:41, 21 February 2019 (UTC)
- @Cryptic: that was not a formal RFC, this is. Additionally, I added some clarifying criteria in this proposal, such as the note about substitute only templates being exempt. Would be nice to have the proposal discussed on its merits rather than based on a cursory previous discussion. --Zackmann (Talk to me/What I been doing) 19:55, 21 February 2019 (UTC)
- Support The reason this gets suggested repeatedly is because it's a good idea. So much time is wasted taking unused templates to TfD when there is virtually no opposition to their deletion. Number 57 20:39, 21 February 2019 (UTC)
- Support - just look at the logs for the past few months, most (or even all?) of the unused templates were deleted without any objection. Regular nomination just clogs the list with pointless discussions full of "per nom" as there is virtually nothing to say. As these templates weren't in use, there is nothing to lose by deleting them. If someone later on wants the template back, they can ask an admin to WP:UNDELETE it and move it to their sandbox. --Gonnym (talk) 21:04, 21 February 2019 (UTC)
- Support – templates aren't articles, so the CSD criteria for removing unused ones should be low. --IJBall (contribs • talk) 21:38, 21 February 2019 (UTC)
- Support – it's too difficult, because too bureaucratic, to have unused templates deleted, so many editors, including me, have given up. Category:Unnecessary taxonomy templates, for example, is full of unused templates, almost all blanked. The suggested criterion would help to get them deleted instead of leaving them just sitting there. Peter coxhead (talk) 21:50, 21 February 2019 (UTC)
- Support much like we delete unused pages with just the article wizard text. Clear the clutter and focus management on the useful. Legacypac (talk) 22:03, 21 February 2019 (UTC)
- Comment should this CSD apply to unused Lua modules too? {{3x|p}}ery (talk) 22:29, 21 February 2019 (UTC)
- Oppose
It cannot be T4, because T4 has been used before and we do not re-use old codes.unless it can be demonstrated that each template to be deleted under the proposed criterion has never been used, either directly or as a substitution. --Redrose64 🌹 (talk) 23:31, 21 February 2019 (UTC)- @Redrose64: so you are opposing the entire proposal because it cannot be T4? That is a pretty simple correction... If it was T5 would you then support it? --Zackmann (Talk to me/What I been doing) 01:34, 22 February 2019 (UTC)
- Support Assuming this accounts for transclusions that have been removed as part of vandalism before the template is deleted, is not just used to bypass TfD in some way, and say around 7 days has passed after being nominated before deletion. Breawycker (talk to me!) 00:15, 22 February 2019 (UTC)
- @Breawycker: great point. You can't just remove the transclusions and then CSD the template to game the system. If a template has a number of transclusions and you think it should be removed, that is a case for XfD. --Zackmann (Talk to me/What I been doing) 01:35, 22 February 2019 (UTC)
- Such activity is a possible behavioural issue rather than a reason to not have a CSD for non-controversal cases. Anyway if someone changes a handful of templates A to template B and the gets rid of template B where is the problem? Legacypac (talk) 15:23, 22 February 2019 (UTC)
- Oppose. The fact that a template is unused is never by itself a reason to delete it, it's only relevant as a criterion by which to select a potentially unneeded template for further inspection at TfD (see also this brief essay). An evaluation of use, or the potential for use, of a template requires knowledge of its context, the function that it serves and the existence of related or similar templates. All this calls for judgement that is above and beyond what goes into dealing with the obvious, clear-cut scenarios that the CSD criteria are there for. A template can be "unused" for a wide variety of reasons. Maybe it's a useful template that nobody happens to know about, in which case it needs not to be deleted, but popularised and integrated into the project documentation. Maybe it's a template that is meant to be used only temporarily, for example until certain issues on a given page have been addressed (niche maintenance templates). It may be a currently unused element of a wider system that somebody might soon need (happens sometimes within the lang-xx family of templates). It may be unused because it was removed in error from the one page where it's meant to be used and nobody has noticed yet. It may be unused because the editor who tagged it for deletion has just removed all of its transclusions. It may be unused at the moment, but its existence may be assumed or required by some other piece of machinery (like a module) in a way that doesn't show up in its list of transclusions. A template may be unused, but it could hold the history of a fragment of article text that has at some point been merged into the article, and hence the template is there to preserve attribution. A template may appear as unused because it's meant to be substed; yes, such a template should be exempt from the proposed CSD, but how do you determine if a template's meant to be substed? (It doesn't always say so in the documentation; I remember there have been TfD discussions where several editors had voted to delete such an "unused" template until someone noticed it was a user warning template and so is always substed.)
I don't think any one editor is attuned to all these possibilities, and that's why such things are better decided by discussion involving several participants. However, TfD does indeed occasionally feel like it's getting flooded with similar nominations, so something probably ought to be done about that. If new speedy deletion criteria are going to be part of the solution, then they should be about easy, clearly-defined subsets of templates; there could, for example, be a CSD criterion for unused navboxes that fail WP:NAVBOX. – Uanfala (talk) 03:08, 22 February 2019 (UTC) - Support. This helps make editing an easier experience by simplifying the set of available templates. The amount of verbiage and time wasted on discussing (but rarely if ever actually using) potential uses of these templates is vast. A template exists to serve the encyclopedia in some way (helping editing or reading) and if it's not used, in the caveats above, it should be deleted.--Tom (LT) (talk) 07:26, 22 February 2019 (UTC)
- Oppose, unused templates can hold interesting history that needs preserving. Also, deleting templates that have been widely used destroys old revisions of articles. And per Uanfala. —Kusma (t·c) 07:45, 22 February 2019 (UTC)
- "this destroys old revisions" is not an argument that TfD seems to be accepting: Template:Persondata was deleted. {{3x|p}}ery (talk) 18:13, 22 February 2019 (UTC)
- Yeah, that argument is a flawed argument from its root, as the whole deletion process of TfD destroys old revisions all the time. --Gonnym (talk) 20:45, 22 February 2019 (UTC)
- It is sometimes relevant. For Persondata, old revisions just have a redlinked template at the bottom. That is not a problem. Deleting templates used within the text (convert-like ones) is a much more serious problem. —Kusma (t·c) 17:33, 6 March 2019 (UTC)
- Yeah, that argument is a flawed argument from its root, as the whole deletion process of TfD destroys old revisions all the time. --Gonnym (talk) 20:45, 22 February 2019 (UTC)
- "this destroys old revisions" is not an argument that TfD seems to be accepting: Template:Persondata was deleted. {{3x|p}}ery (talk) 18:13, 22 February 2019 (UTC)
- Oppose per Uanfala and Kusma. I'm generally not opposed to cleanup but the above-mentioned risks are clearly higher than the potential benefits. Regards SoWhy 08:15, 22 February 2019 (UTC)
- Oppose I've supported and even proposed the ability to PROD unused templates, since it takes a step out of the deletion process but allows users seven days to oppose and potentially discuss its deletion and use. Speedy deletion offers none of that. Nihlus 09:06, 22 February 2019 (UTC)
- Oppose Some unused templates, such as {{Roads legend}} and {{Trillium Line route diagram detailed}}, are nevertheless permanently stored in template space, and they may be linked to Wikidata items which are structurally useful. I would support if there were a provision for certain templates – such as templates explicitly marked as historical, templates with incoming links from articles, template sandboxes, templates for which T5 has previously been declined, and templates which have been otherwise marked as ineligible for T5 – to be ineligible for T5 deletion. Jc86035 (talk) 10:07, 22 February 2019 (UTC)
- Oppose. While I think the basic idea is good, there are too many nuances and exceptions to the point where I believe a CSD to be untenable. I think the proper solution is to expand PROD to templates so the distinction becomes potentially controversial vs. uncontroversial, and I would support such a proposal. -- Tavix (talk) 15:00, 22 February 2019 (UTC)
- I was thinking I'd like to see template prod as well, but that's a discussion for a different page. --Izno (talk) 15:22, 22 February 2019 (UTC)
- No, that was not a template PROD but another CSD proposal disguised as a "PROD". A template PROD would be simply expanding WP:PROD to include (unused) templates. If you read that discussion, I had opposed that proposal for that very reason. -- Tavix (talk) 16:21, 22 February 2019 (UTC)
- Either way, this isn't the page for it, was the point I was making. --Izno (talk) 16:47, 22 February 2019 (UTC)
- No, that was not a template PROD but another CSD proposal disguised as a "PROD". A template PROD would be simply expanding WP:PROD to include (unused) templates. If you read that discussion, I had opposed that proposal for that very reason. -- Tavix (talk) 16:21, 22 February 2019 (UTC)
- I was thinking I'd like to see template prod as well, but that's a discussion for a different page. --Izno (talk) 15:22, 22 February 2019 (UTC)
- Oppose per the above opposers and the many previous discussions where a CSD criterion for unused templates has been rejected. Not all templates need to be used at all times (e.g. {{help me}} may be unused at any given moment), not all subst-only templates are marked as such, some templates should be kept so as not to break old revisions, etc, etc. Thryduulf (talk) 15:27, 22 February 2019 (UTC)
- C1 exists despite the fact that Category:Wikipedians looking for help may be empty at any given moment. Why can't T5 be implemented in the same way? {{3x|p}}ery (talk) 18:20, 22 February 2019 (UTC)
- Oppose per those above who have already highlighted many situations in which a template with zero transclusions should be kept anyway, i.e. there are good reasons why zero transclusion templates should not be deleted without prior discussion. Ivanvector (Talk/Edits) 16:25, 22 February 2019 (UTC)
- Comment @Thryduulf: and a few others have made a very valid point that there are some templates that at any given moment may have no transclusions ({{help me}} for example). Pppery raised a good counter argument about C1 accounting for sometimes empty categories. This is why we have {{Empty category}}. With that in mind, I 100% agree that something would be needed to document that some templates may have zero transclusions at any given moment. I'm curious those who have objected based on this point alone, if we were able to account for this case would you be more supportive of this? I have added a new criteria to the top of this RFC to account for that case. It would seem to me that it would be pretty easy to add some documentation to the new CSD criteria that exempts templates that may at any given moment have no transclusions.
- Additionally, I'm curious if there are additional conditions that would cause some of you to be more supportive of the idea? For example, if we said that the template must be at least 1 year old instead of just 6 months? Thryduulf thank you for raising that point, it wasn't something I had considered and definitely needs to be accounted for. A reminder, the goal of this CSD criteria is to expedite the process of deleting old unused templates that have been sitting around for a long time and are unused. It is not my intention to facilitate a method for gaming the system and quickly nuking templates someone just doesn't like. --Zackmann (Talk to me/What I been doing) 20:08, 22 February 2019 (UTC)
- @Redrose64, SoWhy, Jc86035, Thryduulf, and Ivanvector: please see above comment. --Zackmann (Talk to me/What I been doing) 20:08, 22 February 2019 (UTC)
- My opposition to this criteria is based on far more than just on that one point, and still stands. For example the older a template is the higher the chance of breaking old revisions. If a template has been around for a year without causing problems then I'm not seeing any reason why deletion of it needs expediting. Thryduulf (talk) 20:18, 22 February 2019 (UTC)
- Regarding your question about an {{Empty category}} equivalent for templates, {{Subst only}} will account for most of the templates that have no transclusions, with the caveat that some templates use {{Substitution}}, which allows a custom message, and thus requires examination to determine whether the jist is that it's a template that must be substituted. --Bsherr (talk) 20:28, 22 February 2019 (UTC)
- Still a no for me, the temporarily-unused template (like {{help me}}) is just one of the issues raised. I'm actually more concerned about borked page histories that rely on templates that are later deprecated. I'm not against deleting unused templates, I'm only opposed to doing it without having a discussion to consider all the angles first. Ivanvector (Talk/Edits) 20:36, 22 February 2019 (UTC)
- Oppose. Again? We just discussed this in December. See Wikipedia talk:Criteria for speedy deletion/Archive 72#Proposal to add T4: Unused. I'll just quote my comment from then:
Alas, this has been proposed, many, many, many, many times. The iterations sometimes vary, such as just for user namespace templates, or just for those "not encyclopedic", with varying waiting periods or age requirements. Just some of these discussions are: /Archive 72#Proposed tweak to T3, /Archive 10#Orphaned templates, /Archive 67#New criterion - T4, aka Template PROD, /Archive 44#New CSD - T4 Unused userbox that is more than 30 days old, /Archive 42#T4: Unused template, /Archive_52#Deprecated_templates, /Archive 22#Speedy deletion of unused templates?, /Archive 59#Gauging opinion on a possible new criterion for templates. There have also been several proposals at WT:PROD for this (reportedly four in 2007 alone). There seem to be three reasons this has never been adopted. Firstly, as pointed out by Tazerdadog, templates that are intended to be substituted have no transclusions by design, and a summary process like CSD is not efficient to distinguish those not transcluded by design from those by circumstance. Secondly, CSD is generally for urgent deletions and, although TfD is busy, it is not backlogged enough (with deserved thanks to Galobtter (above), Primefac and others) to warrant the use of a summary process like this, particularly since an unused template is not an urgent cause for deletion. Thirdly, that a template is unused is not, in any guideline, a dispositive reason to delete it; rather, it is just a relevant consideration; therefore, it would be egregious to make it a dispositive reason to speedily delete it before there is consensus on whether this should be decisive for a deletion discussion.
- I want to echo what Tavix said, because he hit it on the head. CSD should be quick and straightforward — the least amount of judgment or gray area required, the better. This proposal would require users to:
- Ensure the template isn't substituted
- Ensure it's older than 180 days
- Check that it has no transclusions
- Check if any redirects have transclusions or history that might have been merged there
- Somehow know whether this template may have been used occasionally but not right now even though it's not substituted(???)
- Know whether any of its redirects may have also been used occasionally but not right now
- That's nowhere near tenable for a SD criterion. A TPROD process might work, but this is too complicated for speedy deletion. ~ Amory (u • t • c) 21:39, 22 February 2019 (UTC)
- As usual, I'm with Amory. Too many criteria for a CSD, any questionable template should be sent to Tfd. Liz Read! Talk! 04:22, 23 February 2019 (UTC)
- Comment/Question: How many templates fit the proposed criteria today? Are we talking about 10, 100, 1000, 10,000, or more? An order of magnitude (backed up by a reasonable method of arriving at that number) would be helpful in this discussion. – Jonesey95 (talk) 22:23, 24 February 2019 (UTC)
- As a first approximation, there's 90684 non-redirect pages in the template namespace that currently aren't transcluded from any other page that were created before 2018-06-27 (quarry:query/33701). So something less than that - the count includes subst-only templates, and template sandboxes, and template documentation pages, and so on. —Cryptic 23:16, 24 February 2019 (UTC)
- Excluding pages with names ending in "/sandbox", "/testcases", or "/doc" brings the total down to 83752. Further excluding templates that themselves transclude {{require subst}} and/or {{subst only}} brings it down to 82204. Even supposing that many subst-only templates aren't documented as subst-only, there's only 518284 total non-redirect pages in the Template namespace. —Cryptic 19:15, 25 February 2019 (UTC)
- Interesting. I wonder why Wikipedia:Database reports/Unused templates shows only about 13,446 templates before it starts to list redirects (and why would it list redirects, which are cheap?). It seems that a better set of queries is needed, perhaps one or two that implement some of the criteria listed at the top of this section. That might allow people who want to take templates to TFD (or label them as subst-only) to have an easier time of it, allowing all of us (or most of us, at least) to achieve our goals. – Jonesey95 (talk) 19:59, 25 February 2019 (UTC)
- I find the deletion of templates very annoying when reviewing old page revisions where they were used. —SmokeyJoe (talk) 22:27, 24 February 2019 (UTC)
- I'm pretty sure the criteria under discussion would prevent that from happening. CSD would apply only to templates that are unused. – Jonesey95 (talk) 22:47, 24 February 2019 (UTC)
- Nope. This is for templates that are currently unused, not for ones that were never used. —Cryptic 23:09, 24 February 2019 (UTC)
- I'm pretty sure the criteria under discussion would prevent that from happening. CSD would apply only to templates that are unused. – Jonesey95 (talk) 22:47, 24 February 2019 (UTC)
- Oppose I'll give a current real example to illustrate why I disagree with this proposal. At Template talk:URL #Infobox input may vary, same output preferred, there was a request for code that "
accepts all input forms, then reformats it as needed into good {{URL}}
". Eventually we've arrived at updated functionality in Module:URL and a new template called Template:URL2 that's much more user-friendly in infoboxes than Template:URL. Take a look at the comparison between the outputs and judge for yourselves whether {{URL2}} has potential, especially as it doesn't throw an error in an infobox that uses Wikidata (which may provide blank input to the template). But {{URL2}} is not used in article space at present, so unless somebody uses it in the next x days, it would be deleted under this criterion. What value does that add to the encyclopedia? What happened to WP:TIND? Why would I spend time creating potentially useful code if I knew there was a deadline imposed for somebody to use it? If potentially useful code keeps being deleted, what will you do when editors asks for new functionality but all the coders are too fed up with having their work deleted to bother with it? --RexxS (talk) 23:50, 24 February 2019 (UTC) Oppose - no matter how many ways these proposals are dressed, I will stubbornly oppose; unused will never be a synonym for unusable (a criterion I would support) and it will never mean its condition is final, and so absolute that every potential for use in the future is also precluded (only its deletion can accomplish that). And, so too is it fact that a template's deletion, after discussion at TfD, does not remotely suggest that discussion itself is not beneficial, or even necessary.
In closing, I'd like to say: I find the proximal nearness of this discussion to its most recent counterpart more than a little disturbing. I hope when it closes, its proponents will accept the consensus achieved (or lack thereof) respect the mandate in its remit, and stifle the inclination to be heard again. These matters are settled, their questions have been thoroughly answered, and enough has been vested already. Moving on is the only course to follow from here, please follow that course!--John Cline (talk) 04:08, 25 February 2019 (UTC)
- Oppose. Mere disuse is a bad reason to delete a template, if for no other reason than per SmokeyJoe and Cryptic. If they're not problematic, keep them, since otherwise you're damaging old revisions for no good reason. Also, per Cryptic's stats, this would involve a very large number of pages, even after you skip the ones that are always supposed to be substituted, the new creations, and the temporary ones. We shouldn't declare such a large number of pages currently speedy-deletable, except after a big community discussion on whether such deletions are appropriate. Look at the way G13 was originally handled; it was much bigger than merely a conversation here. Unless they come out of a big discussion, the only way we should create new speedy criteria is if they cover a class of pages that is rare at the moment because the pages are constantly getting deleted at XFD already. Nyttend (talk) 04:27, 25 February 2019 (UTC)
- Comment (additional to oppose above). Zackmann08 has recently TFD'd hundreds of unused templates. The results seem to indicate to me that, while many are deleted unopposed, clearly not all are deleted, and the deletions are clearly not uncontroversial. So CSD for them is clearly wrong. —Kusma (t·c) 17:33, 6 March 2019 (UTC)
- Oppose for similar reasons as in this recent MfD of unused templates: Wikipedia:Miscellany for deletion/Unused userbox template. — Rhododendrites talk \\ 22:16, 13 March 2019 (UTC)
- Oppose: There are enough corner cases that this could be dangerous. What about templates that are uses as preloads only (which aren't subst only, yet have zero transclusions)? What about templates that aren't technically subst only, but which don't have any hidden comment for tracking and have so far only been substed? How would criteria 4 be objectively judged, as any template could be defined as "sometimes unused"? As we've seen at TfD recently, figuring out whether or not a template is actually used is often not srtaightforward enough to be considered as a CSD. --Ahecht (TALK
PAGE) 14:05, 14 March 2019 (UTC) - Oppose: The old discussion already showed problems, it shouldn't have been simply reproposed after that short a time without addressing the issues raised before. ChristianKl ❪✉❫ 14:38, 14 March 2019 (UTC)
- Oppose especially per Kusma and others. Deleting templates makes older revisions extremely difficult to read. Guettarda (talk) 15:02, 14 March 2019 (UTC)
- 'Comment note that Wikipedia:Miscellany for deletion/Unused userbox template (and MfD for 250 unusued userbox templates) was closed as "keep", Template:Pollachi–Dindigul branch line, Template:Railway stations in the Borough of Scarborough and Template:Railway stations in Nottinghamshire were both brought to TfD as unused but rather than being deleted the templates were used. Wikipedia:Templates for discussion/Log/2019 March 2#Template:Trillium Line route diagram detailed is still open but is likely heading for a keep or no consensus close after being relisted - debate is about whether a template that is linked from article space is "unused". These show that "unused templates" fail the "uncontestable" and "objective" requirements for new criteria. Thryduulf (talk) 10:50, 15 March 2019 (UTC)
- Oppose. Uanfala covered what I was going to. I also agree this is basically forum-shopping since we just went over this in December. — SMcCandlish ☏ ¢ 😼 21:53, 20 March 2019 (UTC)
- Please do not archive this until it has been formally closed (I've requested this at WP:ANRFC) so that there is a clear record of consensus about this going forward. Thryduulf (talk) 19:20, 23 March 2019 (UTC)
- A dated comment to delay automated archiving while we await formal closure. Thryduulf (talk) 22:20, 10 April 2019 (UTC)
- Oppose The criteria are too complex to be determined in a speedy manner. Many taggers will make mistakes, and for deleters it takes too long to check, and may not even be possible to tell, so they are sure to be deleted in error if this gets up. Graeme Bartlett (talk) 22:45, 10 April 2019 (UTC)
Extend R2 to portals
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Two days ago WP:R2 was boldly extended to apply to redirects from the portal namespace. There appears to be some disagreement at least on what exceptions there should be. Could we decide on all that here first? Pinging involved editors: Legacypac, Thryduulf, Tavix. – Uanfala (talk) 19:01, 3 March 2019 (UTC)
- Mainspace and Portal space are both reader facing content spaces. Draft space is for stuff that is not intended for readers. Links between Mainspace and Portal space and vice versa are fine but if a portal is draftified it is exactly like draftifying an article so the redirect should be immediately deleted. When there were 1700 mostly dead portals this was not frequent problem but now we have 4500 new automated portals that are being examined and I expect a bunch will be placed in some Draft holding pen out of view while consideration of their future is done. Legacypac (talk) 19:09, 3 March 2019 (UTC)
- Legacypac, it is not clear that this is a very common occurrence that would require it to be covered by CSD. Also, draft portals can just be left in portal space. If they are not linked to from any articles or other portals, there is no fundamental problem with keeping unfinished portals around in portal space. The mass-produced automatic portals should be deleted, not draftified. Classic portals with many subpages simply can't be moved around in any meaningful way, so instead of being draftified, they should just be tagged with some template that tells any accidental readers that it is unfinished and that they should go read something else. —Kusma (t·c) 19:18, 3 March 2019 (UTC)
- Wait until Wikipedia:Administrators' noticeboard#Thousands of Portals is resolved. I don't see the point in draftifying portals if they are going to be deleted anyway. -- Tavix (talk) 19:13, 3 March 2019 (UTC)
- (edit conflict) Main → Portal and Portal → Main redirects should not be speedily deleted under R2 as they are both reader-facing namespaces with encyclopaedic content. Such redirects will not always be optimal but that is a matter that should be discussed at RfD as deletion is not going to be the best action for all of them. This means that, if R2 applies to portals at all (which I have no strong opinions about) then the main namespace needs to be listed as an exception. Thryduulf (talk) 19:14, 3 March 2019 (UTC)
- agreed 100% with Thryduulf's refinement. I disagree with waiting because the nuclear option is not going to delete all portals, only many portals. There are a bunch of legacy portals that may well be draftified and dealt with seperately. I actually tagged a portal=>Draft redirect R2 but it did not display properly, then I tagged it housekeeping with a not it was R2 and that was accepted. I don't see this change as an expansion, more a refinement of wording based on the principle of the CSD. Legacypac (talk) 19:19, 3 March 2019 (UTC)
- Then how about a separate bullet point for R2 that includes any other namespace to draft, so we don't get a bunch of potentially-confusing exceptions and includes anything that has been draftified (eg: templates, books). -- Tavix (talk) 19:24, 3 March 2019 (UTC)
- Good idea. Add the words "Also any redirect to Draft namespace, except from user namespace." This way anything draftified from any random spot (like I saw someone post a draft as a category recently) can be moved and the redirect nuked. Legacypac (talk) 19:32, 3 March 2019 (UTC)
- Legacypac, redirects created because a page was obviously created in the wrong namespace are already covered under G6. —Kusma (t·c) 19:34, 3 March 2019 (UTC)
- Tavix, the question is why we would want to move pages from non-mainspace to draft anyway. I am unconvinced that this is a good idea, as many namespaces have their own special features. Draft books should be in Book space, just as draft TimedTexts should be in TimedText namespace. —Kusma (t·c) 19:33, 3 March 2019 (UTC)
- That makes sense to me. -- Tavix (talk) 19:42, 3 March 2019 (UTC)
- Good idea. Add the words "Also any redirect to Draft namespace, except from user namespace." This way anything draftified from any random spot (like I saw someone post a draft as a category recently) can be moved and the redirect nuked. Legacypac (talk) 19:32, 3 March 2019 (UTC)
- Then how about a separate bullet point for R2 that includes any other namespace to draft, so we don't get a bunch of potentially-confusing exceptions and includes anything that has been draftified (eg: templates, books). -- Tavix (talk) 19:24, 3 March 2019 (UTC)
- agreed 100% with Thryduulf's refinement. I disagree with waiting because the nuclear option is not going to delete all portals, only many portals. There are a bunch of legacy portals that may well be draftified and dealt with seperately. I actually tagged a portal=>Draft redirect R2 but it did not display properly, then I tagged it housekeeping with a not it was R2 and that was accepted. I don't see this change as an expansion, more a refinement of wording based on the principle of the CSD. Legacypac (talk) 19:19, 3 March 2019 (UTC)
The "special features" argument just convinced me no Portal should be in Draftspace. It breaks them anyway. Can you make that point at AN against the idea of sending Portals to Draftfor more work? Legacypac (talk) 19:58, 3 March 2019 (UTC)
- Legacypac, done. One problem with the current portals discussion is that it is so fragmented... but the AN discussion should fix the main issue soon. —Kusma (t·c) 20:20, 3 March 2019 (UTC)
- Point of order - I'm not following the discussion this was forked from so I don't really get what's happening, but the redirect criteria apply to redirects in any namespace, including portal redirects. If the proposal is to apply the R criteria to portals themselves, then oppose, R criteria are for redirects. Ivanvector (Talk/Edits) 16:31, 21 March 2019 (UTC)
Provide for CSD criterion X3: Mass-created portals
G14
As far as I was aware DAB pages that have a primary topic and there is only 1 other (WP:2DABS) can be deleted as unnecessary DAB pages. This was quite clear in the past but it looks like since G6 was split, the inclusion of situation where there are only 2 topics and there is a primary topic has been lost for some reason. See User talk:Patar knight#Ross Greer (disambiguation) and User talk:Sir Sputnik#Magnus Lindberg (disambiguation). I would note though that DAB entries that are red links and part title matches do still count as "entries" for this purpose. Crouch, Swale (talk) 18:16, 10 March 2019 (UTC)
- Looking at the plain wording of G14 as it is now, if there is a primary topic and a non-primary topic on a 2DAB, then it is still disambiguating two extant articles and ineligible for G14. My experience has been that they are then typically PRODed, so they show up at Wikipedia:WikiProject_Disambiguation#Article_alerts for a week and gives room for editors who work with DABs to review them. This interpretation fits with the framework of CSD as getting rid of unambiguous cases and letting other deletion processes deal with less clear cases.
- For DABs, those "disambiguating" one or zero DAB entries absolutely fail as DAB pages and are arguably actively unhelpful in navigation, while those with two entries do not. Those with two entries are more easily converted into valid DABs with the addition of only one additional entry or might be converted into a 2DAB page with no primary topic if the article at the base name doesn't have a solid claim to be the primary topic. Having a 2DAB page is at worst neutral, and an additional week to potentially save it isn't a big deal.
- Looking through the history of G14/G6 I don't think that it ever explicitly allowed deletion of 2DABs with a primary topic:
- August 2009: Added to G6 as "deleting unnecessary disambiguation pages
- January 2013: G6 is broken out into bullet points
- January 2013: "unnecessary is clarified as "those listing only one or zero links to existing Wikipedia articles."
- March 2017: "links" to zero/one extant article is changed to "disambiguates" zero/one extant articles.
- December 2018: G14 is split off from G6 with minimal changes.
- My interpretation of Tavix's change in March 2017 is that linking the previous wording technically allowed DAB pages with zero valid dab entries but some links to existing articles, either in invalid DAB entries or a "See also" section, to escape speedy deletion. The new wording shows that the linked articles must be part of a valid dab entry, not just any link whatsoever. ---- Patar knight - chat/contributions 01:21, 11 March 2019 (UTC)
- I would support a speedy deletion option for disambiguation topics where there is one clear primary topic, only one other topic, and for which the primary topic page already contains a hatnote to the other topic, with no link to the disambiguation page. In that case, any reader looking up the term is already going to be taken to a page with an existing hatnote leading to the other topic, so there is no point in the disambiguation page existing. bd2412 T 17:46, 12 March 2019 (UTC)
- I fully agree with Patar knight. Dab pages with a primary topic and only one other link are pretty useless most of the time, but the emphasis here is on "most of the time". Such pages are normally dealt with using PROD, and often enough it would happen that somebody might come along and expand the page with additional entries. Or it might turn out that the page is a result of a bad move. Or it could disambiguate between two articles only because of a previous overzealous attempt at cleanup that had removed valid links. Etc, etc. There are too many possible scenarios and too many nuances for CSD to be appropriate, and there are too few pages of this kind getting deleted for there to be a need for an extension of the existing criteria. – Uanfala (talk) 00:07, 13 March 2019 (UTC)
- I agree with Uanfala that extending CSD to cover primary+1 disambiguation pages is not appropriate. In addition to the scenarios they list, in some cases there will be people navigating directly to the disambiguation page where they know or suspect the topic they are looking for is not primary but do not know what its title is. Whether this is likely will depend on factors that cannot be judged by a single admin reviewing CSD nominations. Thryduulf (talk) 09:22, 13 March 2019 (UTC)
Clarification on G8
Proposal: Expand G5 to include undisclosed paid editing
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Main proposal
Currently G5 is actionable for articles created by editors violating a block or ban. This is a personal block or ban: "To qualify, the edit must be a violation of the user's specific block or ban."
Well, an undisclosed paid editor is "banned" from editing, not by Wikipedia policy but by the WMF's Terms of Use. This ban isn't directed at any specific editor, but editing without disclosing payment is blockable.
A discussion on the OTRS mailing list suggests that it would make sense, as an additional deterrent, to treat articles created by such editors as any other G5 article, but the wording of G5 would need to change.
We have the Terms of Use, and we have G5, and the purpose of G5 seems like a good fit for enforcing the Terms of Use ban on undisclosed paid editing.
I suggest adding after the bullet list in G5:
- "In addition, because undisclosed paid editing is prohibited by Wikimedia Foundation policy, articles created by undisclosed paid editors are candidates for speedy deletion under G5."
Or an alternative suggested by Cryptic below:
- "In addition, because undisclosed paid editing is prohibited by Wikimedia Foundation policy, articles created by editors who have been indef-blocked for undisclosed paid editing are candidates for speedy deletion under G5."
What say everyone? ~Anachronist (talk) 22:59, 13 March 2019 (UTC)
- @Anachronist: I like the idea... My question becomes how do you determine that the user IS an undisclosed paid editor? I've personally accused someone of being a paid editor to later find out it was a high school kid who was just really excited about the product. Page certainly needed to be reworked, but didn't really qualify for CSD. I would argue this sort of article really needs to go through WP:AFD so that the paid editor status can be proven/flushed out... Now, if on the other hand, the editor in question is blocked as a result of paid editing, that is another story. Just some food for thought. I like the idea. --Zackmann (Talk to me/What I been doing) 23:30, 13 March 2019 (UTC)
Support.
- Can't support deletion because the criterion is not object, per User:Thryduulf. (we've been here before). However, something' has to be done. Counter proposal is to Quarantine suspected UPE product. --SmokeyJoe (talk)
- Support toughening the current rules to better than toothless. Consistent, or even prerequisite of this, is that at WP:COI most of the occurrences of the toothless "should" are changed to "must".
- COI editors MUST NOT edit articles directly; instead they may make requests and suggestions on the talk page.
- COI editors MUST NOT create articles; instead they may use WP:AfC.
- UPE editors are a worse-problem subset of COI editors, and the boundary is indistinct. Where a page is the sole product, ignoring minor edits, of a UPE editor or editors, an admin may delete it per WP:CSD#G5(UPE).
- Post deletion of UPE product, if the editor later sufficiently declares and complies, the deleted page should not be WP:REFUNDed, instead, the COI editor may start again, ensuring that all COI editing has links back to a declaration older than the edits. To comply conservatively with attribution requirements, if they request an emailed version, email only the references (there is no creative content in a reference list).
- I would like to go further, and require paid editors to use a special alternative account, named with the suffix "(paid)". Eg. User:Example (paid). This account must be a fully declared alternative account, linking to & from all other accounts controlled by the same person. The right to privacy is compromised by engaging in paid editing. Paid editing accounts must not be allowed to vanish leaving their product live in mainspace.
- --SmokeyJoe (talk) 23:37, 13 March 2019 (UTC)
- I don't support excluding Wikipedians in Residence or WMF employees. If they are making edits for which they are paid to make, they should use similar declared alt. accounts and suffixed usernames: User:Example (WiR), and User:Example (WMF). Not because they are problem editors, but to set the example for best practice. --SmokeyJoe (talk) 00:47, 14 March 2019 (UTC)
- Support. Just as I support banning any form of paid editing (except WIR) and deleting their contributions. Making money out of the work of the volunteers who create and maintain this encyclopedia is dishonorable and unethical. Kudpung กุดผึ้ง (talk) 00:11, 14 March 2019 (UTC)
- Support as a straightforward and long overdue extension of WP:DENY. The only snag I can foresee is that proving UPE is hard, and it wouldn't be in the spirit of speedy deletion to use it when there's merely a suspicion. I'd suggest restricting the new G5 subcriterion to articles created by users who have subsequently been indefinitely blocked/banned for UPE. – Joe (talk) 00:21, 14 March 2019 (UTC)
- Very strong oppose. In order for this to meet the requirements 1 and 2 for a new or expanded criterion (i.e. only applying to things that it should) it would need to be restricted to pages created by confirmed (not just suspected) undisclosed paid editors, for pay (i.e. not other articles they have created) who knew at the time of page creation that they needed to disclose and have not, after a reasonable opportunity to do so, disclosed in an appropriate location that they were/are paid to edit, and the creation was not otherwise permitted by the ToU. Given that it would be impossible for a single admin to verify even half of this it is not remotely suitable for CSD. Even if it were, almost all the actually problematic content would be suitable for speedy deletion under an existing criterion anyway (failing requirement 4 with the remainder probably failing requirement 3 also). Thryduulf (talk) 00:42, 14 March 2019 (UTC)
- Actually, Thryduulf makes an important point. It would be fine to G5 known UPE product, but nearly always, it is a mere suspicion, at best a DUCK test. That is why I proposed: Wikipedia:Quarantine promotional Undeclared Paid Editor product. Quarantine suspected UPE, blanked so that it looks not there, subpages so that "Quarantine" is in the title, but available for the author to defend themselves. Note that the proposal is rough with serious comments on altering the details, on its talk page. --SmokeyJoe (talk) 00:52, 14 March 2019 (UTC)
- @Thryduulf: How about modifying the proposal to delete articles created by editors who have already been blocked for undisclosed paid editing? Often these are checkuser blocks. ~Anachronist (talk) 02:58, 14 March 2019 (UTC)
- @SmokeyJoe: See SoWhy's point below - I would support this only if it applies only to pages that were created in violation of the ToU, which is not necessarily all of them. Thryduulf (talk) 09:52, 14 March 2019 (UTC)
- Thryduulf, I appreciate that principle, you want to discriminate between UPE-TOU violators and other pedestrian COI editors, but how can you tell the difference if you don’t ask? And if you ask, how can you expect an answer with neither stick nor carrot? And why not chase the pedestrian COI editors to answer a few little questions? —SmokeyJoe (talk) 10:38, 14 March 2019 (UTC)
- If you can't tell the difference between UPE and other COI editors then that is another reason why this cannot work - I oppose in the strongest possible terms penalising editors for breaching the ToU when they have done no such thing. If the article is non-neutral then fix it or delete it - you can do this already. If the article is neutral then there isn't a problem that requires deletion. Thryduulf (talk) 10:48, 14 March 2019 (UTC)
- Thryduulf. How do you tell the difference between a UPE and another COI editor? Suppose both are newish accounts, all they have done is written a draft on WP:CORP-borderline company&products, a couple OK sources, a half dozen non-independent PR sources, and another half dozen mere-mention sources. This is typical. I don't think I can tell the difference without a little free form discussion. The problem is, most do not even answer. I suspect most are UPEs, but there is no proof. What would you do in this situation? Give the suspect UPEs the benefit of the doubt, and let them through?
Do you have a problem with OK articles in mainspace when they are the product of undisclosed paid editing? --SmokeyJoe (talk) 11:17, 14 March 2019 (UTC)- If there is no proof that an editor has broken the ToU then it is completely inappropriate for us to be treating them if they have. If there is a neutral, BLP-compliant article about a notable topic in the main namespace then the encyclopaedia would be harmed by deleting it (assuming it's not a copyvio) - why does it matter who wrote it? If the Foundation want editors to rigorously enforce the TOU prohibition on UPE then they need to (a) explicitly ask us and (b) give us the tools to do so reliably. Thryduulf (talk) 11:28, 14 March 2019 (UTC)
- That is logical. I think, for an editor that does not self-declare, given the privacy policy, there can never be proof. Are people getting hyped up about UPE for no good reason? Is there evidence of a problem? Beyond NPP and AfC thinking they have to worry about it? --SmokeyJoe (talk) 11:48, 14 March 2019 (UTC)
- I've never seen any evidence that content written by (suspected) paid editors presents any problems that content of the same standard written by other editors does. If an article is irredeemably spammy it should be deleted, if an article is good quality it should not - who wrote it ins't relevant. Thryduulf (talk) 11:54, 14 March 2019 (UTC)
- I have never seen content written by an undisclosed paid editor that was any good. Often it looks superficially accurate but when you start looking at the refs they are typically poor and many often do not support the content they are placed behind. Paid editing is trying to mislead our readers and thus it harms our encyclopedia and our reputation. Those doing it are not interested in becoming editors who contribute high quality content but simple want to promote those who pay them and will try anything to continue to do so. Doc James (talk · contribs · email) 12:23, 14 March 2019 (UTC)
- "I have never seen content written by an undisclosed paid editor that was any good." in which case it can and should be fixed or deleted like any other bad content - that the author was (or might have been) paid is irrelevant. However, I actually suspect that there is good content produced by UPE that doesn't get noticed because it's good and doesn't actually cause any problems. Thryduulf (talk) 13:23, 14 March 2019 (UTC)
- I have never seen content written by an undisclosed paid editor that was any good. Often it looks superficially accurate but when you start looking at the refs they are typically poor and many often do not support the content they are placed behind. Paid editing is trying to mislead our readers and thus it harms our encyclopedia and our reputation. Those doing it are not interested in becoming editors who contribute high quality content but simple want to promote those who pay them and will try anything to continue to do so. Doc James (talk · contribs · email) 12:23, 14 March 2019 (UTC)
- I've never seen any evidence that content written by (suspected) paid editors presents any problems that content of the same standard written by other editors does. If an article is irredeemably spammy it should be deleted, if an article is good quality it should not - who wrote it ins't relevant. Thryduulf (talk) 11:54, 14 March 2019 (UTC)
- That is logical. I think, for an editor that does not self-declare, given the privacy policy, there can never be proof. Are people getting hyped up about UPE for no good reason? Is there evidence of a problem? Beyond NPP and AfC thinking they have to worry about it? --SmokeyJoe (talk) 11:48, 14 March 2019 (UTC)
- If there is no proof that an editor has broken the ToU then it is completely inappropriate for us to be treating them if they have. If there is a neutral, BLP-compliant article about a notable topic in the main namespace then the encyclopaedia would be harmed by deleting it (assuming it's not a copyvio) - why does it matter who wrote it? If the Foundation want editors to rigorously enforce the TOU prohibition on UPE then they need to (a) explicitly ask us and (b) give us the tools to do so reliably. Thryduulf (talk) 11:28, 14 March 2019 (UTC)
- Thryduulf. How do you tell the difference between a UPE and another COI editor? Suppose both are newish accounts, all they have done is written a draft on WP:CORP-borderline company&products, a couple OK sources, a half dozen non-independent PR sources, and another half dozen mere-mention sources. This is typical. I don't think I can tell the difference without a little free form discussion. The problem is, most do not even answer. I suspect most are UPEs, but there is no proof. What would you do in this situation? Give the suspect UPEs the benefit of the doubt, and let them through?
- If you can't tell the difference between UPE and other COI editors then that is another reason why this cannot work - I oppose in the strongest possible terms penalising editors for breaching the ToU when they have done no such thing. If the article is non-neutral then fix it or delete it - you can do this already. If the article is neutral then there isn't a problem that requires deletion. Thryduulf (talk) 10:48, 14 March 2019 (UTC)
- Thryduulf, I appreciate that principle, you want to discriminate between UPE-TOU violators and other pedestrian COI editors, but how can you tell the difference if you don’t ask? And if you ask, how can you expect an answer with neither stick nor carrot? And why not chase the pedestrian COI editors to answer a few little questions? —SmokeyJoe (talk) 10:38, 14 March 2019 (UTC)
- @SmokeyJoe: See SoWhy's point below - I would support this only if it applies only to pages that were created in violation of the ToU, which is not necessarily all of them. Thryduulf (talk) 09:52, 14 March 2019 (UTC)
- @Thryduulf: How about modifying the proposal to delete articles created by editors who have already been blocked for undisclosed paid editing? Often these are checkuser blocks. ~Anachronist (talk) 02:58, 14 March 2019 (UTC)
- Actually, Thryduulf makes an important point. It would be fine to G5 known UPE product, but nearly always, it is a mere suspicion, at best a DUCK test. That is why I proposed: Wikipedia:Quarantine promotional Undeclared Paid Editor product. Quarantine suspected UPE, blanked so that it looks not there, subpages so that "Quarantine" is in the title, but available for the author to defend themselves. Note that the proposal is rough with serious comments on altering the details, on its talk page. --SmokeyJoe (talk) 00:52, 14 March 2019 (UTC)
Support. Paid editing on Wikipedia is reaching crisis-level proportions and we need to deal with it as such. — pythoncoder (talk | contribs) 01:45, 14 March 2019 (UTC)
- Support option 2 as it is more objective. — pythoncoder (talk | contribs) 13:03, 14 March 2019 (UTC)
- Please explain how your suggestion relates to the previous RFC. --Izno (talk) 02:10, 14 March 2019 (UTC)
- I'm still waiting on an explanation of why we think the prior consensus has changed... would anyone care to let those persons know, also? --Izno (talk) 19:10, 17 March 2019 (UTC)
- Over half of AfC submissions are likely UPE or COI edits. This could really cut down the AfC workload. Legacypac (talk) 02:12, 14 March 2019 (UTC)
- The way to make this criterion objective is to foist the uncertainty off on another process. To wit: the content should only be speedyable if its creator has already been indefinitely blocked as a paid editor. —Cryptic 02:24, 14 March 2019 (UTC)
- Support Option 2 as it is clear cut easy to see. I'm going to suggest we try to feed a notice about COI and UPE to every submitter of content at AfC that someone might pay for. Maybe a bot can do that. Even if it get posted to editors that are writing historic topics etc who cares because it will raise awareness without accusing them. Legacypac (talk) 03:51, 14 March 2019 (UTC)
- Oppose particularly at AFC. It is difficult to determine whether the writer is paid or has made a disclosure. So this is not suitable for a speedy deletion. At AFC pages will be examined to see if they are promotional or not. It gives a UPE editor a chance to learn they need to disclose. Graeme Bartlett (talk) 05:31, 14 March 2019 (UTC)
- I'm suggesting a notification system to encourage disclosure. We would only speedy drafts at AfC created by blocked UPE users. Often these drafts get worked on by sock after sock so flushing them from the system would be a good thing. Why waste my volunteer time to ensure the UPE gets his/her paycheck? Why make it easy for them to violate our rules? Legacypac (talk) 05:47, 14 March 2019 (UTC)
- Oppose How do you tell that something was created for undisclosed payments? Idle speculation and "I think so" is not clear enough. Besides, G11 is a thing for spammy articles. Jo-Jo Eumerus (talk, contributions) 06:30, 14 March 2019 (UTC)
- G11 only catches the product of the most inept UPEs. Granted, there are lots of them, but they are noisy inept UPEs that will learn how to avoid G11, and G11 leaves no record for the non-admin reviewers to refer to when they try again, and again, and again. --SmokeyJoe (talk) 06:41, 14 March 2019 (UTC)
- Why does no one like my quarantine idea?
- * The suspected UPE page can't be deleted, because the reviewer rarely can know it is UPE objectively enough for any acceptable CSD.
- * It is not worth a community discussion for every suspected UPE creation, NPP and AfC reviewers have to be trusted on this to do something.
- * The page has to be blanked, so that the UPE is not recognized for the work in progress.
- * The page and every version of it has to have the ugly title, including "Quarantined", so that the UPE can't even send the sponsor a version link. (Achieved by the page move)
- * if the author can declare, or explain that the are not a UPE, then the reviewer can move the page back, no admin functions required.
- The quarantine proposal talk page has productive input on details. I think the concept is the only viable action I've seen. A CSD based on the unknowable author=UPE condition is not workable. A CSD requiring the author to be blocked will miss 99% of the problem. --SmokeyJoe (talk) 06:39, 14 March 2019 (UTC)
- Short answer on that proposal - too complex and too much work for reviewers. A blank in place for drafts might work with a message about UPE/COI much like we do with suspected copyvio. We could make it a CSD with a delayed deletion, it only shows up in the CSD pending list after X days. That can be programmed right into the CSD template. Give the user time to disclose and remove the CSD. Otherwise bye bye. Legacypac (talk) 07:04, 14 March 2019 (UTC)
- It is not meant to be work for the reviewer. It is meant to be less work than giving a reviewing comment. If the concept is agreed to, everything is easily scripted.
- You can't have a CSD for suspected UPE. Anachronist's proposal is doomed for this reason, just like the previous one last time. --SmokeyJoe (talk) 07:20, 14 March 2019 (UTC)
- Short answer on that proposal - too complex and too much work for reviewers. A blank in place for drafts might work with a message about UPE/COI much like we do with suspected copyvio. We could make it a CSD with a delayed deletion, it only shows up in the CSD pending list after X days. That can be programmed right into the CSD template. Give the user time to disclose and remove the CSD. Otherwise bye bye. Legacypac (talk) 07:04, 14 March 2019 (UTC)
- Possibly support #2 delayed. I could entertain supporting some kind of UPE-PROD which gives these editors the chance to disclose per Legacypac above or dispute the UPE and a block based on that. After all, just because an admin has decided to block someone for UPE does not mean they are an UPE. However, I do see the problem that in most cases, such a deletion mechanism will fail due to the uncertainties surrounding UPE and how to prove it. Regards SoWhy 08:29, 14 March 2019 (UTC)
- Support option 2 While I agree that suspicions of UPE are not sufficient for the deletion of a page (that's why we have the
{{undisclosed paid}}
template), if it is confirmed that a user has been editing in violation of Wikipedia's TOU, that's a good reason for deletion. As a comparative analogy, if a user is blocked for copyright violations, we delete pages they have created which are violating copyright, and we do so regardless of whether they were created before or after the block. In this scenario, we are blocking a user for violating the TOU, and the pages that they created prior to being blocked are part of that violation - they should, therefore, be deleted. Yunshui 雲水 08:37, 14 March 2019 (UTC)- I don't think that analogy is apt. If we block an editor for repeated copyright violations, we still have to check whether all their creations really fit G12 because just because they violated copyright in some cases does not mean they did so in all cases. Similarly, someone blocked for UPE does not mean all their articles were created because they were paid for it. Regards SoWhy 08:50, 14 March 2019 (UTC)
- Having dealt with some large cases of copyright issues, after finding copyright violations in 10 random edits from a single user, yes large scale rollback becomes a perfectly reasonable option and one I have carried out. At this point instances that do not look like copyright violations from this individual might just mean that the source they copied from is no longer easily avaliable online rather than the content being "okay". Doc James (talk · contribs · email) 13:17, 14 March 2019 (UTC)
- I don't think that analogy is apt. If we block an editor for repeated copyright violations, we still have to check whether all their creations really fit G12 because just because they violated copyright in some cases does not mean they did so in all cases. Similarly, someone blocked for UPE does not mean all their articles were created because they were paid for it. Regards SoWhy 08:50, 14 March 2019 (UTC)
- Support (option 2) - Articles written by confirmed UPEs should be speedily deleted for two reasons: 1) there is a high probability that the articles were also paid for, and 2) deleting all article created by the UPE would have the same disincentivising effect as it does for socks of blocked/banned users.- MrX 🖋 11:25, 14 March 2019 (UTC)
- Very strong support of both options plus suggestions by User:SmokeyJoe. When a family of 3 plus socks are blocked and all the articles created by these socks are 1) promotional 2) barely / not notable 3) each sock makes a dozen small edits, waits a week, and then creates a perfect article in a single edit. It does not take a rocket scientist (or AI specialist) to identify this as undisclosed paid editing (and by accounts of a prior blocked user). I currently delete these articles as they are created by previously blocked users (we do not need to bury our heads in the sand). In fact all articles that follow this pattern could really be simple deleted. User:SmokeyJoe suggestions are excellent and are definitely required if we are going to allowed paid "promotional" editing to continue at all. My issue with paid editing in the type were those doing the paying have a COI regarding the subject matter in question (ie someone paying for an article about themselves or their business). The NIH/CDC paying someone to help improve articles about hearing loss for example is not an issue as the NIH/CDC do not have a COI with respect to the topic in question. They of course should not and do not work on content about the NIH or CDC itself. Doc James (talk · contribs · email) 12:16, 14 March 2019 (UTC)
- If a family of socks are blocked then their contributions can already be deleted under G5 - why they were socking is irrelevant. The latter part of your comment just proves that paid editing (disclosed or otherwise), COI editing and promotional editing are three different issues - they are overlapping sets but all combinations of 0, 1, 2 and all 3 of them exist. If the content is bad it should be (and can be) fixed or deleted using existing processes so there is no need for this proposal; if the content is good then there is no need to delete it so there is no need for this proposal. Thryduulf (talk) 13:23, 14 March 2019 (UTC)
- Support Proposal 2 - I also believe it is unjust and immensely unwise to block articles because they are believed to be from PAIDCOI editors. However I'm all for scrapping the work of confirmed paid editors. While we might actually do some collateral damage this way, it should make it much harder for the paid editor to keep doing his action as a number of his clients suddenly become grumpy. Nosebagbear (talk) 13:35, 14 March 2019 (UTC)
- Support proposal 2 with a caveat that it is not very retrospective and only includes arrticles created from March 2018 because many upe articles that have been around for a few years have a lot of contributions from legitimate editors and the G5 criteria in practice is applied very unevenly so that some articles will be deleted no matter whether the other legitimate contributions are substantial and significant, regards Atlantic306 (talk) 15:53, 14 March 2019 (UTC)
- Oppose Proposal 2, as it would include all pages they created, even those they were not paid to create. That, to me, is overreaching and suggests if an editor was paid to create one page and created 99 good pages without payment, all 100 pages are eligible for speedy deletion. That's not productive. Neutral for now on Proposal 1. Smartyllama (talk) 18:07, 14 March 2019 (UTC)
- @Smartyllama: Both proposals would include all pages UPE created. The only difference is which users are covered, in proposal one it's everybody suspected to be an undisclosed paid editor (whether this is proven or not), in proposal 2 only those people who have been indefinitely blocked for undisclosed paid editing (whether proven or not). Thryduulf (talk) 18:32, 14 March 2019 (UTC)
- In that case, Oppose Both but Support a potential third option which only includes pages the editor was actually paid to create, as those were the only edits in violation of WMF policy. Smartyllama (talk) 18:43, 14 March 2019 (UTC)
- @Smartyllama: Both proposals would include all pages UPE created. The only difference is which users are covered, in proposal one it's everybody suspected to be an undisclosed paid editor (whether this is proven or not), in proposal 2 only those people who have been indefinitely blocked for undisclosed paid editing (whether proven or not). Thryduulf (talk) 18:32, 14 March 2019 (UTC)
- Oppose. UPEs have long been alleged to be socks at SPI. I don't know the stats, but certainly frequently they are not socks. I have no objection to some sort of proposal to add a G code for UPEs, but expanding G5 in this way would be very messy. As it is, many editors tag articles incorrectly, and there are admins that either don't understand the language of G5 or who IAR-go along with it. I suspect a new G code just for UPEs would also be messy, but let's at least keep our messes separate.--Bbb23 (talk) 18:38, 14 March 2019 (UTC)
- Oppose per Thryduulf and Bbb23. Additionally, the only valid reason for a UPE's contribs to be speedy deleted is already covered by WP:G11. Ivanvector (Talk/Edits) 19:24, 14 March 2019 (UTC)
- Oppose unless it is limited to cases where the article "is the sole product, ignoring minor edits, of a [blocked] UPE editor or editors", per SmokeyJoe. I've seen plenty of blatently promotional articles rescued by uninvolved editors, and we wouldn't want to throw those babies out with the bathwater. --Ahecht (TALK
PAGE) 19:44, 14 March 2019 (UTC) - Oppose I support the general idea but this version does not restrict itself to articles created for pay. If I accept a commission to write a paid article tomorrow and don't disclose it then any article I've ever written could be deleted under this. Hut 8.5 20:42, 14 March 2019 (UTC)
- part of the point of this is to discourage people from doing just that. It wouldn't conceivably apply to you, because an editor with skill and experience and knowledge of WP, would know to declare. DGG ( talk ) 21:33, 14 March 2019 (UTC)
- Sure, I have no desire to take up paid editing but the same principle applies. If an editor gets banned then G5 isn't used to delete everything they have ever written, even though that would be a deterrent to doing something ban-worthy. Hut 8.5 21:56, 15 March 2019 (UTC)
- part of the point of this is to discourage people from doing just that. It wouldn't conceivably apply to you, because an editor with skill and experience and knowledge of WP, would know to declare. DGG ( talk ) 21:33, 14 March 2019 (UTC)
- Support option 2. with the same understanding as for other G5s, that if a good-faith editor has edited it substantially, it does not apply. I have in the past tried to rescue articles of importance and clear notability even for UPEs, and this will make it more difficult, but considering the threat that they pose, it is necessary. I do point out that by adopting it we eliminate the possibility of a UPE reforming and declaring their earlier work. But this is not that much of a change, because even now they would have to declare their earlier work, tho it would not get speedy deleted. -- several such editors have contacted me, and they are not willing to declare earlier work regardless, claiming confidentiality. I would suggest proceeding very slowly withe earlier ones. DGG ( talk ) 21:21, 14 March 2019 (UTC)
- @DGG: What is the actual "threat" here that is not posed by unpaid COI edits? Why does this require speedy deletion of every article created someone we suspect of engaging in UPE - regardless of whether we are right, and regardless of whether that article was created for pay? Thryduulf (talk) 00:58, 15 March 2019 (UTC)
- Unpaid COI editors are normally friends or associates or fans of the subject. They can, at best, be educated about our standards and will try to meet them and sometimes even go on to editing properly; at worst, they will quickly leave. We need to be flexible and open to potential good faith contributors. (If they're the subject in person, then it's a different problem--their sense of self-importance is involved, and they will generally become so obnoxious about it that we can quickly remove them.) Paid COI editors, the ones who are undeclared especially, almost never can be educated about our standards; they can be stopped at a particular article, but many of them seem to return indefinitely. They have no interest in being good faith editors. . (I'm usingCOI in the sense of specific interest in having a particular article , not contributing with a COI to WP generally) DGG ( talk ) 01:45, 15 March 2019 (UTC)
- Another consideration is that COI editors are often in a better place to find sources to support an edit on the topic, given that they are connected to it. Jo-Jo Eumerus (talk, contributions) 06:34, 15 March 2019 (UTC)
- Broken windows theory applies to promotional articles because newbies justify their existence with WP:OSE. There's also cases where articles - they don't have to be spam, mind you - were created for pitching to potential clients (who are not necessarily the subject of the article) or business development purposes. Even though these articles are not created explicitly for pay, they still need to be nuked - perhaps more urgently. MER-C 18:34, 17 March 2019 (UTC)
- Unpaid COI editors are normally friends or associates or fans of the subject. They can, at best, be educated about our standards and will try to meet them and sometimes even go on to editing properly; at worst, they will quickly leave. We need to be flexible and open to potential good faith contributors. (If they're the subject in person, then it's a different problem--their sense of self-importance is involved, and they will generally become so obnoxious about it that we can quickly remove them.) Paid COI editors, the ones who are undeclared especially, almost never can be educated about our standards; they can be stopped at a particular article, but many of them seem to return indefinitely. They have no interest in being good faith editors. . (I'm usingCOI in the sense of specific interest in having a particular article , not contributing with a COI to WP generally) DGG ( talk ) 01:45, 15 March 2019 (UTC)
- Support option 2. The only way to rid ourselves of this menace is to deny the spammers their product. They are not here to improve Wikipedia and we shouldn't expend more than the bare minimum of time needed to dealing with them - they can just overwhelm us with cheap labor. AFD doesn't scale very well and many UPE articles are specifically created to frustrate the notability evaluation process (e.g. by WP:REFBOMBing). Whether they are explicitly sockpuppets or not is irrelevant - many UPE operations are sophisticated enough to evade CU and/or farm out the actual page creations to low wage/third world/freelance meatpuppets with instructions on evading detection. Quarantine should be used when UPE is merely suspected. The analogy with copyright violations is correct - it is policy that contributions of repeat infringers can be presumed to be copyvios and removed indiscriminately. MER-C 18:51, 17 March 2019 (UTC)
- Oppose any version of this. I am as strongly opposed to undisclosed paid editing, and paid editing in general, as anyone. However, no version has been proposed that is Objective and Uncontestable. Robert McClenon (talk) 19:53, 17 March 2019 (UTC)
- Support. — Jeff G. ツ 23:10, 17 March 2019 (UTC)
- Oppose per Jo-Jo Eumerus. How many of you supporters are familiar with the big box at the top of the page? Look under Read this before proposing new criteria if you're not. This proposal is neither objective nor uncontestable, especially as it's quite plausible that an article potentially deletable under this proposed criterion is beneficial to Wikipedia. We shouldn't go deleting good content merely because the creator got paid to create it: we should delete it if the content's demonstrably bad or if the situation's ambiguous, but this proposal would have us delete all paid-edit content, even when it's demonstrably good. Nyttend (talk) 02:50, 18 March 2019 (UTC)
- Oppose - Unless an editor discloses that they are a paid editor, how can it objectively be known that they are? — Godsy (TALKCONT) 03:46, 18 March 2019 (UTC)
- Oppose. Content should be evaluated by its own merits, not those of its creator. Benjamin (talk) 06:27, 18 March 2019 (UTC)
- Oppose Thryduulf and Nyttend raise good points. It's impossible to quickly verify whether the article meets the proposed criteria. And if it's not quick, it's not speedy deletion. feminist (talk) 10:10, 18 March 2019 (UTC)
- A wonderful, impractical dream. A criterion that, if it could be objectively and speedily enforced, would make Wikipedia a better place. ~ ToBeFree (talk) 05:43, 19 March 2019 (UTC)
- Oppose both. As drafted, both proposals are overly broad, capturing work that have substantial contributions by others, weren't written while blocked, weren't written for pay, and combinations thereof. It couldn't work as an extension of G5, which is specifically for pages created in violation of a user's specific block w/o substantial contributions by others. A more narrower and objective CSD criteria could work, or some PROD or quasi-PROD process might be better. ---- Patar knight - chat/contributions 20:53, 19 March 2019 (UTC)
- Oppose. Undeclared paid editing is not even agreed to be a deletion reason. The place for this discussion is Wikipedia talk:Deletion policy#Undeclared Paid Editor (UPE) product; only if UPE AfDs consistently result in SNOW deleted should this be considered as a CSD. --SmokeyJoe (talk) 00:32, 20 March 2019 (UTC)
- Oppose. Thryduulf and Nyttend are right again with this one. --Bsherr (talk) 15:13, 20 March 2019 (UTC)
- Oppose as per WP:BITE. Some newcomers may not know about declaring paid editing. I know the best wiki (talk) 15:55, 20 March 2019 (UTC)
- Oppose per Thryduulf's valid point. It's literally impossible to tell whether the article was a COI creation, which would ruin the point of speedy deletion. --GN-z11 ☎ ★ 18:48, 20 March 2019 (UTC)
- Oppose, mostly per Thryduulf, though I like SmokeyJoe's idea. — SMcCandlish ☏ ¢ 😼 16:05, 21 March 2019 (UTC)
- Oppose If the article isn't bad enough to delete G11, there should be a discussion about it, not a speedy deletion based on the author. Monty845 05:03, 24 March 2019 (UTC)
- Oppose per WP:CREEP and WP:FOC. Andrew D. (talk) 12:03, 25 March 2019 (UTC)
- Oppose per above. Should be done on a case by case basis with community input, not summarily by admins. Wugapodes [thɑk] [ˈkan.ˌʧɹɪbz] 01:03, 26 March 2019 (UTC)
- Oppose per Benjamin: "Content should be evaluated by its own merits, not those of its creator." As well, one editor with 99 good articles shouldn't have those all deleted over one undisclosed-paid one. ɱ (talk) 17:05, 26 March 2019 (UTC)
- Oppose For so many of the reasons listed above that it would take too long to list them all. · · · Peter Southwood (talk): 06:33, 27 March 2019 (UTC)
- Oppose - For all the reasons already stated. If it doesn't qualify under G11, then not only should it get a discussion, there should be even less confidence in the identity of the supposedly conflicted creator. MarginalCost (talk) 01:22, 1 April 2019 (UTC)
- Oppose Outright deletion of content, except as already defined per CSD. UPE is unethical but there is absolutely no basis to say that the content produced as a product of UPE cannot be salvaged at all. Instead of going on a crusade against UPE, if we all improved Wikipedia (by improving UPE bs, for example) we'd be better off. --QEDK (後 ☕ 桜) 18:07, 4 April 2019 (UTC)
Alternative proposal (3)
To remedy some of the opposers’ concerns, I propose that the following be added to the G5 bullet list in lieu of any of the above:
- In addition, because undisclosed paid editing is prohibited by Wikimedia Foundation policy, articles created for pay by editors who have been indef-blocked for undisclosed paid editing, with no substantial edits by other users, are candidates for speedy deletion under G5.
(Changes from proposal 2 marked in red.) — pythoncoder (talk | contribs) 21:47, 14 March 2019 (UTC)
- Support as proposer. — pythoncoder (talk | contribs) 21:47, 14 March 2019 (UTC)
- Strong oppose because these conditions are impossible for a single admin reviewing a CSD nomination to determine. Also, it fails to address other problems noted above - principally lack of need and not being restricted to proven cases. Thryduulf (talk) 00:58, 15 March 2019 (UTC)
- It may be hard to tell if the article was created for pay, but if it is obvious, then this should be possible to enforce with a speedy deletion. However if other good standing editors have adopted the page or removed the speedy delete tag, then this should not be foreced, and AFD considered instead. Graeme Bartlett (talk) 02:54, 15 March 2019 (UTC)
- "It may be hard to tell" and "sometimes it's obvious" (without any indication of who gets to decide, using what criteria, or anything remotely objective) is exactly why this subjective assessment is has no business being anywhere near a CSD criterion. Thryduulf (talk) 10:35, 15 March 2019 (UTC)
Oppose as written, but would support if "with no substantial edits by other users" were added to keep it in line with the rest of G5. Smartyllama (talk) 11:24, 15 March 2019 (UTC)
- This would already be a required step because it's part of the G5 requirements, but I have inserted it above for clarity. — pythoncoder (talk | contribs) 12:02, 15 March 2019 (UTC)
- Support current version as my concerns above have been addressed. Smartyllama (talk) 12:28, 15 March 2019 (UTC)
- Support, although in practice, we won't know which articles have been written for pay. But I guess we'll assume the article about a company CEO was written for pay and the one about an obscure 18th century poet wasn't, and we'll get it right most of the time. —Kusma (t·c) 11:46, 17 March 2019 (UTC)
- Sorry, but whether we "get it right most of the time" is unknowable and even if it were it wouldn't be good enough. Thryduulf (talk) 14:28, 18 March 2019 (UTC)
- Support this too, though prefer proposal 2. MER-C 18:52, 17 March 2019 (UTC)
- Oppose as above. Robert McClenon (talk) 19:53, 17 March 2019 (UTC)
- Support, this one looks like has been ironed out enough. I think I would prefer a separate number for this per Bbb23's concern above, but I understand why it's bundled with G5. -- Tavix (talk) 21:32, 17 March 2019 (UTC)
- Support. — Jeff G. ツ 23:11, 17 March 2019 (UTC)
- Support' along with proposals 1 & 2 (with preference reflected in the proposal number). Best, Barkeep49 (talk) 00:01, 18 March 2019 (UTC)
- Support as this is clear and can be decided by only examining a few pages (history and user log). Graeme Bartlett (talk) 10:10, 18 March 2019 (UTC)
- How? Yes it's easy to tell whether the editor is indef blocked, and not difficult to tell whether there have been significant contributions from others but how do you propose to reliably and objectively tell whether any given article was created for pay? Note that simply suspecting that it might have been is insufficient. Thryduulf (talk) 14:28, 18 March 2019 (UTC)
- Oppose per Thryduulf. feminist (talk) 10:14, 18 March 2019 (UTC)
- Oppose. Just leave G5 alone, it's fine as is for quickly cleaning up ban violations, which is what it's meant for. For suspected paid editors, make a proposal to modify G11, or propose a new criterion. Ivanvector (Talk/Edits) 14:38, 18 March 2019 (UTC)
- @Ivanvector: The idea here is that undisclosed paid editors are already "banned" by default. Therefore, wouldn't it be appropriate to use G5 to clean up these ban violations also? Remember, the proposal in this section applies only to contributions of editors who have already been blocked for undisclosed paid editing, and often those blocks are by checkusers. ~Anachronist (talk) 15:23, 20 March 2019 (UTC)
- Yes, that's a common view (that undisclosed paid editors are violating an implied ban) and one I share, but it's not universal, and so it doesn't meet the "objective" requirement for new criteria. For socks of editors who are already banned by a community process, G5 already applies. As a side note, we were explicitly warned in orientation that suspicion of commercial editing is not a valid rationale on its own for the use of Checkuser - if a user is Checkuser-blocked, you can presume it's not because of undisclosed paid editing but because of some violation of policy related to multiple account abuse. They don't go hand-in-hand as often as people like to think. Ivanvector (Talk/Edits) 16:23, 20 March 2019 (UTC)
- @Ivanvector: The idea here is that undisclosed paid editors are already "banned" by default. Therefore, wouldn't it be appropriate to use G5 to clean up these ban violations also? Remember, the proposal in this section applies only to contributions of editors who have already been blocked for undisclosed paid editing, and often those blocks are by checkusers. ~Anachronist (talk) 15:23, 20 March 2019 (UTC)
- Oppose. Better than the above but still too subjective per Thryduulf. If this does get consensus, it should be a new criterion not under G5, which is for very clear cut violations of the user's original block/ban. If this must be under G5, then it should be restricted to very clear cut cases where someone is blocked for socking (e.g. explicitly mentioned in the block, link to SPI, Checkuserblock) after their first account was blocked for UPE. ---- Patar knight - chat/contributions 21:08, 19 March 2019 (UTC)
- Oppose as not objective. Instead, to move forwards on this, introduce a tracking category on UPE AfDs, to provide evidence on what normally happens. --SmokeyJoe (talk) 00:25, 20 March 2019 (UTC)
- Oppose. Per Thryduulf. --Bsherr (talk) 15:19, 20 March 2019 (UTC)
- Support as the person who originally started all this. I like this better than my option 2 in the section above. ~Anachronist (talk) 15:23, 20 March 2019 (UTC)
- Oppose per Thryduulf again. This is closer, but it's not there yet. — SMcCandlish ☏ ¢ 😼 16:05, 21 March 2019 (UTC)
- Oppose - This would allow adaptive block rationales to make pages deletable, which is an abhorrent notion. — Godsy (TALKCONT) 19:29, 23 March 2019 (UTC)
- Oppose Same as main proposal, if the article isn't bad enough to delete G11, it should go to a discussion, not be deleted due to the author. Monty845 05:05, 24 March 2019 (UTC)
- Oppose per WP:CREEP and WP:FOC. Andrew D. (talk) 12:03, 25 March 2019 (UTC)
- Weak support. I hate G5, but paid-for spam needs to be dealt with better than it is right now. --Piotr Konieczny aka Prokonsul Piotrus| reply here 06:01, 2 April 2019 (UTC)
- Support - If paid editors can easily continue to create new articles with new accounts, they will not stop. We need to do more to actually deter paid editing before it happens. We must make it clear that if you break our rules about paid editing you cannot get away with it. Meszzy2 (talk) 18:10, 5 April 2019 (UTC)
Another alternative (4) (or possibly, an addition)
Articles written by UPE can be brought to AfD, and if the consensus is that they were written by a blocked UPE, this will be a sufficient reason for deletion, regardless of considerations of possible notability and promotionalism
- This way no one person gets to decide, and there is a possibility of making exceptions. The disadvantage of this is having a large number of inconclusive AfD debates, so I'm not sure of this. I'm suggesting it only as a possible alternative to see what people think. (And, of course, it is't actually just one person in Speedy. Good practice is for one person to nominate, and then a second person who is an admin to decide.) DGG ( talk ) 01:52, 15 March 2019 (UTC)
- This sounds reasonable, but AFD would always have been an option. I suppose that the deletion policy is what gets changed by thgis proposal. We shouuld make it clear that in this case it is the UPE that is the only substantial contributor, to avoid the case where pre-existing articles get edited by the paid editor. (in which case ubndoing edits would be appropriate). Graeme Bartlett (talk) 02:57, 15 March 2019 (UTC)
Additional possibility: remove the word blocked. If we are going by consensus the consensus can decide. (There will be an additional proposal to revise blocking policy. We now often presume any upe is likely to have socked, butit would bemuch more straightforward for UPE should be stated explicitly as a reason for blocking. DGG ( talk ) 04:52, 15 March 2019 (UTC)
- AfD can delete any article for any reason per the consensus established in the AfD discussion. User:Thryduulf has provided a fundamental challenge. Who says ToU violation is a deletion reason? Did the WMF? Did editors decide? Links? --SmokeyJoe (talk) 05:28, 15 March 2019 (UTC)
- We can decide. (though the discussion would have to take place at WP:Deletion Policy). We make our own policy. It would add to the list at WP:DEL-REASON. Though we can delete for any other reason also, reasons not on that list are in practice usually strongly challenged. Leaving it to individual AfDs would repeat this discussion each time, whether UPE is an acceptable reason. We need some uniformity of practice in dealing with COI contributors. DGG ( talk ) 14:17, 15 March 2019 (UTC)
- Wikipedians can decide, right. But not on this page. It is not even a WP:DEL-REASON, so it is not even conceivable that it should be a CSD. —SmokeyJoe (talk) 04:27, 16 March 2019 (UTC)
- We can decide. (though the discussion would have to take place at WP:Deletion Policy). We make our own policy. It would add to the list at WP:DEL-REASON. Though we can delete for any other reason also, reasons not on that list are in practice usually strongly challenged. Leaving it to individual AfDs would repeat this discussion each time, whether UPE is an acceptable reason. We need some uniformity of practice in dealing with COI contributors. DGG ( talk ) 14:17, 15 March 2019 (UTC)
- Any article can be brought to AfD for any reason by any editor at any time. This proposal adds nothing to that. Thryduulf (talk) 10:37, 15 March 2019 (UTC)
- Support Yes reasonable. Helps with causes of a bunch of paid for articles by a single account. Doc James (talk · contribs · email) 11:41, 15 March 2019 (UTC)
- Wrong forum this is not a proposal for speedy deletion so the discussion should be happening at an appropriate venue - presumably Wikipedia talk:Deletion policy. However it is unnecessary as all you would be doing is adding a bullet point to Wikipedia:Deletion policy#Reasons for deletion which duplicates existing points 1, 4, 6, 7, 13 and/or 14. Thryduulf (talk) 14:07, 15 March 2019 (UTC)
- of course it would need to be discussed at DelPolicy. And it I think ought to be viewed as a supplement to the discussion here which might use narrower criteria. For example, CSD would be blocked, AfD would be any UPE even if not yet blocked. The reasons given for deletion at DELPOLICY already overlap. It adds clarity to be specific. , whith something that can be quoted at the AfD. We do not want to add confusion to what many new users see as an already complicated and confusing process. DGG ( talk ) 14:30, 15 March 2019 (UTC)
- Oppose as above. Robert McClenon (talk) 19:53, 17 March 2019 (UTC)
- Oppose AfD can choose to delete an article under existing policy for promotionalism. This just muddies the waters. Best, Barkeep49 (talk) 23:59, 17 March 2019 (UTC)
- AfD can only delete under promotionalism if it reaches the level that would justify a WP:G11 deletion anyway. Paid Editing might cause the same problem, but it is not the same reasoning.
- actually, based on the last few years of decisions at AfD, Deletion for promotionalism it is interpreted more broadly. This is justified by WP:DEL4, " Advertising or other spam without any relevant or encyclopedic content )" The CSD criterion for G11, WP:G11 is "Unambiguous advertising or promotion: ... This applies to pages that are exclusively promotional and would need to be fundamentally rewritten to conform with Wikipedia:NOTFORPROMOTION " The difference, as for all speedy criteria, is that is has to be unambiguous. Situations that are debatable need to be debated at AfD, and promotionalism there can be and is interpreted however the consensus decides. Many is the promotional page where I have declined G11, nominated for AfD as promotional, and seen deleted by consensus. DGG ( talk ) 17:59, 18 March 2019 (UTC)
- Support with Bartlett's amendment, - This is a different case to those above, and I think it is a reasonable one. It handles comments that it either isn't appropriate for a CSD standard time scale, or doesn't belong because it's too long for CSD (a somewhat tricky double argument!). Given the significant opposition primarily focused on being discussion in this forum, @DGG:, closing this here, moving to Del Policy while both leaving a forwarding and pinging each participant in this proposal would be reasonable/wise. Nosebagbear (talk) 17:40, 18 March 2019 (UTC)
- I agree with closing this here,. The discussion has moved on. DGG ( talk ) 17:59, 18 March 2019 (UTC)
- AfD can only delete under promotionalism if it reaches the level that would justify a WP:G11 deletion anyway. Paid Editing might cause the same problem, but it is not the same reasoning.
- Wrong forum per above. UPE should obviously be a factor to consider at AFD that would militate towards deletion barring exceptionally good other factors (e.g. substantial, transformative contributions by others, very high quality material explicitly endorsed by other editors, etc.) but implementing changes to AFD is outside the scope of this page. ---- Patar knight - chat/contributions 21:14, 19 March 2019 (UTC)
- Oppose in favor a restarting at WT:DEL, should UPE be a WP:DEL#REASON? --SmokeyJoe (talk) 00:26, 20 March 2019 (UTC)
- Oppose per Barkeep49, et al. This isn't a CSD proposal so is off-topic here, and it's not necessary since AfD can already delete such pages as hopelessly promotional. — SMcCandlish ☏ ¢ 😼 16:05, 21 March 2019 (UTC)
- Oppose If the AFD discussion on the merits results in a keep consensus, the identity of the author should not override that. Monty845 05:10, 24 March 2019 (UTC)
- Oppose per WP:CREEP and WP:FOC. Andrew D. (talk) 12:03, 25 March 2019 (UTC)
Alternative wording (5)
Here's my concept for this new criterion:
"In addition, any articles confirmed to have been created by a confirmed undisclosed paid editor, are promotional, and have no substantial edits by others, are subject to this criterion due to violating Wikipedia terms of service."
Since other proposals were said to be too broad (not counting the wrong venue proposal), I thought I'd throw my hat into the ring with this potential wording of the new criterion for G5.
Hopefully this is short but concise enough to do. Kirbanzo (userpage - talk - contribs) 00:56, 24 March 2019 (UTC)
- Support as proposer. Kirbanzo (userpage - talk - contribs) 00:56, 24 March 2019 (UTC)
- Oppose If the article is problematically promotional, it can already be deleted G11. If it isn't bad enough for G11, its worth discussing at AFD. Monty845 05:12, 24 March 2019 (UTC)
- Oppose per Monty845. It also doesn't distinguish articles created for pay from articles created by someone who was paid to create other articles. Thryduulf (talk) 09:37, 24 March 2019 (UTC)
- Oppose per WP:CREEP and WP:FOC. Andrew D. (talk) 12:03, 25 March 2019 (UTC)
Alternative proposal 6
I think that this would be a good reason for speedy deletion, but I think that G5 is the wrong criterion. I think that we should, in stead, add this to G11 (spam), which is directly related to the issue of undisclosed payed editing. עוד מישהו Od Mishehu 15:10, 24 March 2019 (UTC)
- @Od Mishehu: Please could you be specific about the wording you are proposing as there are several different formulations above "this" could refer to (and the exact wording matters at speedy deletion). However, unless your proposal addresses the reasons for opposition that are unrelated to it being part of G5 (which I think is most of them) it is very unlikely to gain consensus. Thryduulf (talk) 19:11, 24 March 2019 (UTC)
- Oppose per WP:CREEP. There are too many half-baked proposals now and this is vexatious. No deal is better than a bad deal, eh? Andrew D. (talk) 12:03, 25 March 2019 (UTC)
Update P2 proposed
Currently P2 says "Any portal based on a topic for which there is only a stub header article or fewer than three non-stub articles detailing subject matter that would be appropriate to present under the title of that portal."
However even WikiProject Portals has 20 articles as the minimum to support a portal, something that many other editors think is too low. We are now finding many portals that lack 20 articles created by various users. To save a lot of wasted MfD time let's move the 3 to 20 in this criteria. Legacypac (talk) 02:01, 16 March 2019 (UTC)
- Oppose, because speedy deletion is intended for content that blatantly violates CSD standards, and as presently worded, the criteria is sufficient. For example, a portal with nineteen selected articles could then be speedily deleted simply because it hasn't been updated to include new content. Makes it too easy to quickly throw away the work of other editors from simply counting articles, essentially qualifying deletion via bureaucratic bean counting. Per this "nineteen article" example, such and similar portal examples would be better discussed and assessed on a case-by-case basis at WP:MFD. North America1000 03:54, 16 March 2019 (UTC)
- Support, but prefer a threshold of 10 non-stub articles pending a wider consensus. I have some sympathy with North's "nineteen article" example, and setting the limit at half that gives plenty of room for the occasional lapse, while still satisfying Legacypac's well-justified concern that too much effort is being wasted on portals with 3–10 articles.
- However, that "nineteen article" example is problematic. The minimum of 20 is a target set by the create-squillions-of-pointless-portals brigade at WP:WPPORT, and I believe that community consensus would set a radically higher threshold. Insofar as that threshold of 20 is upheld, it should be regarded as the absolute rock-bottom bare minimum for a portal to survive, so editors should be aware that creating a portal without a significantly higher number of articles is placing the portal at real risk of deletion if the numbers slip.
- So while, I'd refer ten, I'd support a threshold of 20 if other editors back it ... but either way, this proposal is just an intermediate step to ease the burden of cleaning up the flood of portalspam. It should all be revisited as part of a wider discussion on what size and number and type of portals the community will support. --BrownHairedGirl (talk) • (contribs) 04:32, 16 March 2019 (UTC)
- Oppose per NorthAmerica. This would be a good reason to discuss the portal at MfD but the higher the number the greater the chance that there will be disagreement about whether an article does fall within a topic area or not so it needs discussion. Given that there is no draft space for portals and editors who dislike the idea of portals so much that they are desperately trying everything they can think of to get as many of them deleted as they humanly can, it is very important that we take extra care not to speedy delete something that would be better merged for example. Thryduulf (talk) 11:44, 16 March 2019 (UTC)
- This isn't unreasonable. It is, however, premature. We should first find a wider consensus on how many articles are truly needed to support a portal - and I strongly suspect it will be multiple orders of magnitude higher than what the WPPORT echo chamber would like - and only then set a CSD threshold at some percentage of that. —Cryptic 12:15, 16 March 2019 (UTC)
- Not yet. The breadth threshold for portals is already being debated in several forums. I am disappointed that those discussions have not yet reached a conclusion, but producing yet another competing standard would not help. Certes (talk) 12:42, 16 March 2019 (UTC)
- 20 is the number the Portal fans set a long time ago. Precious few topics on Wikipedia are notable enough to have a page yet are unreleated to less than four other pages. Therefore P2 is pretty much pointless as written. Legacypac (talk) 18:44, 16 March 2019 (UTC)
- @Legacypac, CoolSkittle, BrownHairedGirl, Pythoncoder, and Cryptic: As Espresso Addict notes below, it is not going to be possible for a single admin to objectively determine whether a threshold above a small number (certainly less than 10) is met. This means it is inherently unsuitable for speedy deletion. It would be a good metric for a deletion discussion though, but per Certes it is important that if we have a threshold that we have only one, so it should be discussed at a single appropriate venue (which is not this discussion). Thryduulf (talk) 09:52, 17 March 2019 (UTC)
- @Thryduulf, I think you misunderstand the issues here.
- This discussion is about criteria for speedy deletion. It is quite common (even routine) to have content guidelines which set thresholds higher than the speedy deletion criteria, so there is no reason why there should be only one criterion. All that matters is that the speedy criterion extends no higher than the lower limit of the general guideline. --BrownHairedGirl (talk) • (contribs) 10:05, 17 March 2019 (UTC)
- (edit conflict) @BrownHairedGirl: I agree that two thresholds are common, but that's not my point. I was making two spearate points - First that many of the thresholds proposed here for speedy deletion (10, 20, etc) cannot be assessed objectively by a single admin and so cannot work as a speedy deletion criterion. Secondly I was pointing out that discussions of a non-CSD threshold (which some commenters seem to be unclear is different) should be discussed elsewhere. Thryduulf (talk) 10:12, 17 March 2019 (UTC)
- Paging through the list of portals-actually-deleted-as-P2 that I generated below, it's clear that even "3" has been entirely ignored in practice. —Cryptic 10:10, 17 March 2019 (UTC)
- @Cryptic: Do you mean that portals with more than three non-stub articles are being deleted or that ones with fewer are not being? If the criterion is being routinely abused then that deserves a separate discussion about how to resolve that - e.g. is it only one admin or multiple admins? If it's not being used, then we need to look at why (again probably a separate discussion). Thryduulf (talk) 10:16, 17 March 2019 (UTC)
- I meant the former, but the latter's probably true too. There hasn't been a year with more than 10 correct P2 speedies since 2007. Its original purpose - to prevent Portal:My meaningless little company and/or website - was obsolesced by the introduction of criterion G11. —Cryptic 10:38, 17 March 2019 (UTC)
- @Cryptic: Do you mean that portals with more than three non-stub articles are being deleted or that ones with fewer are not being? If the criterion is being routinely abused then that deserves a separate discussion about how to resolve that - e.g. is it only one admin or multiple admins? If it's not being used, then we need to look at why (again probably a separate discussion). Thryduulf (talk) 10:16, 17 March 2019 (UTC)
- @Legacypac, CoolSkittle, BrownHairedGirl, Pythoncoder, and Cryptic: As Espresso Addict notes below, it is not going to be possible for a single admin to objectively determine whether a threshold above a small number (certainly less than 10) is met. This means it is inherently unsuitable for speedy deletion. It would be a good metric for a deletion discussion though, but per Certes it is important that if we have a threshold that we have only one, so it should be discussed at a single appropriate venue (which is not this discussion). Thryduulf (talk) 09:52, 17 March 2019 (UTC)
- 20 is the number the Portal fans set a long time ago. Precious few topics on Wikipedia are notable enough to have a page yet are unreleated to less than four other pages. Therefore P2 is pretty much pointless as written. Legacypac (talk) 18:44, 16 March 2019 (UTC)
- Support with a threshold of 10 for now per BrownHairedGirl. CoolSkittle (talk) 14:52, 16 March 2019 (UTC)
- Support 10 or 20 per BrownHairedGirl — if 20 is what the biggest portal fans (and thus people who want the lowest number of portals deleted) on this site want, it should definitely be the minimum standard for portals. This would help get rid of some of the useless nanoportals that aren't eligible for X3. — pythoncoder (talk | contribs) 19:27, 16 March 2019 (UTC)
- Comment. I think this could be quite difficult for an admin to quickly assess in practice. Take, for example, the late unlamented Portal:Spaghetti. In addition to an adequate main article, there's Spaghetti alla chitarra, ~7 clear non-stubs under the subcat Dishes plus ~3 other borderline stub/start. I was surprised not to find that English school meal staple 'spag bog', but it's under Bolognese sauce, which apparently traditionally should be cooked with flat pasta, but does talk about the spaghetti dish. Also Tomato sauce, which has a classic spaghetti dish highlighted in the infobox. Also spaghetti sandwich, mysteriously not categorised. There could well be many other uncategorised articles that relate to spaghetti that might be findable by search, or just by meandering through the main article & the other found articles. Then you've got peripheral (but categorised) content such as Spaghetti-tree hoax, Flying Spaghetti Monster & 3 related articles, plus 16 pages of metaphors relating to spaghetti. There are ingredients (well, flour). There might well be books on spaghetti, well-known manufacturers, chefs, restaurants &c. Then you've got the notion that articles in Italian might be translated or mined to expand the stubs. Does the admin press delete or not? Unless >=9/10 admins would agree, it isn't going to be viable as a speedy category. Espresso Addict (talk) 07:15, 17 March 2019 (UTC)
- Can anyone tell how many deletions happen under P2 in the last X months or years? Legacypac (talk) 07:32, 17 March 2019 (UTC)
- Not all that many. There's only been 449 deletions, ever, in the portal namespace that mentioned "p2" or "P2" in the log comment (quarry:query/34444). Maybe a handful more if you account for admins who don't mention the code in their deletion comment, but those are rare these days. A bunch less if you filter out the false positives (Portal:San Francisco Bay Area/Cities and Counties Intro mentions P2 only to say that's being declined as a reason, for example) and the plainly, blatantly invalid ones (Portal:Laser, Portal:Disco, Portal:Bon Jovi, and about 60 subpages of Portal:Paleontology jumped out at me). —Cryptic 09:41, 17 March 2019 (UTC)
- The current P2 is essentially useless and unused, and could be abolished with little harm. In more than ten years of CSD patrol, I have not used it once. I can, however, imagine a good portal that navigates through only 10 pages and a gazillion images. While I support a very low bar for deletion of autogenerated crap portals, that is a different issue than the number of articles covered by a portal. —Kusma (t·c) 10:20, 17 March 2019 (UTC)
- Support. I like the effort to make P2 actually useable and would prefer an expansion of P2 over the X3 option that currently floating around. I'm also fine with 10—I could go either way. -- Tavix (talk) 21:27, 17 March 2019 (UTC)
- Support 10, 20 or any number up to 100. I'd also support expanding P2 to apply to semi-automatically created portals with fewer than 100 pages, or portals with fewer than 100 notable topics under their scope, if either suggestion would garner more support. This threshold is still tiny and the vast majority of useless portals we have would not fail it. Given how incredibly low readership is for portals, and given that Wikipedia is written for readers, there's no point in us hosting portals on topics with so few pages under their scope. — Bilorv (he/him) (talk) 22:32, 17 March 2019 (UTC)
- @Bilory: How on earth do you propose to quickly, reliably and objectively (all required) determine whether there are 10 "notable topics" (definition please) within a portal's scope? For example would Trolley pole, Tunnelling shield and The Institution of Railway Signal Engineers be within the scope of a Portal:Railway infrastructure? Thryduulf (talk) 14:40, 18 March 2019 (UTC)
- No one has been arguing it is impossible to quickly, reliably and objectively deturmine if 3 related articles exist. 10 or 20 is just a slightly higher number. If you believe the entire CSD is unworkable and has been that way for years make that point instead. This CSD should be used to clear the most crappy narrow portals. If there are not 20 bird species in a family for the Portal, Portal fails the test. If a writer has a head article article and articles on two books, writer fails the test. It's not that hard. Most of us can count to 10 or 20 and the Admin who feels incapable of making the assessment can leave it for an Admin who is more capable. Legacypac (talk) 15:33, 18 March 2019 (UTC)
Example: new Portal:Skunks which stinks. There are only 12 kinds of skunks so would fail P2=20. The 12 species are linked from the nav box and the portal and easy to count. The portal, complete with a contextless unlabeled chemical bond chart, adds nothing to skunks, it is just a detour from the useful article. Even more clear cut, a portal on Western spotted skunk would have zero other articles and fail P2 as written now. Legacypac (talk) 16:10, 18 March 2019 (UTC)
- Indeed, it is almost always going to be easy to find (or not find) three articles that objectively fall within a portal's scope, however this is not the case infinitely and so a cut-off has to be drawn somewhere at a point that it will be possible to determine objectively in all cases. Taking your skunks example, there are 12 types of skunk, but there are also list of fictional skunks, Enchantimals, Pepé Le Pew, Pogo (comic strip), Punky Skunk, Stinkor, Skunks as pets, Mephitis (genus) and Brachyprotoma obtusata, meaning there are 21 articles at least. Now you might argue that some of those shouldn't count - but that's the point, e.g. whether fictional skunks count as within the scope of a portal on Skunks is subjective (meaning the criteria fails the objectivity requirement). In this case though I would suggest that Portal:Skunks be merged into a Portal:Mephitidae or Portal:Musteloidea (meaning the criteria fails the incontestable requirement). Thryduulf (talk) 17:52, 18 March 2019 (UTC)
- This, I think, is the crux of the thing. Once you get past a limit of a few (say, for example, three), it becomes time consuming and subjective, which makes for a bad CSD criterion. Once something becomes slow, subjective, or complicated enough that the typical sysop might not be guaranteed to do the same thing every time, it's better to use a more deliberative process like XfD where these cases can be discussed. The problem here is not with P2, but rather with the recent spate of portal creation. We can deal with that without complicating P2. ~ Amory (u • t • c) 19:39, 18 March 2019 (UTC)
- The portal links to Category:Skunks and Category:Fictional skunks, which in total contain 31 articles. This is an objective criterion that could be used. And if a category doesn't exist for a topic then we certainly don't need a portal on it. — Bilorv (he/him) (talk) 20:21, 18 March 2019 (UTC)
- And Portal:Skunks shows no evidence of including fictional skunks in the scope so that is a big stretch. The lede article does not mention fictional skunks. Legacypac (talk) 21:35, 18 March 2019 (UTC)
- You're making my point for me: There is no consensus whether the portal does or should include fictional skunks, therefore it is not objectively and uncontestably determinable how many articles are within the scope and so it is unsuitable for speedy deletion. Thryduulf (talk) 11:16, 21 March 2019 (UTC)
- And Portal:Skunks shows no evidence of including fictional skunks in the scope so that is a big stretch. The lede article does not mention fictional skunks. Legacypac (talk) 21:35, 18 March 2019 (UTC)
- Indeed, it is almost always going to be easy to find (or not find) three articles that objectively fall within a portal's scope, however this is not the case infinitely and so a cut-off has to be drawn somewhere at a point that it will be possible to determine objectively in all cases. Taking your skunks example, there are 12 types of skunk, but there are also list of fictional skunks, Enchantimals, Pepé Le Pew, Pogo (comic strip), Punky Skunk, Stinkor, Skunks as pets, Mephitis (genus) and Brachyprotoma obtusata, meaning there are 21 articles at least. Now you might argue that some of those shouldn't count - but that's the point, e.g. whether fictional skunks count as within the scope of a portal on Skunks is subjective (meaning the criteria fails the objectivity requirement). In this case though I would suggest that Portal:Skunks be merged into a Portal:Mephitidae or Portal:Musteloidea (meaning the criteria fails the incontestable requirement). Thryduulf (talk) 17:52, 18 March 2019 (UTC)
- Support portals usually link to at least one category to establish the scope of the portal, so it should usually be fairly unambiguous how many articles are in scope. If there is no category then there aren't likely to be many articles in scope. The ambiguity argument doesn't seem to have prevented P2 from working up to now either. Three articles is far too small a scope for an effective portal and is below even the standard used by the relevant Wikiproject. Hut 8.5 21:50, 18 March 2019 (UTC)
- Sure, but the CSD criteria aren't designed to catch everything for deletion, only the most obvious/egregious/clear-cut cases. A3 or A7, for example, don't set a low a bar for an effective article because they don't define what makes a good article, just an obvious example of something that doesn't. Similarly, a portal relating to 4 or 11 or 21 articles may well be deleted just as so many articles that do no qualify for A3 or A7. Setting an overly strict standard to ensure the community has their say doesn't mean the criteria are ineffective, it's by design. At any rate, I'm open to the idea that three is too few to be make a valuable criterion, but I continue to think the real issue will be slowing down reviewers. ~ Amory (u • t • c) 23:58, 18 March 2019 (UTC)
- Not only are categories useful for gauging the scope of portals, they are required portal items. If there is no category, then the portal shouldn't have been created in the first place because it can't contain all the items required for a portal. – Finnusertop (talk ⋅ contribs) 16:15, 19 March 2019 (UTC)
- Sure, but the CSD criteria aren't designed to catch everything for deletion, only the most obvious/egregious/clear-cut cases. A3 or A7, for example, don't set a low a bar for an effective article because they don't define what makes a good article, just an obvious example of something that doesn't. Similarly, a portal relating to 4 or 11 or 21 articles may well be deleted just as so many articles that do no qualify for A3 or A7. Setting an overly strict standard to ensure the community has their say doesn't mean the criteria are ineffective, it's by design. At any rate, I'm open to the idea that three is too few to be make a valuable criterion, but I continue to think the real issue will be slowing down reviewers. ~ Amory (u • t • c) 23:58, 18 March 2019 (UTC)
- User:Amorymeltzer what reviewers? Do you mean portal creators? I'm happy with a low bar but obviously 1:1 Portal:Article is too low. 1:3 is also effectively meaningless. Many of the past P2 tags look like blank pages or tests so could have been speedy deleted G2 or other ways. Legacypac (talk) 00:39, 19 March 2019 (UTC)
- Sorry; I meant reviewers as in people reviewing the CSD itself aka sysops. ~ Amory (u • t • c) 00:57, 19 March 2019 (UTC)
- Oppose. Per Northamerica1000 and Thryduulf. I concur that increasing the availability of speedy deletion will be at the expense of the alternatives to deletion that can be considered at MfD. --Bsherr (talk) 15:22, 20 March 2019 (UTC)
- Oppose, but expand PROD to include portals so that these uncontroversial cases of deletion can be made via the PROD process. feminist (talk) 01:38, 21 March 2019 (UTC)
- Oppose per Northamerica1000, et al. Also, there is no evidence that MfD is failing to get the job done, including with mass nominations. Extant process is sufficient. — SMcCandlish ☏ ¢ 😼 16:08, 21 March 2019 (UTC)
- Oppose per the group. 3 is a reasonable objective threshold. There is a significant range of grey area above that level in which portals are unlikely to be kept after a discussion to evaluate their merits, but each one should be evaluated in a discussion, not deleted or not deleted under the selective evaluation of individual administrators. Ivanvector (Talk/Edits) 16:17, 21 March 2019 (UTC)
- Oppose, three is reasonable. I also want to point out to all commenters here that Legacypac is also simultaneously trying to delete portals with well over 20 articles inside here: Wikipedia:Miscellany for deletion/Portal:Alhambra, California. These two discussions are inherently related, and should have been disclosed together. ɱ (talk) 17:13, 26 March 2019 (UTC)
- Says the editor who claims portal guidelines don't apply to his portal and he does not need to follow them. Follow his link. Legacypac (talk) 21:11, 26 March 2019 (UTC)
- Oppose per User:Ɱ. Also, there is no need to speedy delete a portal with less than 20 articles. Making the threshold that high would lead to speedy deletion for portals currently under construction. 3 to 20 is a good range for grey area that should not result in simple speedy deletion. Speedy deletion is not means to address grey area portals. PointsofNoReturn (talk) 19:16, 26 March 2019 (UTC)
- No it would not. The number of articles within the scope exists before someone even starts a portal. If a portal has only 5 or 10 articles within the scope it should not be created or exist. See WP:POG. This is not a gray area. Legacypac (talk) 22:34, 30 March 2019 (UTC)
- While it is probable that such portals will be deleted it is not certain as it is not always clear exactly which articles are in scope and which are not - see for example the difference of opinion regarding whether fictional badgers are within scope of the badgers portal - and adjusting the scope is an alternative to deletion that will be appropriate in some cases - e.g. a portal about Brian May is unlikely to have enough articles in scope but a portal about Queen (band) is more likely to be, but then are articles related to May's scientific career within scope? These are questions that cannot be answered by a single administrator acting alone. Thryduulf (talk) 13:34, 31 March 2019 (UTC)
- No it would not. The number of articles within the scope exists before someone even starts a portal. If a portal has only 5 or 10 articles within the scope it should not be created or exist. See WP:POG. This is not a gray area. Legacypac (talk) 22:34, 30 March 2019 (UTC)
- Oppose per Northamerica1000 and others above. Could lead to unwarranted deletions. There are valid alternatives. --Crystallizedcarbon (talk) 20:25, 26 March 2019 (UTC)
- Oppose An "underpopulated" portal is akin to a stub article. Everything has to start somewhere, and the requirement that everything has to be a fully complete, finished, product before it can be published is completely opposite to how Wikipedia works. WaggersTALK 11:27, 2 April 2019 (UTC)
- Comment the discussion at Wikipedia:Miscellany for deletion/Portal:Andy Gibb, a portal with 19 articles at the time of nomination, has attracted good-faith recommendations to keep, further demonstrating that expanding P2 to cover portals with fewer than 20 articles fails the uncontestable requirement for speedy deletion criteria. Thryduulf (talk) 12:03, 4 April 2019 (UTC)
Mentioning the requirements for new criteria in an edit notice?
There are plenty of examples of people on this page proposing new criteria or commenting on proposals for new criteria who seem unfamiliar with the requirements (Objective, Uncontestable, Frequent, Non-redundant) detailed in the page header. To help reduce this (especially for those who arrive via a link to a section), how about adding a slimmed down version of the header in an edit notice, linking to the header for full details? Perhaps something like:
Although it might be possible to condense it still further - the point is to alert not overwhelm, and other improvements are almost certainly possible as well. Thryduulf (talk) 11:36, 21 March 2019 (UTC)
- Suggest Speedy closing proposals that clearly fail the new criterion criteria, and Speedy close CSD proposals that are not even accepted reasons for deletion. —SmokeyJoe (talk) 11:56, 21 March 2019 (UTC)
- I wouldn't object to that in general, although in some cases they only need a slight tweak and in others speedy closing will be a lot of drama (especially I think anything related to UPE). I don't think that's incompatible with my suggestion though. Thryduulf (talk) 12:23, 21 March 2019 (UTC)
It's worth mentioning here that I've just created a new shortcut to the header listing the the requirements: WP:NEWCSD. Thryduulf (talk) 13:45, 21 March 2019 (UTC)
- Fails #4, redundant to the header.I don't think it's a matter of not noticing the new-criteria criteria. (And if it is, the first step is to remove the irrelevancies distracting from it - we absolutely don't need a "don't delete WP:CSD, we copied some text out of it four years ago" box; we don't need the "be polite and sign your posts" box; and the archive lists doesn't need to be full-width.) What we've been seeing lately is A) legitimate disagreement over whether criteria meet NEWCSD, and B) a suicidal headlong rush to vote on everything. B exacerbates A, because there isn't time to hammer out the obvious amendments that make a proposal meet NEWCSD before some dolt's plastered it all over CENT and started a formal RFC, and then we suddenly have four competing proposals for people to vote against without even fully reading. —Cryptic 14:28, 21 March 2019 (UTC)
- I should hope I'm not a dolt because you're making that statement as a general comment. ;) --Izno (talk) 14:51, 21 March 2019 (UTC)
- The purpose is very much to complement the header, for people who don't see it and/or read it, that's why it explicitly summarises it and links to it for full details. It's very much not redundant to it. The excessive number of proposals being made (sometimes on the wrong page - e.g. the X3 proposal at AN) without understanding the requirements is why we need to make it more in-your-face. And just like when evaluating a page against the criteria, if there is good faith disagreement about whether a proposal meets the requirements for a new criterion it doesn't. If people disagree whether something is objective it clearly isn't. If people disagree whether everything that covered by the criterion should be deleted then it is not uncontestable. The "we copied text" box is needed for proper attribution - see WP:CWW. The archives box though could be made less prominent, but I'm not sure it would make that much difference. Thryduulf (talk) 15:56, 21 March 2019 (UTC)
- Strongly support making such an editnotice. The vast majority of proposals at this page fail at least one of these obvious criteria, and it's a drain on editorial time. One of those "RtFM" things. — SMcCandlish ☏ ¢ 😼 16:11, 21 March 2019 (UTC)
- Support - clearly an improvement. If rule #4 prevents us from implementing this obviously helpful advice, ignore the rule. Ivanvector (Talk/Edits) 16:19, 21 March 2019 (UTC)
Given the comments above I have added an edit notice (Template:Editnotices/Page/Wikipedia talk:Criteria for speedy deletion). Please tweak it. Thryduulf (talk) 13:52, 25 March 2019 (UTC)
Expand A3 to all namespaces except draftspace (A3 -> G15)
I believe A3 can apply to things like a template that looks like this:
Obviously, this would be deleted per G1, but if it weren't nonsense, it couldn't be deleted, as A3 doesn't apply to templates at the moment. I also have a potential template here. InvalidOStalk 17:41, 22 March 2019 (UTC)
- Support An empty page is still an empty page but I don't expect it will be used much though since new users rarely create pages in other namespaces. I'm assuming you also exclude user space. Crouch, Swale (talk) 17:44, 22 March 2019 (UTC)
- A3 should apply to all name spaces except Userspace. Mostly this would simplify how we deal with Drafts. We see a lot of pages in Draft, including many but not all pages found in Category:AfC submissions declined as a test, Category:AfC submissions declined as lacking context, and Category:AfC submissions declined as blank that meet A3. Usually a G2 test CSD is accepted by Admins for the blank pages but that is not a fully intuitive application of G2. A3 is clearer and this change would bring consistency. Legacypac (talk) 18:50, 22 March 2019 (UTC)
- Oppose. Patent nonsense is already G1, test pages are already G2 and G7 covers pages blanked by the author, which would seem to cover most of these cases (fails the "non-redundant requirement of WP:NEWCSD). Lack of context is not a reason to delete a draft - this can be added. There is no evidence presented that there is any reason why these drafts need to be deleted before they're eligible for G13 and there is no evidence presented that problematic pages in other namespaces are overwhelming normal deletion processes (fails the "frequent" requirement). Simply expanding A3 to other namespaces would catch a lot pages in user, template, Wikipedia and portal namespaces that have no content of the type the criterion mentions but are nevertheless useful pages - including most noticeboards, reference desks, the help desk TOC templates, navboxes, template testcases, etc, etc, (stupendous failure of the "uncontestable" requirement). Thryduulf (talk) 19:25, 22 March 2019 (UTC)
- Your opinons on CSDs such as X3 show you lack the ability to present factual or logical points around CSDs. This is no better. Legacypac (talk) 19:34, 22 March 2019 (UTC)
- Thank you for the unnecessary ad hominem. Would you now like to address the actual points I made or would you rather stick with personal attacks? Thryduulf (talk) 19:57, 22 March 2019 (UTC)
- Your opinons on CSDs such as X3 show you lack the ability to present factual or logical points around CSDs. This is no better. Legacypac (talk) 19:34, 22 March 2019 (UTC)
- I don't see how WP:A3 applies really to that template (if it were less nonsense it could be a valid navbox template); I'd think WP:G3 would cover it. (WP:G1 may not even really apply since there are portions of the template that are understandable) Galobtter (pingó mió) 20:01, 22 March 2019 (UTC)
- Oppose - G1 wouldn't apply because that's clearly intended to be a template (exempted per "if you know what it is, G1 doesn't apply") but it is G3 obvious vandalism, and if not then it's G6 "this is obviously going to be deleted"/WP:IAR, and interestingly this might have been a use case for the long-deprecated T1. I would also strongly consider blocking the creator and then deleting this per G5, but that's specific to this particular example. Ivanvector (Talk/Edits) 20:07, 22 March 2019 (UTC)
- Oppose. I don't really get it. WP:A3 is for lack of content. The example you give has content. So why would it be covered by a G-A3 criterion? Also, what Thryduulf said. Regards SoWhy 21:35, 22 March 2019 (UTC)
- Oppose I'd have thought many perfectly legitimate article talk pages would fall into the scope of such an A3. Here is one I created: Talk:Dudley Wolfe. Now, I know most people nominate for CSD appropriately but some do not and a couple of administrators delete all CSD nominations they come across on the basis that they'll later restore any that are shown to be wrong. It's quicker like that. This would be disruptive and overwhelmingly unsuitable as a changed CSD criterion. Thincat (talk) 21:49, 22 March 2019 (UTC)
- Oppose - something like this is already covered under existing CSD criteria (G2 or G3 usually, but depending on circumstance other criteria like G5 or G10 would apply) and a lack of content on the page is to be expected for userpages that have been blanked (courtesy or otherwise). —A little blue Bori v^_^v Bori! 22:10, 22 March 2019 (UTC)
- Oppose, far too wide-reaching, and no obvious flood of pages that MfD/TfD can't delete quickly enough. I probably don't need User:Kusma/AJH anymore, but it was a useful page containing only external links that once helped me write an article. Why should pages like that (many of them exist in user space, as a WikiProject subpage, as a Portal subpage, probably in many other places) suddenly become speedy-deletable? If they are not in article space, they do not cause embarrassment and may actually be helpful for writing an encyclopaedia. —Kusma (t·c) 21:17, 24 March 2019 (UTC)
Twinkle now logs user-inputted options and should better handle noms in module space
This is about CSD tagging and not the criteria themselves, but just FYI if attempting a CSD of a Module, Twinkle should now place the tag on the documentation subpage, like is supposed to be done at WP:TfD. Also, the CSD log will now include the user-inputted options, like user for G5 or xfd for G6. Other recent changes here; please let me know if there are any issues! ~ Amory (u • t • c) 15:47, 3 April 2019 (UTC)
Speedy deletion of modules
Hi. There have been a number of times when I've tried to tag modules for deletion. The tag goes on the doc page, and doc page gets deleted, but the actual module isn't. Can an admin please delete Module:User:Xinbenlv bot/msg/inconsistent birthday per WP:CSD#G7. Separately, is there a better way to communicate to admins that, despite the tag being on the documentation page, the module itself is the target of the deletion request? --DannyS712 (talk) 00:59, 6 April 2019 (UTC)
- You can use a custom rationale on most tags. --Izno (talk) 13:36, 6 April 2019 (UTC)
- Ask a friendly admin directly - preferably one who works in module space, so they know how to verify it's not being used. Failing that, to make it clear you're not talking about the doc page itself, you can either enclose the speedy deletion template in <includeonly> tags, or manually categorize it into Category:Candidates for speedy deletion with some explanatory text. —Cryptic 00:38, 8 April 2019 (UTC)
- @Cryptic: okay, thanks --DannyS712 (talk) 00:51, 8 April 2019 (UTC)
G6 for post-merge delete?
- The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.
- Proposal withdrawn Graeme Bartlett (talk) 22:51, 10 April 2019 (UTC)
The Twinkle CSD menu includes G6 for An admin has closed a deletion discussion ... as "delete" but they didn't actually delete the page, but this isn't listed as a use case under WP:G6. I propose adding another bullet point:
- Deleting a page per an XfD close which specified deletion but didn't perform it.
Any objections? -- RoySmith (talk) 14:58, 6 April 2019 (UTC)
- PS, see this discussion for background. -- RoySmith (talk) 15:00, 6 April 2019 (UTC)
- That's already in there; fourth bullet point in the "templates" section of G6:
* {{Db-xfd|fullvotepage=link to closed deletion discussion}} - For pages where a consensus to delete has been previously reached via deletion discussion, but which were not deleted.
- It's also in the main section, though I do suppose it says
... as the result of a consensus at WP:TfD.
I suppose this could be modified to just read "XfD". Primefac (talk) 20:56, 6 April 2019 (UTC)- I believe that's because non-admins can close TfDs as delete, but not other XfDs. ansh666 22:28, 6 April 2019 (UTC)
- The particular case here (as described in the discussion on my talk page), was I closed an AfD as (essentially), Merge, then delete. As is common in merge closes, I left the actual merge for somebody else to execute. In this case, the merge was done by Nthep, who is an admin and was able to delete the page after they were done merging. They used WP:G6 as the reason. But, in theory, the merge could have been done by a non-admin, and then tagging the page for G6 deletion would have made sense.
- "Merge and delete" is almost never a valid outcome, since attribution is needed for a merge. ansh666 17:58, 7 April 2019 (UTC)
- In general, I agree with you. It almost always makes sense to "Merge and redirect". In fact, I think it made sense to do that in this case too, but as the closing admin, my job is to summarize the discussion, not cast a supervote. So, let's for the moment, assume we have a valid "Merge and delete" outcome. What's the best way to implement that? -- RoySmith (talk) 23:54, 7 April 2019 (UTC)
- Move the history to another existing redirect that doesn't have history that needs preserving. There's some alternatives at WP:Merge and delete. —Cryptic 00:17, 8 April 2019 (UTC)
- In general, I agree with you. It almost always makes sense to "Merge and redirect". In fact, I think it made sense to do that in this case too, but as the closing admin, my job is to summarize the discussion, not cast a supervote. So, let's for the moment, assume we have a valid "Merge and delete" outcome. What's the best way to implement that? -- RoySmith (talk) 23:54, 7 April 2019 (UTC)
- "Merge and delete" is almost never a valid outcome, since attribution is needed for a merge. ansh666 17:58, 7 April 2019 (UTC)
- The particular case here (as described in the discussion on my talk page), was I closed an AfD as (essentially), Merge, then delete. As is common in merge closes, I left the actual merge for somebody else to execute. In this case, the merge was done by Nthep, who is an admin and was able to delete the page after they were done merging. They used WP:G6 as the reason. But, in theory, the merge could have been done by a non-admin, and then tagging the page for G6 deletion would have made sense.
- I believe that's because non-admins can close TfDs as delete, but not other XfDs. ansh666 22:28, 6 April 2019 (UTC)
- It's also in the main section, though I do suppose it says
- Oppose. Lazy catchall. If the deletion is authorised by an XfD, use “deleted per [the XfD]”, not per G6. G6 should never be used for pages with a nontrivial history, and G6 post merge sounds like a terrible violation of that. —SmokeyJoe (talk) 21:00, 6 April 2019 (UTC)
- Fix twinkle. Twinkle documentation errors should be fixed, not policy altered to suit twinkle documentation. —SmokeyJoe (talk) 21:02, 6 April 2019 (UTC)
- Two things. First, the list under G6 isn't exclusive, i.e. those are not the only reasons that G6 may be used. Tagging a page under G6 where the XfD for that page was closed as delete is fairly obviously "uncontroversial maintenance", so there isn't a need to add a bullet point. Second, tagging a page as G6 for an AfD result of merge - even if the nominator recommends deletion post-merge, which should frankly never be done for attribution reasons - would no longer be uncontroversial by default, and therefore shouldn't be done. ansh666 22:24, 6 April 2019 (UTC)
- I'm with SmokeyJoe. If you're deleting something because of a deletion discussion, that deletion discussion is tautologically the reason for deletion. It's not a speedy deletion, and shouldn't be labelled as one. It especially shouldn't be labelled as a G6, which is intended for cases where there's both no non-temporary loss of content, and no controversy whatsoever. (Before someone brings it up, for G4, you're speedying it because of similarity to previously-deleted content, not directly because of the previous deletion discussion.) —Cryptic 00:17, 8 April 2019 (UTC)
- I'm with SmokeyJoe and Cryptic. In addition G6 is already the most convoluted and most misused speedy deletion criterion, we should not be adding more things to it, especially something that will encourage inappropriate merge and deletes. Thryduulf (talk) 08:45, 8 April 2019 (UTC)
- It's clear this isn't going anywhere, so I withdraw my proposal. -- RoySmith (talk) 15:25, 10 April 2019 (UTC)