User talk:Smallbones/Proposed commercial editing policy

I would just make it clear that prohibitions imposed on firms also apply to the freelancers who work for them. A freelancer can't get away with paid editing for a firm just be saying that they are not an employee of the firm in the conventional sense. BD2412 T 02:49, 13 July 2020 (UTC)

• • will do. Smallbones_(smalltalk) 13:14, 13 July 2020 (UTC)
• As it stands, someone who works in sales, or company law dept etc, as their day job is totally barred from editing Wikipedia articles , eg about sport, ancient history, local church etc. There needs to be some way round that, unless we want to class a large number of people as spawn of the devil. PamD 05:55, 13 July 2020 (UTC)
  • I've added that commercial editors may "register an alternative non-commercial account, per legitalt-Designated roles. Both accounts must declare their connection and link to each other. The non-commercial account may not, in any way, contribute to any commercial editing.

AfC

This includes a specific prohibition against AfC? Except that's the route we want them to take for article creation, so it confused me.

We also need to enable the help desks (TH, HD, AFCHD, etc) since though they sometimes cause irritation there, it also mitigates the problems. I suspect we also can't prohibit a number of other pages without causing more issues than we resolve (AN, ANI etc) Nosebagbear (talk) 08:15, 13 July 2020 (UTC)

@Nosebagbear: I'd personally like to prohibit (almost?) all commercial editing at AFC, though there may still be some "non-commercial paid" editors who could submit a draft there. I believe there are many editors who agree with me, e.g. somebody at WP:AN who said 90% of the paid editing requests are "crap", and others who complain that even after they are turned down, the requests keep on coming back multiple times. The problem IMHO is that commercial editors are completely irretrievably biased when it comes to notability. I'd like the notability decision to be made by an experienced editor who will take the time to write the article themselves. The writing is also very bad there usually, so when a real professional paid editor comes through with a slickly written but biased draft, some reviewers will just accept it very quickly, without "going underneath the hood" and seeing what the article is really doing, e.g. not looking for the missing pieces. It also seems to me that nobody ends up taking full responsibility for an accepted article - neither the commercial editor who says "it was approved by an independent editor", or the reviewer who might downplay their role - they just reviewed the article, they didn't write it. Frankly, I think both the author and the reviewer need to take 100% responsibility for publishing the article.

There's an analogy that has been driving much of what I've written on the User page that hasn't come through so well. An article written by a commercial editor is essentially just a press release or a page that would have better been written on the company's website. In both cases the "press release" is written by somebody paid by the company and expresses the company's POV. It is probably original research - we only have the company's word that the "press release" is true. Both forms are trying to convince the news media (or Wikipedia) that they (we) should publish the facts as the company sees them. But the AfC submission is given an advantage over a real press release or company webpage. The Wikipedia commercial editor gets to lobby on the talk page to keep the company's version intact or to modify other editor's writing. The company doesn't even usually admit authorship so we can't even quote them. All-in-all an AfC submission from a commercial editor is just a press release that we've awarded super-powers. By limiting commercial editors to posting links to press releases, company websites, or even other sources, we take those super-powers away. It turns out they are just press releases and should be treated by our own editors as such.

So is there a way to let commercial editors make AfC submissions from their own talk pages? Perhaps an edit request from their own talk page linking to a press release or company website (with a CC-BY SA license attatched to that page) and saying, "Would you please submit this to AfC?" It might seem like just an extra step in an intricate song and dance, but at least we'd be able (required) to attribute the text, and most editors would easily see what the submission really is - a press release. Smallbones_(smalltalk) 16:51, 13 July 2020 (UTC)

More relevant though is our ultimate end goal - to reduce the problematic content on Wikipedia. Severing legitimate routes just drives even more to just damn it as an option. Most of that would usually be detectable, but would cause an even bigger time drain. AfC was specifically formed for handling reviews for editors we'd cut off from the standard method. I don't believe it should be limited - both on an ethical position and on a pragmatic one. Nosebagbear (talk) 19:30, 13 July 2020 (UTC)

The companies and their commercial editors would still have the option of linking to their press releases or the company's webpages, which is pretty much the way companies have always tried to attract attention from the legitimate press - put out a press release. There's no ethical question there - it's just "if you want to have that information published here, we have to write it up. You can't do it yourself." It is a matter of trust between Wikipedians and our readers. As a practical matter, if people want to avoid disclosure their options would be to lie on the commercial editing disclosure - which should be pretty easy to catch; or they can go through the old, tired routine of creating a sleeper account, qualifying it with enough non-related edits in order to create an article, and then coming up with a superficially slick article from a brand new editor. That's also pretty easy to catch. All we need now is a quick way to enforce the rules, which I haven't written up yet, but which will be based on MER-C's work. We don't owe these guys, anything if they don't want to follow our rules. Smallbones_(smalltalk) 02:20, 14 July 2020 (UTC)

User:MER-C/Paid2019

I drafted something similar last year but did not have the time to follow it through: User:MER-C/Paid2019. Three amendments are proposed:

• Disclosure is necessary but not sufficient
• No commercial editing (the definition is more concise than this proposal)
• No content appraisal.

(1) is already accepted, but need to be explicitly stated. MER-C 10:03, 13 July 2020 (UTC)

Thanks. There is a lot there, some of which is already included here. I'll include more especially under enforcement. Smallbones_(smalltalk) 13:14, 13 July 2020 (UTC)

Multiple notification

A person editing from a *(commercial) account can be assumed to be editing for pay at all times in article or talk space. Why is it necessary or desirable for them to declare this rather obvious status in any other way? Cheers, · · · Peter Southwood ^(talk): 14:06, 13 July 2020 (UTC)

They still need to disclose who is paying. A commercial account can take on multiple jobs funded by different parties. MER-C 15:07, 13 July 2020 (UTC)

MER-C, Does that actually happen in real life? The economics are that accounts are free, and more of a liability than an asset. All that happens if you keep an account for a substantial amount of time is you increase the odds of enough evidence accumulating to sink you at SPI or some other forum. So, any paid editor with any brains at all makes a new account for every job they take. They would be stupid not to, given our rules and our absurd sense of AGF. -- RoySmith (talk) 19:44, 13 July 2020 (UTC)

The few paid editors who claim to follow our paid editing rules certainly disclose their clients and employers. If others don't want to do the very minimum, disclose per the Terms of Use, what else can we do with them, other than ban/block them and their clients and employers? Let's make it easy to do that. Smallbones_(smalltalk) 02:27, 14 July 2020 (UTC)

Contractors

As it is, "contractors of these firms" that are editing Wikipedia or hire firms to edit Wikipedia casts a really wide net. Perhaps that can also be limited with a similar clause as for employees ("employees of the legal, public relations, marketing or sales departments") to focus on contractors who are actually editing Wikipedia as part of their work. isaacl (talk) 15:20, 13 July 2020 (UTC)

That's probably correct. I'm afraid I'm getting too wordy and at the end of deciding what the content should be, I should probably invite in a good copyeditor. Are you interested? Smallbones_(smalltalk) 16:54, 13 July 2020 (UTC)

I can do it, though you and I often don't agree on how to word things, and we have different takes on what needs to be said regarding commercial editing (I tend to think that when existing policy already covers a scenario, wording specific to commercial editing is not needed), so not sure if it is the best fit. isaacl (talk) 19:58, 13 July 2020 (UTC)

Government agencies

I don't think it's a good idea to try to distinguish government agencies promoting culture, education, or health. I don't feel confident that the corresponding agencies of all the governments of the world will be editing Wikipedia benignly. (I don't think I trust any government tourism agency, for instance, and am doubtful about cultural agencies, too.) isaacl (talk) 15:43, 13 July 2020 (UTC)

I'd rather not restrict, e.g. the Smithsonian, the British Museum or either National Gallery, or the National Archives or the Center for Disease Control or local libraries. Figure out how to exclude these folks and let me know. Smallbones_(smalltalk) 17:28, 13 July 2020 (UTC)

I think museums can get covered under a museums exception (personally I have qualms about that, given the wide scope of museums out there, but the English Wikipedia discussions I have seen generally haven't expressed the same concerns). I'm not as clear about giving an exception to local libraries; that covers a lot of employees (including part-time staff). Some country's archives or disease control agencies may be dedicated to spreading accurate, neutral information, but it's not clear to me that all of them are, or that all of them are sufficiently free of political influence when it comes to outreach initiatives. isaacl (talk) 20:16, 13 July 2020 (UTC)

Well, this or any other attempt to further restrict paid/commercial editing is not going to fly unless we include a GLAM exemption. The commercial editing exemption for libraries would only be for librarians writing about the libraries themselves. They wouldn't need an exemption if they were writing about something else. Librarians are natural allies of Wikipedians. If the choice is getting rid of many commercial editors and having a few more puffy articles about libraries, vs. keeping the paid editors, I'll take a few puffy articles about libraries. I can only imagine what all those exempt librarians will write about: "great restrooms!" "open 363 days each year!" "wonderful local history collections!" Smallbones_(smalltalk) 02:42, 14 July 2020 (UTC)

I don't see galleries, libraries, archives, and museums as synonymous with government agencies promoting culture, education, or health. I think an exemption crafted around official initiatives from these institutions would be community-supported. I'm not as clear that exempting all employees would be accepted. Nonetheless, my larger point is about government agencies; I think political considerations make an exemption less desirable. isaacl (talk) 03:45, 14 July 2020 (UTC)

Also note as currently written, you've included all employees from government agencies as commercial editors; there's no restriction based on what they're writing about. isaacl (talk) 03:49, 14 July 2020 (UTC)

Requirements on commercial editors

I'm not clear on the need or practicality of having a new commercial editor declare who a previous commercial editor was. There may be many; would they be required to ask their employers for all of the previous Wikipedia accounts and list them all? I get that we can say it's their problem and they can't edit Wikipedia if they can't comply, but it feels unduly burdensome to me. Perhaps it would be better to create a registry sorted by origin and let new commercial editors update the list with their account. That way we'd have a single, consolidated list to trace commercial editing behaviour (in theory, anyway).

Regarding a prohibition on hiring someone if a previous hire was blocked: it seems odd to single out indirect editing as being forbidden but not direct editing. Perhaps it would simpler to ban the editor, which would encompass both? isaacl (talk) 15:38, 13 July 2020 (UTC)

a) my intention is only that the last commercial editor needs to be named, so that only one at a time can edit (other wise it's too much like sock- or meat-puppeting). I do want to make sure though that there is no switching every 2 or 3 days. i.e. no tag-teaming. Smallbones_(smalltalk) 17:14, 13 July 2020 (UTC)

b)I intend that all of these be blocked or banned: the on-Wiki commercial editor. the paid editing firm (e.g. Wikiprofessionals), and their client (e.g. xyz Inc.). To do this both firms need to be blocked from hiring future commercial editors.

So the net is - if the commercial editing that your firm has been involved in breaks the rules, you can't get around any block just by hiring a new commercial editor or commercial editing firm.

Believe it or not, I've been involved in a situation exactly like this. Multiple paid editors, some declared, some not. Ultimately they get blocked for sockpuppeting, then about 6 months later, a new paid editor shows up. I just gave up at that point, and you know I don't give up easily.

BTW, there is a very easy way around these blocks/bans that might satisfy almost everybody. First the editor or the firms declare all their paid editing since June 2014 and then petition for reinstatement. Smallbones_(smalltalk) 17:14, 13 July 2020 (UTC)

Since paid editors are already required to disclose their employer, I think it makes more sense to gather up this information for one origin (that is, one company or corporate hierarchy) onto a single page, rather than have a winding thread of disclosures from editor to editor.

Using meatpuppets and other forms of proxy editing already contravenes the banning policy. So banning editors also disallows the editors from hiring someone to do their bidding. isaacl (talk) 20:03, 13 July 2020 (UTC)

I wish it was that easy. The responses that I got from admins (among all the non-responses) was that the editors were blocked, not the company. Sometimes you just have to spell thse things out. Smallbones_(smalltalk) 02:50, 14 July 2020 (UTC)

I was thinking about the restriction in terms of commercial editors. Regarding firms: you're essentially trying to ban firms in a roundabout manner, by saying they can't hire anyone if they've previously contravened policy. Putting aside that this ignores existing employees, as long as editors are treated as individuals and not firms, it won't really change how admins are handling the situation. I think it would be better to address the problem head-on: obtain a community ban on all employees (direct or indirect) of the firm in question, present and future. isaacl (talk) 03:31, 14 July 2020 (UTC)

To put it another way: we should not be crafting rules to cover scenarios A, B, C, and so forth as it basically gives paid editors the ability to set the rules of the game, encouraging them to think of more scenarios that aren't covered. We should play the game by our rules: once a paid editor breaches the community's trust by making unacceptable edits, ban them, and at some point, if they are employed, ban all employees of the employer. That's what we need to get the community behind: editors not here to write a neutral point of view encyclopedia should be asked to leave.

Regarding this edit: since this proposal is already specifying exactly which pages can be edited and has described permitted actions, perhaps it's easier just to list the specific approved ways of work, and leave everything else as prohibited (the content appraisal processes listed all take place on pages excluded by the proposal). In particular I don't like adding on the "accept or solicit payment" condition for any restriction, since the entirety of the rest of the proposal essentially assumes every edit made under the commercial account is in service of the editor's employment tasks. isaacl (talk) 22:09, 13 July 2020 (UTC)

You may be right on this one, but I'd like to first get everything down that people will accept as far as prohibiting certain behavior. Your comment is mostly about the style of writing it up, which will come later. Smallbones_(smalltalk) 02:50, 14 July 2020 (UTC)

But you've already listed explicitly what is permitted: editing article talk pages to provide suggestions for information, and addressing biography of living person issues. I don't see a point in generating lots of lists with everything else. isaacl (talk) 03:19, 14 July 2020 (UTC)

Coordination

I imagine there will be a big RFC where various proposals regarding strengthening paid editing policy, big and small, are put to the community for consideration in a consolidated RFC. What will be the means for doing this? MER-C 16:18, 13 July 2020 (UTC)

Yes, of course we need to have an RfC, though maybe consolidation can best be done here. First I intend to get a well-thought out, thoroughly checked version of this that I and many other editors can support. Then there needs to be an RfC somewhere - maybe at Village Pump Policy? (any other suggestions?). The one place I do not want to do this is on the talk page (or sub-page) of WP:Paid. I want this to supplement, not a replacement of WP:PAID. I suppose this is important enough to keep open for at least 3 weeks. Smallbones_(smalltalk) 17:24, 13 July 2020 (UTC)

Smallbones, re your reply to me at WP:AN, I'm not sure what the best way is plug in parts of MER-C's enforcement proposals into here. And since both pages are changing, it's not worth duplicating over stuff, since it'll possibly change.

My thoughts: once both proposals are developed they should be compiled into a well structured RfC, with each mini-proposal split into separate sections for approval (so one part failing doesn't cause everything to fail, and so the discussion doesn't become a confusing mess). Best to keep that consolidation separate from either of the two separate proposals, because it'll probably look a fair bit different since your proposals don't entirely overlap and do cover different proposals, and then just copy that consolidation over to the discussion venue once ready. This proposal creates new policies for paid editing, whilst MER-C's creates mainly new enforcement actions for paid editing, promo and spam, and policy restatements/adjustments for commercial editing, so there's not that much overlap, which is a good thing. imo VPP would be best, because these proposals touch multiple areas (COI, paid, promo and general spam).

For the most part, MER-C's proposal provides enforcement actions for the proposals in this discussion, too, since general sanctions provide discretion, with perhaps some exceptions (eg on enforcing the (commercial) username requirement, which would need extra clarification for enforcement). I think this proposal should focus on enforcement of those things not automatically covered by that only, and upon consolidation of both proposals the gaps in both will be addressed. If you both like this approach, I can volunteer to try consolidate into a series of (slightly independent) proposals for a single RfC when you're both done (no obligation to accept)

This one is slightly harder than that to consolidate since it's an entire policy, rather than a series of amendments/proposals. I'm not entirely sure how to butcher it up into a series of proposals yet, but I think that would be required for RfC, because there's a number of controversial points here. Some aspects of this proposal may be approved, others may not. If it isn't butchered up, there's no option to clearly pick-and-choose parts to approve. ProcrastinatingReader (talk) 19:13, 13 July 2020 (UTC)

Non-profits

@DGG: you might look through what you wrote in the draft about non-profits and social movements and conclude that I removed it all. Well, not all of it exactly. I did even add something on non-profits who act as for-profits. But for the most part, in *this proposed policy* I do not want to prohibit non-profits even in some cases from editing on Wikipedia unless they hire a commercial editing firm. While non-profit and for-profit editing by organizations have some similarities, they also have many differences. I think it would confuse many editors to mix the two. I'll encourage you to come up with a different proposed policy on "Editing by non-profit organizations and social movements." But commercial editing is a big enough topic as it is without mixing in something different. Smallbones_(smalltalk) 18:28, 13 July 2020 (UTC)

I continue to disagree on this. Most of my recent despamming has been with non-commercial entities, and I consider them at least as much a hazard. I have recently completed a cleanup of the article on every US law school. Medical schools will be next. Almost all of such articles are written by obvious coi editors; half of it is usually not encyclopedic content. Such organizations--and every college and university in the world care every bit as much for their public reputation and their funding as any ordinary business. I'm lately concerned with small charities, almost none of which have true references providing substantial coverage from third-party independent reliable sources, not press releases or mere announcements. For many of them their prime source of pr is likely to be wp.

there is one principal difference --the coi is so easy to detect & the promotionalism so obvious that they are even easier to work with.

there's another group, individual professionals, eithe MDs/lawyers/etc/, and college faculty. For the first group , almost every article is written by pr firms, andtheiy're notably formulaic. staff. I remove most of them entirely. Forthe second, about half are the school's pr staff, who are usually the last competent of all the coi editors--I written elsewhere about their traits. I go by the publications--to the extent an academic has notable work, I willl try to rescue the article if its is not disgustingly puffy. There are other groups I work on also, mainly fine artists and classical musicians. Almost every article on a LP in thesefields is PR, but I leae mist of them to the people in those fields--i have enough to do with the academics.

What I do not particular care bout is social movements as such.I tend to be very inclusive here, despite the PR. I've felt this way since the beginning--politics and religion can be so hard to judge that I am very pemissibe about notability. As for those closely associated with a living person tying to promote their views, i tend to merge with the person.

Whatever others work on, I'm planning to continue my emphasis. If anything , i consider commercial enterprises less dangerous--any reader would realize they're probably spam. The firms to be really restrictive on are the ones selling alt med etc. And if the real danger is the rings of cheap undeclared eds, they work on anything. Perhaps the solution is to turn it into a multi-part essay, DGG ( talk ) 22:34, 13 July 2020 (UTC)

I've got no problem with what you're working on other than it seems endless - I've got my own seemingly endless task here.

There are ways that this proposal can help in what you're doing.\:

• If the universities, record labels, etc. hire a commercial editing firm to do their dirty work, they are mostly covered here. I'll check whether it's clear here that the same should apply if you hire a PR firm to do your editing.
• Perhaps more important - if admins and others get used to the idea that we can dispose quickly of the obvious problems with commercial editing, perhaps the whole ethos regarding promotion, adverts, etc. will change. I'm not working against you here. Smallbones_(smalltalk) 03:07, 14 July 2020 (UTC)

DGG, this proposal is probably to be merged with User:MER-C/Paid2019 when presented for RfC. That proposal deals more with the spamminess aspect, e.g. medical topics. Take a read through that, I think it'll add the administrative powers to do what you're looking for. Maybe schools etc can be added to the topic list - I've also seen lots of COI spam from schools, especially private schools. ProcrastinatingReader (talk) 10:44, 14 July 2020 (UTC)

a good deal of school spam Is students writing thearticle, using material from the school handbook on website I just remove The worst when I see it -- Of course this happens In other areas also. promotionalism is so widespread In ourout society that people use it without thinking. It can be a problem differentiating spam that comes from a relatively naïve or innocent source and spam that comes from a paid editor. That's why I like to concentrate on removing and preventing spam, rather than trying to decide whom to punish. , DGG ( talk ) 22:16, 14 July 2020 (UTC)

As DGG notes, the focus should be on content, not the user who created it. But, I think I have a somewhat different reason why that's true. The cost of a new account is zero. It's whack-a-mole. Block one account, they just create a new one (or dozens). We invest a lot more in tracking them down then they do in creating them. At most, it's a minor inconvenience.

What really matters is the content, since that's all anybody cares about. If we block 10 socks, but the content stays, the editor gets paid. Removing the content is the important thing. Anybody who's spent more than about two days at WP:AfC can spot a paid article a mile away. All we need now is the gumption to nuke them on sight, rather than playing silly games with COI templates. -- RoySmith (talk) 23:17, 14 July 2020 (UTC)

I agree with RoySmith in general, but I may have less skill in discerning the motives of editors--it still happens at least once a month that I find I have assumed someone to be a paid or other coi editor when they are actually a good faith editor who does not know how to write non-promotionally, (sometimes verified by the direct evidence of someone I know & trust personally). And of course it still happens much more than once a month with a relatively simple straightforward article that I fail to catch that it is actually a contribution by an undeclared paid editor. It is very hard to know about people based only on what they say online. DGG ( talk ) 00:41, 15 July 2020 (UTC)

Reporting violations

Report violations of Wikipedia rules by their employers or clients. If they cannot convince their employers or clients to follow Wikipedia's rules they must report the violations and then resign their position. What exactly has to be reported, and to who? Does this 'resign their position' mean their editing here, or their job with their employer? ProcrastinatingReader (talk) 22:05, 13 July 2020 (UTC)

ProcrastinatingReader, I agree this particular passage is a problem. All we can (should) care about is what they do on the wiki. It's well outside our remit to insist they tattle on their employer or make employment decisions based on our policies. -- RoySmith (talk) 23:03, 13 July 2020 (UTC)

RoySmith, that too. My thoughts were along the lines of legal issues (if 'report' means report to us, here on Wikipedia, it'd probably be a privacy law violation to actually expect a paid editor to do that, so it effectively becomes a catch-22 pretext for banning). But 'report' could also mean report to employer that Wikipedia has paid policies, or something. Not really sure, the passage is ambiguous. ProcrastinatingReader (talk) 23:05, 13 July 2020 (UTC)

Not a violation privacy in the usual sense, but confidentiality—employees are generally not authorized to speak about business details to the public. isaacl (talk) 23:20, 13 July 2020 (UTC)

1st, there is no confidentiality for employers and clients under the ToU, to the extent that their existence and company names must be reported. 2nd, let me just give a couple of examples of violations that should be reported:

1. the commercial editing company tells the editor that he must create a new account, without disclosing it or linking to it, and editing a different article, a violation of WP:sock. Since the editor is required to put Wikipedia's best interests above his employers, he clearly has to report this, let's say to an admin or to WP:COIN. If he doesn't attempt to say "no I can't do this" to his employer, or convince them that he can't do it, he clearly has to resign as a Wikipedia editor. He could presumably do this and then go back to his employer and say, "please give me some other work to do." But, at least for those hired to edit Wikipedia, that tactic is unlikely to work - so in effect he'd be resigning both positions. How can we require that the editor reports his bosses rule breaking? It's pretty easy, the Wikipedia community sets the rules, and if the company doesn't want to follow them, they can forego editing. In practice what would likely happen is that the editor would follow the dictates of his wallet and create the sock account. Then it could be very easy to ban the lot of them. Sometimes - perhaps 5-10% of the time, he might actually report the company (there's likely a lot of turnover in this type of job) so over time we'd catch a many rule-breaking commercial editing companies that way.
2. Another example, perhaps more realistic, is that an editor is required by his employer to use a phony news article created by his employer, similar to a situation documented by The Wall Street Journal last November about Wiki-PR. That would violate WP:OR, and WP:RS and WP:Wikipedia is not for things made up one day Smallbones_(smalltalk) 04:18, 14 July 2020 (UTC)

By confidentiality, I was referring to employment contracts which usually restrict what employees can say about their employer. I agree the English Wikipedia community is free to set whatever rules it decides upon; just be aware that most people aren't going to breach their contracts (even if resigning, they can still be sued for breaking confidentiality). isaacl (talk) 05:05, 14 July 2020 (UTC)

Yeah. The WP TOU doesn't word itself in a way to imply that Wikipedia takes priority over your employer. It just says if you want to edit wiki, you must make the following disclosures. It implies that you get consent to make those disclosures, and if you can't, you don't edit. This policy is explicitly stating if you can't get consent for it, disclose to an admin and resign. There's a big difference here.

It's also unreasonable to think that people who can't follow the basic paid editing policy now will suddenly start disclosing against agreements and resign, so this mostly feels like a pretext to block. The examples are good, but the wording of the actual policy itself feels a bit iffy. ProcrastinatingReader (talk) 10:39, 14 July 2020 (UTC)

Too verbose

For the most part, I agree with what this is trying to say. The problem is, it's taking several pages to say it. Much of this duplicates existing policy. Some of it is just plain out of scope because it intrudes on off-wiki activity. Could you condense the core of what you're trying to do here into an Elevator pitch? -- RoySmith (talk) 13:22, 15 July 2020 (UTC)

Elevator pitch

• Commercial editors may only edit articles in the case of a BLP violation
• They may only edit article talk pages to link to press releases or to company webpages, which can give the same information that the commercial editor would otherwise post to an article talk page.
• No other commercial edits are allowed, including those to WP:AfC
• All their edits must be identified in an easily trackable manner (unlike the method for paid contributors' edits)
• Since Wikipedia editors love to argue about "What is a paid editor?" which has a definition set in stone by the WMF, there are sections that give broad classes of commercial editors, and exclude broad classes of other editors. These can be argued about and changed as circumstances require.
• (not quite done yet) Enforcement is easy and straight-forward.

Smallbones_(smalltalk) 16:40, 15 July 2020 (UTC)

Yes, keeping it concise like this is great in how it focuses on what is new and avoids duplication of existing guidance; the only thing that needs expansion is defining commercial editors. If you'd like I can take a pass at a draft. isaacl (talk) 16:47, 15 July 2020 (UTC)

@Isaacl, RoySmith, and MER-C: Please havea go at it. Smallbones_(smalltalk) 21:15, 15 July 2020 (UTC)

I've created a sandbox draft. isaacl (talk) 02:41, 16 July 2020 (UTC)

Username requirement

The username requirement is interesting — it would certainly be nice to be able to see when looking through an edit history or watchlist who is a paid editor. This would require a lot of rename requests, and I'm not sure how much disruption that would cause to handle. Would there be another technical means to achieve a similar result — for instance, create a list of paid editors and then add some code so that their names are highlighted or something? Just throwing out thoughts here.

Also, I'd encourage separating out the different proposals, since some may be more controversial than others, and you don't want one controversial element to sink the whole thing when another element might actually have support. {{u|Sdkb}} ^talk 04:05, 16 July 2020 (UTC)

Paid vs commercial confusion

From the original document I had mis-interpreted this, or at least, its definition seems (or seemed, if changed) different to that in the elevator pitch one.

I'd interpreted in, functionally, as saying those who are paid-editing as their job (i.e. are paid to create an article on a subject they're not linked to, whether freelancers or WikiPR-esk) are commercial editors, as opposed to the broader category of those editing as part of their job or even just as non-tasked employees are "just" paid editors. Nosebagbear (talk) 09:29, 16 July 2020 (UTC)

I also interpreted it the same. Commercial editor being an editor not linked to the subject, an organisation/individual which does paid editing for various clients. ProcrastinatingReader (talk) 10:02, 16 July 2020 (UTC)

Firms that mainly sell goods or services other than Wikipedia editing or writing services is in the original draft from Smallbones. But in any case, I don't think it should matter if the editing is being done by employees of company A or a firm hired by company A; both should be treated the same way, since the hired firm is a proxy for the company A. isaacl (talk) 14:40, 16 July 2020 (UTC)

Except employees are, on average, significantly less duplicitous. I also am willing to support more stringent restrictions on dedicated commercial editors (though not as broad as proposed) but not the general, very broad, category of paid editors. They aren't a true proxy because they don't act in the same fashion. Nosebagbear (talk) 15:08, 16 July 2020 (UTC)

I think loquaciousness and greater time resources is part of the concern behind limiting permissible actions. isaacl (talk) 15:27, 16 July 2020 (UTC)

Isaacl, this is meant to be a subset of the paid editing policy, rather than a replacement, if I understand it correctly. If I work for Microsoft and edit its article, I'm a paid editor not a commercial one. ProcrastinatingReader (talk) 15:10, 16 July 2020 (UTC)

There is a paid-contribution disclosure policy from the terms of use, but no explicit paid editing policy at present. As you can see from the "Commercial editors include" section in Smallbones's draft, from which I have quoted a small passage, the draft does include edits made by employees of Microsoft as commercial editing. isaacl (talk) 15:21, 16 July 2020 (UTC)

Isaacl, perhaps I'm being a bit slow here. Firms that mainly sell goods or services other than Wikipedia editing or writing services, if they edit Wikipedia, or hire a commercial editing firm as long as Microsoft didn't hire an external firm, a Microsoft employee editing the article wouldn't be classed as a commercial editor, but just a normal paid editor? ProcrastinatingReader (talk) 15:24, 16 July 2020 (UTC)

Microsoft is a firm that mainly sells goods or services other than Wikipedia editing or writing services. fyi, there is no need to ping me on this page. isaacl (talk) 15:35, 16 July 2020 (UTC)

Insufficient Permissive Actions

Commercial editors (whatever they are defined as) have their permissible actions narrowed to such a degree that we're shooting ourselves in the foot, even if you agree with the base policy.

They currently can't edit at the Teahouse or Helpdesk, so if they have any questions about these rules, or (say) editing a talk page, or submitting a rename request to move into compliance, they can't do it. All of those would actually make it harder for them to meet the ruleset.

They also can't edit at AN and ANI, so can't report actions taken against them or the pages they're watching, such as non-BLP vandalism which this ruleset wouldn't let them work on.

I'm still not sure why we can let them post sources that we generally wouldn't accept for anything but the most basic information but prohibit posting any regular source. Sure, they presumably won't post sources that cast them in a negative light, but that hardly rules out a majority of sources. The current phrasing effectively bars them from any participation while trying to make them think they can still provide something other editors will consider, which seems duplicitous. Resubmitted from elevator pitch page Nosebagbear (talk) 15:09, 16 July 2020 (UTC)

Agree that this has holes. There's also other obvious holes (eg, what about template talk namespace for articles relating to the company?), and the more we write to clarify exemptions, the more verbose and confusing it becomes.

I think we need to remember what the problem is here. Paid editors that declare their affiliations and follow the requested edit process aren't the problem. We don't need to toughen down on those that actually bother follow our policies already. Some of these editors actually produce decent (as in, not promo spam) content. This proposal has stemmed from a discussion on ANI regarding UPE. What we're trying to fight here are undisclosed paid editors, and to a wider degree, just general promotional spam across various areas on Wikipedia (as DGG mentions).

I really feel what we should be focusing on here is giving admins a wider toolset to take quick action (specifically, delete or implement prohibitions) against that spammy garbage. That's what we urgently need. The challenge is coming up with a policy that gives admins broad power to deal with that, without giving too much power that has proven to be a pitfall in prior proposals. To the extent a new commercial editing policy is required, it should mainly be to aid both transparency, and increasing abilities to sanction a certain subset of paid editing-like behaviour (ie "commercial editing") so that it doesn't all keep slipping through the cracks. And whatever we write has to be simple, multiple editors on this talk (including myself) appear to have confusion on the scope of this policy already. How are we going to expect new editors to understand? ProcrastinatingReader (talk) 15:35, 16 July 2020 (UTC)

I feel that by limiting the permissible actions by commercial editors (albeit perhaps with a slightly longer list than in the sandbox draft), it makes it easier for admins to deal with problematic commercial editors, since there would be less discretion for arguing over what actions are allowed. How to determine when it is reasonable to assume that an editor is in fact an undeclared paid editor is probably better covered in a guideline than policy, so it can be more easily adapted to evolving best practices. I agree more consensus on this aspect is needed. isaacl (talk) 16:08, 16 July 2020 (UTC)

I'm inclined to agree with ProcrastinatingReader. The dispute is usually over whether someone is an UPE - if someone isn't a disclosed paid editor they couldn't be bound by this set of rules. If they were an undisclosed, but known paid editor, they'd already have breached a clear brightline. Issues are mainly: identifying if someone is a UPE; stopping blocked UPEs coming back; and stopping new UPEs coming in. This additional policy wouldn't handle those. While a very expansive reading of "commercial editor" would knock out disclosed (or, at least, clear and willing to disclose PEs) paid editors out of the AfC queue, I think we'd just end up with more UPEs - which is the opposite of what we want. There probably are beneficial ideas - the idea of a standardised username form/add-on would be beneficial, but otherwise it's more enforcement mechanisms that are needed. Nosebagbear (talk) 17:14, 16 July 2020 (UTC)

My personal preference would be similar to DGG above and look at content. Promotional content = bad edit; ban editor after a certain threshold. It doesn't matter to me if the editor was paid, partisan, or just bad at writing. But I recognize that I'm more hardline than a lot of the rest of the community on this, and that Wikipedia probably would be a lot different if my approach were followed. isaacl (talk) 18:58, 16 July 2020 (UTC)

Agree on looking at content, but disagree with cracking down on disclosed paid editors. This is only going to create more UPEs. Our problem is with UPEs and spam, policy needs to fix that, not create more UPEs. And honestly, if we're going for policies which limit paid editors, we need a better process for paid editing reviews. The requested edit queue is always backlogged (and for good reasons). But when you got an employer breathing down your neck it's probably a little hard to say "sorry, but the one paragraph I wrote is in review, it'll be about a month". We can say that isn't our problem, but if we can't create a decent environment for paid editing they'll skip requesting edits. If we make our paid editing policies limiting to death, we'll create more UPEs. Creating effective policy isn't easy. A balance needs to be had between ideal goals, what the community will accept, and what will ultimately be effective. ProcrastinatingReader (talk) 19:51, 16 July 2020 (UTC)

Regardless of whether or not you are paid, if your edits are creating an undue burden on other editors, at some point your participation is incompatible with collaborative editing. As I alluded to, I recognize that current community sentiment is to work with such editors for an extended period of time to try to get them to contribute productively.

Regarding managing requests from paid editors, I would not support giving them priority based on their deadlines: they should be dealt with like any request from any editor. Somewhere there's a discussion about paid editing where I said if we're going to put in place measures where there will be more requests, we need to also put something in place to deal with the queue more expeditiously. So I agree with trying to make the whole queue move faster, but in a volunteer environment, that's hard. isaacl (talk) 20:07, 16 July 2020 (UTC)

I just realized I accidentally omitted from the sandbox draft some pages that were listed in this proposal as permissible to edit; I've updated the draft. isaacl (talk) 16:28, 19 July 2020 (UTC)

Over reaching

I think this proposal tries to do too much. It wants to put some teeth behind what we consider to be a paid/commercial editor. That alone would probably take a lot of discussion. It wants to implement new requirements on registering/naming paid accounts. That would take a lot of discussion. It wants to change transparency for paid editors. That is not a simple thing. It also wants to place broad new limits on paid editors. That definitely would take discussion. The thinking here seems to be that perhaps bundling all these things together makes it becomes one hard discussion rather than several. A reasonable theory but ultimately one that I think is wrong. I support some of what this proposal does but not other pieces. I am against the pieces I oppose strongly enough that I would have to oppose the whole thing rather than have the chance to endorse what I see as the good stuff. I read some other skeptical voices on this page and think I'm perhaps not alone. I would actually suggest it would be easier to gain consensus, not easy mind you just easier, for banning all paid editing (minus the carve outs listed) than for this proposal. Best, Barkeep49 (talk) 22:46, 18 July 2020 (UTC)

Two comments on what you said. First: Let not the perfect be the enemy of the good. Every proposal starts with some good ideas. We don't have the luxury of perfecting everything. I think this is good enough to fly.

Second: I followed the discussion about this being too big to appeal to enough editors to gain consensus, though I don't agree. However, your remedy was to go even bigger and ban all paid editing? I don't think that will gain consensus and don't understand the logic of going there, having said the first part. - Bri.public (talk) 19:18, 20 July 2020 (UTC)

Bri, I agree with not letting the perfect be the enemy of the good. But, in this pretty sheltered environ, I see anyone with concerns getting dismissed by the supporters. I don't think it'll play out that way in front of a wider group.

My underlying message is that this is both incredibly ambitious and very complex. I think a simpler proposal, such as to just ban paid editing, is far simpler while still being ambitious. I am not sure that would pass but it my suggestion that it is more likely than this to gain consensus. Best, Barkeep49 (talk) 19:30, 20 July 2020 (UTC)

I'm not sure who you are considering as supporters at this point: I think a lot of commenters have been raising their concerns and engaging in discussion on them. Smallbones has a specific vision regarding commercial paid editing which has been discussed in various venues, so the proposal doesn't surprise me, nor does the editor's stalwartness on certain points. Personally I think trying to make it an all-or-nothing proposal is more of a "perfect is the enemy of good" situation, and would prefer a more granular approach where specific options can be approved or disapproved separately. isaacl (talk) 21:05, 20 July 2020 (UTC)

I definitely agree an "old school" RfC where people can throw out whatever ideas they have for consensus or not (similar to how we do Arbitration Election rules or Wikipedia:Requests for comment/2020 left sidebar update) is going to help us move forward on paid editing better than this proposal as an up/down thing. Best, Barkeep49 (talk) 21:21, 20 July 2020 (UTC)

Yes, I also made this suggestion above in the "Coordination" section. I think it'd be a much cleaner way to do this. ProcrastinatingReader (talk) 09:02, 21 July 2020 (UTC)

List of individual ideas pertaining to combating paid editing

Since multiple individuals above have suggested they have more interest in a broken-down list of individual proposals, which wouldn't rise/fall as one, I've created a list of proposals.

It's definitely non-final, I just added those that came to mind - please bring your own.

The aim is not to submit final proposals - Smallbones' policy here is a very advanced version, whereas the list is, in many ways, more of a fact-finding mission. Any idea not slated will be taken through to community consideration. Anything that passes there that needs substantive additional work, will then be fleshed out and brought for confirmation.

This saves needing individuals to do an insane amount of reading, while saving significant creation work.

I'll read through this TP and ping those who have stated a specific interest. Nosebagbear (talk) 15:12, 21 July 2020 (UTC)

I think this is a more practical approach. Otherwise we'll be stuck with opposition. to various details. DGG ( talk ) 04:35, 28 July 2020 (UTC)

Paid editing tag

A vague idea, but coming off a discussion on IRC, the UPE tag is actually a relatively effective de facto badge of shame imo. For companies that are paying paid editors to write articles for them on Wikipedia, having a big red UPE at the top of your article is probably worse than not having an article at all. Yet, many of these articles get AfD'd, later recreated and the tag will be gone. This suggestion is a bit retribution-y, but if an article in a spammy area (companies, startups, BLPs especially 'entrepreneurs') can be shown to be UPE (and there are obvious tells), I think the UPE tag should persist through recreations, even if the recreations aren't UPE, at least for some period of time. I think it'd make companies think twice before hiring a spammy UPE org to write them an article. Again, very vague idea and it's probably awful, but no progress is made if ideas aren't raised so here goes. ProcrastinatingReader (talk) 00:19, 25 July 2020 (UTC)

The problem being that it's punitive on anyone else who writes on the article. It also risks discouraging anyone who runs it through AfC, which is what we'd rather they do. It certainly does have a badge of shame effect, but I'm reticent that the collateral outweighs the gain. One particular problem being that most companies presumably only hire editors to write one article about them (potentially more than once if it's deleted, but still 1 article). While the badge of shame might discourage them from risking the issue again, I think we'd struggle to get it known well enough to prevent others from UPE. Nosebagbear (talk) 12:45, 25 July 2020 (UTC)

It also risks discouraging anyone who runs it through AfC speaking of, here's another bad idea: a broader AfC restriction for articles created in certain spammy topics (startups, entrepreneur BLPs, all the areas DGG mentioned above, etc). All non-extended-confirmed (or autopatrolled) editors need to go through AfC to create an article in the topic (optionally: only for their first 3 articles). Any editor can draftify articles that don't meet this criteria, and probably some usage of the ORES topics API could help here too. One obvious downside is that it adds to the already-long AfC queue, so we'd have to do some data collection to predict how much load we'd be adding. ProcrastinatingReader (talk) 13:21, 25 July 2020 (UTC)

I think we can deal better with the AfC queue if we didn't have to go through the effort of finding articles that try to evade it. I will be suggesting several simplifications there. DGG ( talk ) 04:38, 28 July 2020 (UTC)

Alternate accounts and user privacy, and innocent people caught up by the policy

If users are going to be required to register their alternate accounts, there needs to be a way to do so that is private, especially if the user was an editor before he came under this policy. There should also be an "out" for people whose jobs are clearly not related to promotion to continue to edit Wikipedia as long as they only edit pages that an alternative non-commercial account would be allowed ti edit.

Remember, some people may work for firms like this in areas that have nothing to do with Wikipedia. If I'm a janitor for XYZ Promo Inc. or a janitor for a firm that contracts with them for cleaning services or a janitor who as an independent contractor provides cleaning services for them, I'm covered under this policy as it is currently written. I shouldn't have to create a dummy "commercial" account and publicly identify it with my existing account just so I can keep editing the way I have edited for years just to comply with the policy. It should be enough for me to email OTRS if and when I become aware of the fact that I'm covered by the policy. If I never become aware that I am covered for any reason other than deliberate bad-faith indifference then I shouldn't be sanctioned until I do become aware that I may be covered. For example, if I'm an independent contractor for XYZ Foods Inc. and they were bought out by XYZ Promo Inc. suddenly I am covered by the policy, even if I have no clue that the merger happened. davidwr/(talk)/(contribs) 20:38, 2 August 2020 (UTC)

If we add the condition that commercial editors are a subset of those who are required to make a paid-contribution disclosure, as per the terms of use, then this policy will only cover those who are compensated for editing (or are deemed to be compensated for editing, based on the clarifications that have been added to English Wikipedia's paid-contribution disclosure page). This is what I had written in my draft revision. isaacl (talk) 21:20, 2 August 2020 (UTC)

It's going to be tricky to avoid gaming the system. For example, if I'm that janitor and I am at an "all hands" meeting where the Chief Communications Officer proudly talks about editing Wikipedia, I should be covered from that point onward the same as someone in the PR department with respect to editing articles that I know would benefit my company or its clients, but not covered for other articles. In other words, once I know my employer or a company I contract for is in Wikipedia-manipulation business, I have a choice: Either steer clear of anything I know would be covered under the standard paid-editing-disclosure or for that matter standard conflict-of-interest rules entirely, or accept the burden that comes with complying with this proposed policy. On the other hand, people in the PR and other "PR-heavy" or "Officer-level" departments shouldn't have that choice: They must stop editing Wikipedia altogether or comply with the burdens of that policy, if nothing else as a show of good faith. That said, a person in the PR dept. whose job doesn't include editing Wikipedia or encouraging others to do so should be allowed to comply with the policy non-publicly, through OTRS or some similar process if they will not be making any edits that the policy is intended to cover. davidwr/(talk)/(contribs) 16:09, 3 August 2020 (UTC)

Publicity employees are currently deemed to be compensated for editing (on a side note, I participated in adding that clarification). Employees in other departments are not, which I think is a reasonable (if imperfect) reflection of whether or not an edit is being made at the behest of the company. The conflict of interest guidance covers scenarios not involving paid edits. If a PR employee isn't making "edits that the policy is intended to cover", their edits aren't considered paid and they don't have to disclose anything. As I wrote previously, personally I think we should use the nature of a user's edits to identify and block editors making promotional edits. Nonetheless, that doesn't preclude also having restrictions on a specific subset of paid editors. isaacl (talk) 05:48, 27 August 2020 (UTC)