Wikipedia talk:WikiProject Editor Retention/Project Handshake

"phew"

I have created this page, Wikipedia:WikiProject Editor Retention/Project Handshake, as a development page for a discussion which began at Wikipedia talk:WikiProject Editor Retention#Discussion about a New WER Welcome. ―Buster7 ☎ 07:04, 9 January 2019 (UTC)

User_talk:RyanRo04#WELCOME! is a the first prototype welcome, created by User:Edaham on 9 January 2019 (UTC) and intended to learn more about the new editor and open a progressive discussion.―Buster7 ☎ 07:13, 9 January 2019 (UTC)

PHEW statement

After a limited "standard" welcome (still to be determined which one), a friendlier, more personal "phew" statement is attached. ―Buster7 ☎ 15:44, 9 January 2019 (UTC)

*Phew*

With that out of the way, I hope you will look around and find other areas of interest and become an active editor. Let me know your specific field of interest and I will see if I can introduce you to some members who can work with you and get you started on some article editing.

Icon

This was an early suggested logo for Wikipedia project WER created by the User Mark Miller. ―Buster7 ☎ 13:22, 9 January 2019 (UTC)

What is the problem we are trying to solve or how we propose to solve it

• The following is a conversation at User:Iridescent's talk page. It seemed pertinent and informative. I have taken the liberty to edit what was unnecessary banter--―Buster7 ☎ 00:14, 13 January 2019 (UTC)

I’ll give you an example. At the moment I’m making some suggestions at editor retention. My suggestion is to add a question to the welcome temple, “Can you tell me a bit about your interests and the areas you want to edit?”. My reason for suggesting this is 1) its nice to invite conversation 2) if answered the editor can then pass on the new editor to relevant wiki projects. Supposing you were to poll a group of editors on their user experience, before during and after the introduction of such an initiative. A positive change over time would be quite a useful bar of soap in a sock to anyone who claims people aren’t doing enough to increase hospitality, for a start. Having made the proposal, and thought about your responses, these are the lines along which I’m now thinking. Therefore, seeking out and identifying “unhealthy” pages, or defining what that means might not be a useful exercise. What I think is probably more useful is to hypothesize areas where data might correlate if you took a problem and addressed it with some initiative. From a long term point of view article content and community satisfaction aren’t really separable in terms of importance. They’re mutually supportive in many ways. I can see why wmf might be concerned about “health”, I can also see how as outsiders, they might be likely to come up with some heavy handed or less than helpful ideas. FYI I wrote this in a hurry, I might come back and edit it if you haven’t replied. Edaham (talk) 06:19, 12 January 2019 (UTC)

I think you're vastly overestimating the size of the Wikipedia community if you think polling a self-selecting sample would give meaningful data. The headline 48,261,578 figure for the number of accounts represents everyone who's ever registered an account, even though most of those accounts have either never edited or not edited for years; the actual size of the active Wikipedia community is c. 3500. (The raw data on editor participation and activity is here, if you want to verify this; the column labeled ">100" is the one you want to be looking at as those are the editors who are actually active, rather than just poking their noses in occasionally to keep an eye on their particular topic of interest.) A sample would by necessity be self-selecting; we have no means to compel people to participate in surveys. It's well documented that self-selecting participants give the answer they think the questioner wants to hear, and the people who'd choose to participate in surveys would naturally over-represent people with strong opinions (either the "Wikipedia is the greatest force for good of modern times!" fanbois or the "Wikipedia is an irredeemably malignant force and it's a duty of all good-thinking people to disrupt it by any means necessary!" Hasten the Day hardliners). In addition, because of the limited size of the community small shifts in editor participation can have relatively significant effects on Wikipedia (I can rattle off any number of instances where the comings or goings of just a single editor has had a drastic impact on Wikipedia's coverage of a given field; not just niche topics, but core fields like "paintings" or "England").

Ultimately, market research techniques intended for analysis of broad trends by teams of professional researchers, or of a professional workforce whose employees work full time and can be compelled to answer the questions, are at best going to be of limited value, and at worst actively disruptive, when applied to a small community of enthusiasts. (If you want to go down the rabbit hole on this, start at Wikipedia:Research and follow links; remember that we've been one of the highest-profile websites in the world for well over a decade now, and almost everything one can think of in terms of research has already been thought of.)

Take your example of adding questions to the {{welcome}} template. We have roughly 5000 account registrations each month; assuming one in ten of those replies to the questions, that makes 500 replies. Every one of those 500 editors will then get offended if they go to the trouble of answering the questions and nobody follows up with them; assuming there are at most ten active editors at WikiProject Editor Retention who actually have the inclination to do the heavy lifting,* that commits each of those editors to engaging in-depth with 50 editors each month; even in the first month that will be a huge time-sink, and assuming the editor-retention aspect is successful it will create a compound-interest situation in which after six months, each of those ten editors will be de facto mentors to 500 editors apiece. To put that in perspective, this talk page is one of the most active pages on the entire project and it takes a significant chunk of my time just responding to comments on it, but even I have only 500 watchers and most of those are passive; to engage simultaneously with 500—or even 50—editors would be a full-time job, and anyone trying to actually do so would burn out within days.
*A big "if"; in my experience, WP:ER consists less of people with a genuine interest in why editors come and go, and more of people who've lost arguments kvetching about how it's Just So Unfair that nobody on Wikipedia appreciates their genius and threatening to leave unless we immediately accede to their demands.

Likewise, I think you're working under a misapprehension regarding WikiProjects, if you feel passing on the new editor to relevant wiki projects would be a good thing. There are perhaps a dozen Wikiprojects that are actually active; what you're actually proposing is that unless the new editor's interests are military history, trains, medicine, videogames or a few other topics that have active projects, new editors be directed to dead talkpages where their questions will go unanswered, and consequently frustrate them even further. The existing system—in which new editors edit an article in which they're interested, and consequently come to the attention of other editors with an interest in that topic who engage with them as to what they're doing right and wrong, all the while being nudged towards The Wikipedia Adventure to get a feel for how the technical aspects of Wikipedia work without causing a nuisance by experimenting in the mainspace—isn't some arbitrary system we've made up from whole cloth, but the project of two decades of evolution and experimentation regarding what works and what doesn't. ‑ Iridescent 08:01, 12 January 2019 (UTC)

I think introducing new people to wiki projects isn’t a great idea. Introducing new people to experienced people is a good idea. Sending someone to a project is pot luck. Taking the time to introduce them to a person who made the most recent few edits to their area of interest (or the closest fit possible) is probably better, and that’s what I’ve been suggesting. I agree with you about some of the passers-by at ER. I’ve seen some grumbles in the short time I’ve been there. I however, think its quite a good idea to think in the direction of strengthening new users - so at least there’s one new person there to join those who take it seriously. Regarding your thoughts on marketing data, you’ve certainly highlighted some pitfalls. In a project with no deadline however, I think its worth exploring further, if perhaps more tentatively. Edaham (talk) 15:30, 12 January 2019 (UTC)

As the "caretaker" at WP:WER I like to roam around and search out comments that may be helpful to discussions that are ongoing at WER. As an off-shoot to one of those discussions, User:Edaham presented an idea that may have potential. He briefly mentions it above and it is in the very genesis stage. I wonder if I might "cherrypick" some of the discussion above as they relate to Edaham's idea and incorporate them at Wikipedia talk:WikiProject Editor Retention#Discussion about a New WER Welcome. Also, mostly all of the kvetching done at WER is not by members since most members are no longer active participants in the project. The remaining few do what we can to steer discussions toward a purpose beyond complaining. ―Buster7 ☎ 17:26, 12 January 2019 (UTC)

Slightly adjacently, you going to run out of Eddies sooner rather than later  :) although that does somewhat tie in with Iridescent's point about a dislocation between the number of editors / and the number actually participating I suppose. ——SerialNumber54129 17:52, 12 January 2019 (UTC)

Yes, of course. I don't mean to denigrate people like you and Dennis, who are genuinely trying to help, but I think it's reasonable to say that the drive-by attention-seekers posting variations of "if Wikipedia continues claiming that there's no evidence to support my crank theory I'm going to resign, you're the Editor Retention project so if you don't force everyone else to do what I say you'll have failed in your duty" gives a false impression that the project's much more active than it is.

English Wikipedia editors with >100 edits per month

The number, and viciousness, of internal squabbles is a fraction of what it once was, and the decline in editor numbers has ended (see right). Despite the WMF's sky-is-falling hyperbole—which is based more on Jimmy Wales's crackpot Civilination project and his determination to 'prove' that Wikipedia is a hostile environment so he can justify spending donor funds trying to solve the alleged problem, than on any objective evidence of problems—Wikipedia is fairly calm compared to the Sue Gardner/Lila Tretikov era when there was a genuine possibility of open civil war and forking.

The primary driver of editor retention problems nowadays isn't the nature of community interactions, but that improved writing standards and more complicated formatting have made the learning curve much steeper than it used to be. VisualEditor was supposed to address this, but the disastrous mess the WMF made of the launch alienated most of the existing editor base who consequently refused to adopt it, and as a consequence we now have a two-tier community of pre-VE editors who see Wikipedia in terms of markup, and post-VE editors who see Wikipedia in terms of text, talking past each other and getting annoyed with each other because neither group understands the other; in addition, because VE editors find the back-office areas (which still function in wikitext) hard to engage with other than by means of automated tools, they're de facto ineligible for adminship as they find it hard to demonstrate experience in admin areas, cementing the gerontocracy of the 2006–08 intake. (The WMF is aware of all these issues, but after the Flow, MediaViewer, Superprotect and VE debacles they're not likely to attempt to impose another cultural change by technical means without community consensus, and whenever the community are given the opportunity to state their priorities they remain stubbornly wedded to bikeshedding and deckchair-arranging, probably because the WMF's "everyone is equal" mantra gives equal weight to the views of people from tiny moribund projects as it does to people from the big Wikipedias and from Commons.) ‑ Iridescent 18:16, 12 January 2019 (UTC)

Don't worry; those of us still trying to shepherd the formation of new ideas are well aware the project is essentially moribund. I have to take care not to be overly pessimistic and quash newcomers, while at the same time trying to prepare them to not be disappointed if their proposals don't progress the way they had hoped. Although I'd state the underlying problem differently than you—I'd say that continued growth in the number of new editors is more likely to come from those who aren't interested in using wiki text—the net conclusion is the same as you stated: the wiki text editor is a barrier. And supposed consensus-based decision-making traditions stalemate change, since a small number of vocal editors can dissipate focus and prevent agreement. The editor retention project can't address the real systemic issues, but I hope it may be able to find some small ways to encourage editors. isaacl (talk) 19:31, 12 January 2019 (UTC)

And the upset editors are fun, as we get upset editors complaining that the project isn't stopping other upset editors from making their invalid complaints on the project talk page. isaacl (talk) 19:36, 12 January 2019 (UTC)

I think we're on the same page; I'd agree that the wikitext editor is a huge barrier, and in an ideal world it wouldn't exist. Unfortunately, even when VE is working perfectly—which will probably be some time around the year 2168—it will still be fundamentally incompatible with wikitext, as even when it's outputting material that looks fine the underlying code it generates looks like a soup of markup codes. (Moving files from a WYSIWYG word processor like MS Word into a code-based application like the old WordPerfect had exactly the same issue.) In the very long term, we should probably be thinking about the feasibility of getting rid of wikitext markup entirely, but I can't imagine there ever being a consensus for that. In my experience, the other huge barrier is that the minimum standards are higher; back when this was a "good article" new and newish editors could start writing and not worry about getting the formatting and referencing right provided they weren't completely screwing up, but nowadays we jump on editors for not including references or for getting the formatting wrong. I'm not saying we shouldn't be doing that—making sure editors learn early on that verifiability is non-negotiable is important—but it must be fairly dispiriting for someone who's always prided themselves on their writing skills to be slapped down by an anonymous stranger on the internet for failing to comply with some arbitrary rule in WP:MOS. ‑ Iridescent 20:08, 12 January 2019 (UTC)

As most of you must know, Dennis is retired. He makes a visit every now and then but we can't convince him to stay. isaacl and I do what we can to keep the doors open and the place relatively tidy. All are welcome. Drop in. Nominate someone for an Editor of the Week award. ―Buster7 ☎ 23:41, 12 January 2019 (UTC)

I agree that the referencing in wiki markup is a problem. My articles use the {{sfn}} system - save for some older creations such as Coropuna - and I don't recall a single newby or inexperienced user getting it right when they try to edit them. Not one. Even an admin had recently difficulties on one of my articles. Jo-Jo Eumerus (talk, contributions) 20:54, 12 January 2019 (UTC)

See #VE and referencing in the megathread above. Because VisualEditor can't handle {{sfn}}, and we (probably correctly) push new editors towards VE rather than inflict wikitext on them from the outset, new editors can't get it right when they try to edit them, even if they wanted to. As I may have mentioned once or twice, I strongly believe the benefits of introducing a single citation format would be worth the significant hassle its introduction would cause; whatever system we came up with we'd get used to it fairly quickly, and we'd no longer need to have the "oh don't be put off by our having 2000 citation styles, you're only expected to be familiar with the twenty or so that are in common usage" discussion, which invariably makes Wikipedia look ridiculous. ‑ Iridescent 21:05, 12 January 2019 (UTC)

Twinkle welcomes

• Wikipedia:Twinkle/doc#Wel (welcome)

Twinkle's welcome functionality

• The wel tab will welcome the user with a selected template
• The Type of welcome gives a menu/list of WikiProject-related welcome templates

Twinkle's welcome functionality

• Twinkle is capable of supporting a custom list of templates that displays below the standard set of Welcome templates. If you have installed Twinkle, you can add a list of templates that you would like to be added to the Welcome dialogue by changing the "Custom welcome templates to display" option in your Twinkle preferences. Any templates added to this list, when selected, will simply be placed on the user's talk page with no heading, no arguments, and will be followed by your signature.

Custom list of templates

• Standard welcomes
• IP user welcomes
• WikiProject welcomes
• Non=English welcomes