Topic on Talk:Growth/Discussing potential Growth ideas

Discussion "New editors suggest edits (3)"

MMiller (WMF) (talkcontribs)
Arthur Crbz (talkcontribs)

Oppose We need to encourage people to edit Wikipedia not to suggest edits: "do it yourself". This is what we are doing here...

Dave Braunschweig (talkcontribs)

Support This proposal is not that different from the approved edits approach already in use on Wikibooks. The idea could take two different forms. One option is the suggested or proposed edit where the user edits the existing text with a proposed alternative. The other is more of a comment suggesting an edit, which might be in the form of a suggestion that an edit be made but asking another editor to actually make the change. This would be somewhere in between editing and adding to the Talk page, which is also intimidating and unfamiliar to new users.

The only initial difference between these two approaches is in the wording of how is implemented. If it is Suggest vs. Publish, they are editing. If it is Suggest an Edit or Suggest a Change, they are commenting. I don't know which of these would be better, but something to consider.

Leaderboard (talkcontribs)

Oppose That is confusing, and really theoretical. As a sysop on en.wikibooks, we already implement a type of edit verification in the form of PendingChanges. I think that does the job just fine (the only drawback is that users are served the latest draft by default, and hence bad edits will show up till they are reverted).

Asking an user to 'propose' changes will simply make things scary; the user may think "what if it is refused...?" and hesitate to make the edit. Conversely, Wikibooks' approach makes this transparent to the user, who'll edit as if this 'verification' never existed.

Kerry Raymond (talkcontribs)

I think we need to tease this problem apart into two separate pieces. The first part addresses the new user who doesnt know HOW to edit, so giving them a free text (just type) window to ask for a change to the article and funnelling that onto the article's Talk page seems an OK solution to that problem. Anything that involves them knowing anything about HOW to edit (or where to find the Talk page etc) must not be required as part of the solution. The problem with the first piece is whether there is anyone willing and able who is watching that article who will take any action on the proposed change. The second part of the problem is whether the proposed change is accceptable in the opinion of the person who is considering implementing it. And my watchlist shows me that many first edits are vandalism (both obvious and subtle), change facts without adding/changing citation, and add inappropriate content (opinions, chatty, spammy) or delete cited information (often whitewashing) and most won't have an edit summary or else it's an unhelpful one ("I changed some details"). And then there is copyvio, need for citation, etc. Now it is easy to deal with new user edits that are obviously bad faith (undo!) but harder to deal with the ones that may (assuming good faith) be genuine. In the absence of a citation, I have to search to see if I can find a source. If I can't find a source, I have to go back and ask them (via the Talk page that I know that they probably aren't reading) if they can tell me what the source is. Or why they think that information should be deleted? So either we use the same chat mechanism (already described for getting help) to allow for suggestions to be made and discussed.Or we make the suggested change more of a wizard that starts with "What do you want to suggest?"

  • Fix a simple presentation mistake like spelling, grammar
  • Change some information that is wrong or misleading
  • Add some new information not already in the article
  • Delete some information from the article
  • Etc

and then leads to a series of different forms. The simple fix could be 3 boxes (existing words, proposed words, rationale). The rational could be optional here as it may be obvious "Theer" -> "There". The other forms would need more rationale and in some cases a separate box for a source (which mightl ead to a cite-manual option and interrogate for the minimal details for a web citation, book citation, etc so that the implementing user can be presented with a ready-to-go-citation to ease their implementation effort.). Now you might think "surely this is more work for the new user than just editing it in the first place" and, indeed for an experienced user, it is more work, but for the new user, it's not, partly because they don't know how to edit (or lack confidence) i and partly because this dialogue starts with their *intention* which the normal editor tool bars do not and gathers the necessary information relating to their intentions which is often needed to understand the appropriateness of the proposed edit (and is normally lacking with post-edit watchlist reviewing). An in-between step is to make edit summaries mandatory and structured into a dialogue.

علاء (talkcontribs)

In arwiki any edit should reviewed by editor before published (Pending changes system) + flagged revisions system.

It's very good and encouraged users to publish there edits without any afraid, but the problem that we have, there's a huge accumulation in pending changes log, and few editors working in it.

And "we try" to send notes in user page if we refused her/his edit(s), and in the same time, to send a notification if we accept the user edit (like here). Note that the notification system in Wikipedia, send a notification for you if your edit refused/rollbacked, but not send if it accepted.

I think if we send a notification that "Thank you. Your edit(s) accepted! We look forward to reviewing your next edit" will encouraged users to make more and more edits without any afraid, and we should work to make this users know that if they did any wrong thing we can fix it.

Jetam2 (talkcontribs)

The idea is interesting but a lot depends on how it is implemented. Suggestions in the form of a discussion text create unnecessary work for the reviewer and editor-to-be. An extra page would have to be checked and the original one actually saved. It could work better if the suggestions are real edits that are somehow temporary until reviewed and deemed worthy and permanent. But this is not that different from the present reality: edits are real and visible unless someone watching recent changes corrects/reverts them. In either case, optionality is a good step.

MMiller (WMF) (talkcontribs)

Thank you all for weighing in on this "New editors suggest edits" idea. Our team discussed all the feedback, and I've summarized it here. We're now figuring out whether to pursue this idea, and we'll be back to discuss more if we decide to work on it (please sign up for our newsletter to get updates on our plans).

Our team really liked @Kerry Raymond's idea of a structured workflow or form that a new editor could use to suggest an edit productively, especially because it could strongly encourage them to add a source. Such a workflow could be a good way for experienced editors to find good-faith contributors who just need help with the technology.

I wanted to ask a follow-up question to @Kerry Raymond and to whomever else has an opinion: do you find many new editors who are afraid to be bold and edit? In other words, in your experience, how much does the problem here actually exist?

Kerry Raymond (talkcontribs)

I see new users in real-life in two different situations, one is edit training, the other is edit-a-thon events. The behaviour tends to be different.

People who go to edit training seem to fall into 2 groups, the tried-but-failed and the not-bold. The tried-but-failed have tried to edit in the past on their own and failed (for technological reasons, got reverted, got a nasty message) and gave up and now think that with some training they might be able to have another go at it. These people are clearly keen to edit. The not-bold are people who think they might like to edit Wikipedia and would never reach out and click that Edit button on their own; they lack confidence in some aspect of the task whether it be not confident with IT, not confident that they are an "authority/expert", not really sure that "people like me" are allowed to contribute. I note that the main page may say "the encyclopedia that anyone can edit" but these folks either haven't seen it or have seen it but misunderstood it. The problem is that many people think of "edit" as like the "editor of a newspaper", not as a contributor/reporter (to use the newspaper analogy). For this reason, when I run edit training, I try to get it advertised as "learn to *contribute* to Wikipedia" because the "edit" word is misunderstood and then in the edit training session, I try to explain the nonemclature so they know what we mean by "edit" and "editor". So yes there are a lot who are not bold or or have been bold and failed and are not bold anymore. So these folks are generally wanting to contribute but feel they need to be shown how first.

The majority of people who attend my edit training sesions are either librarians (who tend to take to Wikipedia editing fairly easily -- being IT literate and understand citation) or the general public. Both groups are predominantly female. General public groups are typically older people, often retired. A few of the older people may have low IT skills although they self-assess as competent (they tend to be able to handle email, Facebook, web browsing, uploading photos from their camera or phone, but lack experience in producing content which would give them skills like copy-and-paste, having multiple tabs open in a browser at one time, knowing what a URL is and where to find it on your screen). A problem with some older people relates to the more limited education they would have had in the time they grew up, and they tend to regard "what an expert once told them" as authoritative rather than grasp the need for citations. "My grandfather was a pioneer in the Great Dry Plains district and he said that ..." is a "fact" for them, so we get some reliable source problems with this deference to "experts talking" over published sources.

At edit-a-thons, you see a much more confident group. Most edit-a-thons I experience are about the gender gap, and mostly women (and some men) who come along are motivated to close the gender gap rather than to become regular contributors to Wikipedia. They tend to be younger or middle-aged, university-educated professional women (often academics, university students, scientists as universities often run such events). They are confident with IT and mostly comfortable with writing and citing. They have been told they will be briefly taught how to edit as part of the event and they assume they will pick it up rather easily (and indeed they do). Many will jump in and be bold at the event even without any training. So this group tends towards boldness but motivated by the gender gap and probably do not intend to contribute after the event ("too busy"). Their main problem is that they will violate copyright on university websites because "the university won't mind" (which is probably true, but not a reason Wikipedia accepts without some OTRS process) .

So the not-bold are definitely out there. As are the not-aware and those who don't realise that clicking the "Edit" button allows you to change Wikipedia. Also Visual Editor is not the default so I suspect many who do click Edit, get shown the markup, and don't proceed because it looks like incomprehensible to them. I believe someone has some stats on what proportion of the clicks on Edit but never result in a Save. One assumes a fair number of these aborted edits by new users are "OMG it's all too hard" reactions, so fall into my bold-but-failed category.

HLHJ (talkcontribs)

Illiterate editors are excluded from the wikis, a major systemic bias. Illiterate people can listen to wiki content through voice synthesis, and find it with voice commands, but editing is harder. Suggested-edits which consisted of sound recordings or voice-recognition-generated text (for smaller data transfers and storage space) could let them contribute; a literate contributor could then polish their edit, source, and insert it. As a halfway step, a user interface for voice-input edits to talk pages might be helpful. The interface could also be used to let people using Wikipedia through voice assistant chatbots (Mycroft, Alexa, OK Google) contribute. HLHJ (talk) 20:25, 11 November 2018 (UTC)

MMiller (WMF) (talkcontribs)

Thanks, @HLHJ. We hadn't thought about illiterate editors before. We'll keep that in mind as we continue our work.

Reply to "Discussion "New editors suggest edits (3)""