[Wikimedia-l] Re: Mopping with the tap open

[Clover Moss]

Oh sorry, I didn't realize someone else had already beat me to linking to
it! Apologies for the duplicate material.

On Wed, Mar 6, 2024 at 7:03 PM  wrote:

> Hello
>
> Some wikis have added the requirement to add citations at the edit summary
> step. But it is clearly too late in the process, as users just want to
> publish. Some users will add citations as a second step, but it might be
> too late, as the edit has a great chance of being reverted meanwhile.
>
> You might be interested in the Editing team's current project, Edit check <
> https://www.mediawiki.org/wiki/Edit_check>.
>
> This project aims to provide in-context help by checking on the edit. The
> first iteration is "Reference Check": if a user adds a paragraph with zero
> source, they are encouraged to add one. We are currently testing it at 22
> Wikipedias, to verify if the prompt to add citations is not blocking users.
>
> You can test it at your wiki using an URL parameter:
> 1. Edit any article in the main namespace using the VisualEditor.
> 2. Add =1 to the URL in your browser. -- For example in Dutch, as
> Romaine started the thread:
> https://nl.wikipedia.org/w/index.php?title=Zon=edit=1
> 3. Reload the page with the new URL.
> 4. Create a new paragraph, that is at least 50 characters long without
> adding a citation
> 5. Press the Publish… Notice the prompt that appears
> 6. Test is completed, don't save your edit unless you know what you are
> doing.
>
> All edits are tagged, so that you can find them in Recent Changes or in
> your Watchlist. If a user selects "no" after the prompt, they have to
> select a reason why. That reason is tagged as well, easing experienced
> users' work on patrolling and improving these edits.
>
> We will soon add a message if the added citation is listed on
> MediaWiki:Spam-blacklist or MediaWiki:BlockedExternalDomains.json.
>
> As Edit Check only checks the first paragraph added, the next iteration
> will be to add multi-Reference checks. We are currently working on the
> design for multi-checks.
>
> Of course, Edit Check is not limited to adding citations. We can imagine
> other ways to close the tap. Your suggestions are welcomed, as are your
> questions.
>
> Thank you,
> --
> Benoît Evellin - Trizek_(WMF) (he/him)
> Community Relations Specialist
> Wikimedia Foundation
> ___
> Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines
> at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and
> https://meta.wikimedia.org/wiki/Wikimedia-l
> Public archives at
> https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/message/RWQIXLQEBNC62THG5J4TY7OCHCKRAPUF/
> To unsubscribe send an email to wikimedia-l-le...@lists.wikimedia.org
___
Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
Public archives at 
https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/message/V2X2AEELZMGMB6AFV3NQPIB4TAI2CRKD/
To unsubscribe send an email to wikimedia-l-le...@lists.wikimedia.org

[Wikimedia-l] Re: Mopping with the tap open

[Clover Moss]

Hi Romaine. You might be interested to learn about an ongoing project
called Edit Check [https://www.mediawiki.org/wiki/Edit_check].

On Wed, Mar 6, 2024 at 8:59 AM Romaine Wiki  wrote:

> In the past days, a new Wikipedia contributor edited Wikipedia and made a
> great contribution, except... This user added zero sources, and the article
> in what the edit was made was about a living person. So the verifiability
> is a problem and in conflict with the policy Biographies of living persons.
> This was just one example of thousands that have to be dealt with every day
> in Wikimedia. And every day the community tries to maintain the quality of
> Wikipedia and has to deal with this kind of edits.
>
> I asked myself the question: why did this new contributor not add any
> sources?
>
> I logged out, went to an article and clicked edit. Made some modifications
> (in the Visual Editor), and then clicked Publish changes. In the steps I
> took to edit the article, I got nowhere a message that Wikipedia wants to
> have sources for the information I added. Nowhere!
>
> I hope that every experienced user by now understands the importance of
> adding sources. But we cannot expect from new contributors to already know
> this. They need to be informed that adding sources is needed. They do not
> go first read the manual of Wikipedia with all the help and project pages,
> they just start editing right away. They think, link in many other
> platforms, that if they do something wrong, they get a message while
> editing/uploading/etc.
>
> For some strange reason, if you edit Wikipedia, you get no notification at
> all that you need to add sources, even while this is one of the most
> important pillars of Wikipedia. The result is that a lot of work of these
> new contributors gets lost, because the information is removed from the
> articles because of a lack of sources. If those new users would have got a
> message in the Visual Editor during the editing, a lot more contributions
> would be able to stay in Wikipedia, less new contributors would get
> demotivated, and it would reduce the workload of existing users who do the
> maintenance every day.
>
> As with the influx of edits without sources nothing is done, the Dutch
> expression "mopping with the tap open" (Dutch: dweilen met de kraan open)
> applies here.
>
> Romaine
>
>
>
>
>
> ___
> Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines
> at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and
> https://meta.wikimedia.org/wiki/Wikimedia-l
> Public archives at
> https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/message/2J32V233R72OWB5W2DKGXIGBPVC6Y75B/
> To unsubscribe send an email to wikimedia-l-le...@lists.wikimedia.org
___
Wikimedia-l mailing list -- wikimedia-l@lists.wikimedia.org, guidelines at: 
https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and 
https://meta.wikimedia.org/wiki/Wikimedia-l
Public archives at 
https://lists.wikimedia.org/hyperkitty/list/wikimedia-l@lists.wikimedia.org/message/Z7LGNY3K4FLOMTF77P5GAQT4QXQZOM4B/
To unsubscribe send an email to wikimedia-l-le...@lists.wikimedia.org

[Wikimedia-l] Re: Mopping with the tap open

[Steven Walling]

On Sun, Jun 9, 2024 at 11:51 AM Steven Walling 
wrote:

> How much does Edit Check decrease the total number of saved edits and
> unique editors?
>

I should have looked before asking in the results you linked to. Short
answer: “On mobile, edit completion rate decreased by -24.3%”

In other words we lose 24% of saved edits in order to decrease the revert
rate by 8.6%. This tradeoff does not seem good.

On Sat, Jun 8, 2024 at 7:07 PM Peter Pelberg  wrote:
>
>> If those new users would have got a message in the Visual Editor during
>>> the editing, a lot more contributions would be able to stay in Wikipedia,
>>> less new contributors would get demotivated, and it would reduce the
>>> workload of existing users who do the maintenance every day.
>>>
>> Romaine – and everyone here who resonated with what Romaine expressed
>> above – I thought you might value knowing that a recent A/B test
>>  of Edit
>> Check  (the idea Benoît
>> shared here
>> )
>> supports the assumptions you're making above and in this thread more
>> broadly.
>>
>> Specifically, the A/B test showed:
>> * People [i] shown the Reference Check are *2.2x* more likely to publish
>> a new content edit that includes a reference and is constructive (not
>> reverted within 48 hours).
>> * The highest observed increase was on mobile where people are *4.2x*
>> more likely to publish a constructive new content edit with a reference
>> when Reference Check was shown
>> * New content edit revert rate decreased by *8.6%* if Reference Check
>> was available.
>> * Contributors that are shown Reference Check and successfully save a
>> non-reverted edit are *16%* more likely to return to make a non-reverted
>> edit in their second month (31-60 days after).
>>
>> You can read the full report that Megan Neisler
>>  prepared here: Reference
>> Check AB Test Analysis
>> .
>>
>> If anything you see brings questions/ideas
>>  to mind, now is a
>> wonderful time to share them. Reason: the Editing Team is
>> actively planning  how
>> to expand Edit Check and needs volunteer expertise to shape this experience.
>>
>> ---
>> i. "People" defined as people who are unregistered or published <100
>> edits.
>>
>>
>> --
>> Peter Pelberg (he/him)
>> Lead Product Manager, Editing Team
>> Wikimedia Foundation
>>
>>
>> On Mon, Mar 11, 2024 at 7:48 AM Paulo Santos Perneta <
>> paulospern...@gmail.com> wrote:
>>
>>> For 10 years or more, already, reliable sources have been mandatory in
>>> the Wikipedia in Portuguese, and any unsourced edit can and should be
>>> reverted and the user warned.
>>> Adding to that, since at least 2016, we use the abuse filters to block
>>> any edition lacking sources. Newbies like the one described by Romaine
>>> would receive a daunting red warning from the abuse filter system about the
>>> necessity of adding reliable sources in order for their edit to be saved -
>>> and the opportunity to go back and fix the problem. This has greatly
>>> improved things there, in that subject.
>>>
>>> Back in 2009, about 1 month after joining Wikipedia I found myself in a
>>> serious conflict with other, well established users, about a well sourced
>>> edit I wanted to add, which was being reverted by the veteran users in
>>> favour of unsourced (and false) information. At the time, I had to comply
>>> and swallow it, as the newbie I was. One year later, now with a reputation,
>>> I returned to the theme, reverted the whole thing and opened a public case
>>> there about falsification of information by said veteran user(s) - and that
>>> time it stood. This whole episode deeply marked me, and made absolutely
>>> clear that in Wikipedia there can be no tolerance for whatever lacks proper
>>> sources - something we actually often indulge in in paper encyclopedias, in
>>> my own experience. I'm very glad that the era of rampant tolerance with
>>> people adding unsourced content - something that was already against all
>>> good practices back in 2001 - is now a distant, sad memory. The quality of
>>> our Wikipedia skyrocketed since then, changing the paradigm from "Wikipedia
>>> is not reliable" to "Wikipedia is actually quite reliable, so much that I
>>> actually want to be there" all over the Lusophone world - and bringing new
>>> problems of its own. But that's undoubtedly the way to go, and it's sad it
>>> took so much time to actually implement what should have been there already
>>> from day 1.
>>>
>>> Best,
>>> Paulo
>>>
>>>
>>>
>>>
>>> Romaine Wiki  escreveu (quarta, 6/03/2024 à(s)
>>> 13:59):
>>>
 In the past days, 

[Wikimedia-l] Re: Mopping with the tap open

[Steven Walling]

How much does Edit Check decrease the total number of saved edits and
unique editors?

On Sat, Jun 8, 2024 at 7:07 PM Peter Pelberg  wrote:

> If those new users would have got a message in the Visual Editor during
>> the editing, a lot more contributions would be able to stay in Wikipedia,
>> less new contributors would get demotivated, and it would reduce the
>> workload of existing users who do the maintenance every day.
>>
> Romaine – and everyone here who resonated with what Romaine expressed
> above – I thought you might value knowing that a recent A/B test
>  of Edit
> Check  (the idea Benoît shared
> here
> )
> supports the assumptions you're making above and in this thread more
> broadly.
>
> Specifically, the A/B test showed:
> * People [i] shown the Reference Check are *2.2x* more likely to publish
> a new content edit that includes a reference and is constructive (not
> reverted within 48 hours).
> * The highest observed increase was on mobile where people are *4.2x*
> more likely to publish a constructive new content edit with a reference
> when Reference Check was shown
> * New content edit revert rate decreased by *8.6%* if Reference Check was
> available.
> * Contributors that are shown Reference Check and successfully save a
> non-reverted edit are *16%* more likely to return to make a non-reverted
> edit in their second month (31-60 days after).
>
> You can read the full report that Megan Neisler
>  prepared here: Reference
> Check AB Test Analysis
> .
>
> If anything you see brings questions/ideas
>  to mind, now is a
> wonderful time to share them. Reason: the Editing Team is
> actively planning  how
> to expand Edit Check and needs volunteer expertise to shape this experience.
>
> ---
> i. "People" defined as people who are unregistered or published <100 edits.
>
>
> --
> Peter Pelberg (he/him)
> Lead Product Manager, Editing Team
> Wikimedia Foundation
>
>
> On Mon, Mar 11, 2024 at 7:48 AM Paulo Santos Perneta <
> paulospern...@gmail.com> wrote:
>
>> For 10 years or more, already, reliable sources have been mandatory in
>> the Wikipedia in Portuguese, and any unsourced edit can and should be
>> reverted and the user warned.
>> Adding to that, since at least 2016, we use the abuse filters to block
>> any edition lacking sources. Newbies like the one described by Romaine
>> would receive a daunting red warning from the abuse filter system about the
>> necessity of adding reliable sources in order for their edit to be saved -
>> and the opportunity to go back and fix the problem. This has greatly
>> improved things there, in that subject.
>>
>> Back in 2009, about 1 month after joining Wikipedia I found myself in a
>> serious conflict with other, well established users, about a well sourced
>> edit I wanted to add, which was being reverted by the veteran users in
>> favour of unsourced (and false) information. At the time, I had to comply
>> and swallow it, as the newbie I was. One year later, now with a reputation,
>> I returned to the theme, reverted the whole thing and opened a public case
>> there about falsification of information by said veteran user(s) - and that
>> time it stood. This whole episode deeply marked me, and made absolutely
>> clear that in Wikipedia there can be no tolerance for whatever lacks proper
>> sources - something we actually often indulge in in paper encyclopedias, in
>> my own experience. I'm very glad that the era of rampant tolerance with
>> people adding unsourced content - something that was already against all
>> good practices back in 2001 - is now a distant, sad memory. The quality of
>> our Wikipedia skyrocketed since then, changing the paradigm from "Wikipedia
>> is not reliable" to "Wikipedia is actually quite reliable, so much that I
>> actually want to be there" all over the Lusophone world - and bringing new
>> problems of its own. But that's undoubtedly the way to go, and it's sad it
>> took so much time to actually implement what should have been there already
>> from day 1.
>>
>> Best,
>> Paulo
>>
>>
>>
>>
>> Romaine Wiki  escreveu (quarta, 6/03/2024 à(s)
>> 13:59):
>>
>>> In the past days, a new Wikipedia contributor edited Wikipedia and made
>>> a great contribution, except... This user added zero sources, and the
>>> article in what the edit was made was about a living person. So the
>>> verifiability is a problem and in conflict with the policy Biographies of
>>> living persons. This was just one example of thousands that have to be
>>> dealt with every day in Wikimedia. And every day the