We need review of https://en.wikipedia.org/wiki/Wikipedia:Autoconfirmed_article_creation_trial/Landing_page_mock-up.
Should we ditch Puzzly? Are the button colors right? Anything else we should change?
kaldari | |
Aug 21 2017, 9:09 PM |
F9209497: Screen Shot 2017-08-29 at 11.46.52 AM.png | |
Aug 29 2017, 6:55 PM |
F9144020: IMG_4348.JPG | |
Aug 22 2017, 1:11 AM |
F9143623: create-warning.png | |
Aug 21 2017, 11:46 PM |
We need review of https://en.wikipedia.org/wiki/Wikipedia:Autoconfirmed_article_creation_trial/Landing_page_mock-up.
Should we ditch Puzzly? Are the button colors right? Anything else we should change?
Just had a chat with @kaldari
few notes
Persona : non auto confirmed user
Sorry, there is no article about this topic on Wikipedia
Options
Open question: Product to decide what behavior we want to encourage on this step. that will dictate the primary action to be promoted
Next steps for design > come up with screens with above based on the product call.
Actually, I lied to Nirzar. Users would not come directly to the landing page from redlinks. They would have to go to the redlink page and then click on the "Create" tab or the "Start the XXXX article" link. Sorry about the confusion.
Alright so based on new information, the user has shown intent to create an article. The new page will be a pause for warning page for this user.
Something like this
Where does 'Proceed anyways' (go forward) take them?
Article creation wizard, it doesn't say "article wizard" on purpose because people don't know what that is.
They have clicked "create" and then they want to still create so "proceed" is the action. article wizard can explain what article wizard is :)
@Nirzar, @DannyH: Here's a crazy idea, but what if we just redirected them straight to the Article wizard? I feel like if that's the main thing we're going to be directing users to (and the community feels strongly that all new articles should be created through the wizard anyway), why do we need the extra step of the landing page? I imagine the more steps we force the user to go through, the less likely they are to actually bother submitting their article. We can still have ArticleCreationWorkflow count how many people are redirected, we'll just be removing one step from the funnel. Thoughts?
just think of people who get disoriented looking at article wizard :)
we need to warn people before we push them into complicated workflows. that's not an ideal solution, the ideal solution is fixing the complicated workflows but that would be out of scope.
I agree with @kaldari. We have too many steps here. Let's go through them below. If I try to go to a 'redlink' page on enwiki right now, here's what I see:
Now if they click 'Create'/'Create source'/'Edit', they see another wall of text like
And now (if they are still with us), they might click through to create an article, they'll land up on Article wizard which is another huge wall of text: https://en.wikipedia.org/wiki/Wikipedia:Article_wizard
I think these are way too many steps for a newcomer to encounter when it's probably their first time creating an article. If we were the newcomer going through these steps, we'd probably give up on creating an article at this point.
Suggestions:
I'd personally favor #1.
Yeah, we talked about it on the ACTRIAL talk page, and we agreed that we're going to take them directly to Article Wizard.
That directly effects T173766: Instrument event-logging data for ArticleCreationWorkflow. What are we recording now instead of which link was clicked?
I think we're not recording anything? But let's wait until Ryan's back next week to make sure.
We will only be counting the actual redirects, probably via statsd. @MaxSem: Does statsd sound like the best solution for that? i.e. counting every time a non-autoconfirmed user is intercepted and redirected to the Article Wizard?
We should stick with the landing page and not go direct to the Article Wizard. The trouble with these discussions is that they are dominated by users involved in the AfC/NPP process. They generally have a hostile attitude to new users and content because they are exposed to the worst of it. Myself, I am more focussed on outreach and content creation. Typically these are events like editathons which recruit new users to add content about people such as women scientists. Such events stay well away from the AfC and its processes because they are a massive drag on getting things done. The focus in such events is getting content into main space and the sandbox is the preferred staging point.
As a fresh example, see Women in Red where an organiser from an "Australian Women Write Wiki group" appeals for help in getting some content out of draft space. I did this immediately because the subject was quite notable and belonged in our encyclopedia. Content that gets stuck in draft now risks deletion because there are several fanatics now deleting thousands of pages there without proper oversight or attention. The AfC/Draft process has been designed mainly to prevent the creation of content and to delete anything which isn't perfect. Its philosophy is a return to the failed idea of Nupedia which had a review/approval process and so utterly failed to produce much content.
So there is not a consensus in support of the AfC/Article Wizard approach -- there are numerous volunteer groups who do their best to avoid it. Please keep the generic landing page approach so that these different groups and views can be accommodated.
@Andrew_Davidson: What if we added a user sandbox button to the beginning of the Article Wizard for people who just want to start writing something?