Wikipedia:VisualEditor/A/B test

From Wikipedia, the free encyclopedia
Preliminary results are now available at meta:Research:VisualEditor's effect on newly registered editors/Results


This page documents the A/B test of the VisualEditor, which was launched to 50 percent of newly registered users on Monday, 24 June, with the intention being to discover how much of an impact it makes on editing rates when compared to editing via markup. Read below to find out more.


What we are doing[edit]

Within the next few days, the VisualEditor will be launched as the default editor for 50 percent of newly created accounts as part of an A/B test. The remaining 50 percent will continue to use the wikimarkup editor. The VisualEditor version deployed will have all of the features of the software currently deployed on enwiki, along with the ability to edit and add templates and use references. If you're unfamiliar with the current setup, you can try it out by going to your preferences and ticking the "Enable VisualEditor" box.

As with the existing opt-in deployment, the VisualEditor will only be enabled for the article and user namespaces; wikimarkup will still be required in others. While we realize that there are currently quite a few bugs with the VisualEditor, editors who have been testing the software have helped locate many of these for repair. The version as deployed on Tuesday should have the most crucial ones fixed, and if new ones come up we'll fix them as soon as possible. We will be monitoring the situation closely. Should major bugs lead to substantial problems, we retain the ability to disable the VisualEditor at any time.

Why we are doing it[edit]

The primary goal of the deployment is to get an estimate for what kind of conversion rate we can expect. At the moment, we know that very few people who create an account, and indeed not all of those who hit the 'edit' button, ever complete an edit or become members of the community. One of the reasons for this is that Wikipedia, unlike almost everywhere else on the Internet in 2013, expects users to learn a markup language to properly contribute. Accordingly we suspect that the VisualEditor will lead to far more new users being able to edit, which is undoubtedly a good thing. It'll increase the number of people editing Wikipedia.

One unknown with this increase is the impact it will have on existing workflows. All of these, both negative (handling vandalism) and positive (handling user requests for help, for example) are set up around the current conversion rate. As a community, we're not used to having a flood of newcomers, and there may be negative side effects of bringing a large number of newcomers in at once, even if it's ultimately a net positive. This test will help give us an idea of how many users we can expect with the VisualEditor, whether any increase causes problems, and if there's anything we can do to help mitigate those problems.

A secondary (but very much secondary) goal is to see what bugs come out with more sustained use of the VisualEditor so that we can implement necessary fixes before further rollout.

How people can help[edit]

We're going to have (hopefully!) quite a few new editors editing with the VisualEditor. If they are, their edits will be tagged "VisualEditor". There are undoubtedly going to be quite a few small, niggly bugs with the VisualEditor, even if the large ones are fixed, and new bugs may appear in the course of the VisualEditor being used. Accordingly, please be extra careful to assume good faith with the newcomers; if they've done something wrong it may have been completely unintentional and a flaw in the VisualEditor.

If you do spot the VisualEditor inserting errors into articles, feel free to revert the edit – just remember to explain to the newcomer why you've done so. Any bugs you find can be reported at the feedback page and will be dealt with as soon as possible by our developers. If you have any other comments or suggestions relating to the software, they can go there, too.