Talk:Account Creation Improvement Project/Archive 1

From Outreach Wiki
Jump to navigation Jump to search

Bug 21318[edit]

Good idea. Might I recommend Bugzilla bug 21318 [1], as explained in this discussion. Minor improvement to make sure all new accounts have a welcome message accessible at all times (via their user talk page, but not a talkpage posting which turns that redlink blue), until they specifically say they don't want it any more. Could perhaps be improved on, but the basic point is: Mediawiki:Welcomecreation (much different on en.wp) is a message easily clicked past, and once it is, it's hard to retrieve. Rd232 16:51, 19 September 2010 (UTC)

Hi Rd232, thanks for letting everyone know about this feature request. I've put my email address on the list so I will get notified about any changes. --Frank Schulenburg 15:47, 20 September 2010 (UTC)

Suggestions[edit]

This is a great start! It's wonderful to get some data on what's happening after users create accounts. The edit-to-account creation ratio is quite a bit higher than I thought it would be. I don't think it's a foregone conclusion that users create accounts to edits. Anecdotally, users create accounts for a variety of reasons (e.g, I want to see what features I'd get with an account, I want to have one, I wanted one for reading). But given that 30% of our account creators end up editing within 10 days, it does seem like a good percentage create an account with the intent to edit. I also think it's interesting that the dewiki and enwiki have percentages that are pretty close even though dewiki has a simpler account creation path.

A few things I think are worth looking into:

  • What is the error rate for account creation? Part of this is already addressed in this section. But I think we need to distinguish between the different reasons users don't make it through the flow. Reasons include:
  • a) they leave due to an error (e.g., fail at the captcha and they get frustrated)
  • b) they leave because the form is too complicated (e.g., they get overwhelmed by the form, they have trouble finding the fields
  • c) they decide it's not for them.
b) and c) are difficult to distinguish, but we should be able to get a measurement of a)
  • What's the distribution of edits over time? I'd like to see how many of these users make their first edit within the same session, during the 1st day, 2nd day, etc. This will help us understand how much potential leverage we can have when these users are either communicated with or get some kind of help/pointers at different times. For example, finding out that most edits are made during the same session could suggest that we test different things right after the account is created (e.g., tutorials, directing users to topics, etc).
  • Where do they make their edits? It would be very interesting to see where these edits occur (e.g., Article vs. Talk).
  • Why do users create accounts? I'd like to do a survey of some users after account creation and ask why they created their account. We should also ask what their expectation are.

Howief 00:02, 21 September 2010 (UTC)

Hello, yeah! nice start. I have another concrete suggestion for these data:

  • What's the percentage of users who created a new account that eventually created their own user_page or user_talk page?
  • Distribution of time elapsed between both events.

--GlimmerPhoenix 14:29, 22 September 2010 (UTC)

Hi GlimmerPhoenix, that's an excellent idea. I'm currently working on some of the data Howie suggested (i.e. distribution of edits over time: how many new users make their first edit on the day of account creation) and the next thing I will look at is the percentage of users that create their own user page. Thanks a lot for your feedback! --Frank Schulenburg 16:25, 22 September 2010 (UTC)
Update: I've created a new section called How many new users make at least one edit in their user namespace shortly after account creation?. Right now it contains data only for the German Wikipedia, but I will add more numbers within the next few days… Thanks again for the good feedback. --Frank Schulenburg 16:37, 23 September 2010 (UTC)

Implementation[edit]

The user signup form is a big pet peeve of mine. It should be super super simple, like Twitter and other services, to create an account on Wikipedia. I have fiddled with the user signup and account code, and raised this issue on English Wikipedia [2]. If given some (small) support (mentorship? code review? feedback?), I might be able & willing to take on some interface design, coding & other implementation tasks. Without support, I'll just get frustrated, though. Aude 19:54, 21 September 2010 (UTC)

I agree. My suggestion is this:
a) Log in ----> 1) User name, 2) Password, 3) Forgotten your password? ----> You're now logged in
b) Create an account ----> 1) 1) Choose password and user name, 2) Links to the rules, 3) Captcha ---> 1) You've now created an account, 2) Links to the rules, 3) Links to suggestion of things to do.
Short, divided for different functions and easy to overview. Each step should have a progress bar at the top (you are on page 1 of 2, etc). No need to make it more difficult than that.//Hannibal 20:36, 21 September 2010 (UTC)

English Wikipedia ACC system[edit]

Please note that the large majority of people who request accounts via the alternative toolserver en:Wikipedia:Request an account - designed for people who cannot use the ordinary method, e.g. IP create blocked, or can't read the CAPTCHA - never actually edit at all. See for example the ones I have created myself. Chzz 00:18, 23 September 2010 (UTC)

On a related note, having created hundreds of accounts via the ACC system, I can say that more than 80% of the accounts I created were not used by the requesters. Having filled out the ACC form and confirming their email address, all it takes is to click on the link that will login to wp via their temporary password. I think the reason why people choose not to edit goes way beyond the account creation "trouble" and complication reason. Bejinhan 03:46, 23 September 2010 (UTC)

Prizes suggestion[edit]

Could there be a tiered-prized competition? For example, the prize that'll cost the most(paid trip to SF, laptop, etc.) is at the top and descending the ladder would be the lesser-costing prizes(t-shirt, movie, vouchers). That way, we could have more prizes. Furthermore, new users would have to compete for them. It's fun and will make them want to aim for the "top" prize. Bejinhan 03:50, 23 September 2010 (UTC)

Hindi Wikipedia[edit]

Hi All, Barry Newstead asked on the Wikimedia India mailinglist, what the reason for the low account creation to edit ratio for the Hindi Wikipedia might be. The thread is very interesting and gives some possible answers: http://lists.wikimedia.org/pipermail/wikimediaindia-l/2010-September/000979.html --Frank Schulenburg 13:37, 23 September 2010 (UTC)

Survey after account creation[edit]

Hi All,

I would like to follow up on Howie's suggestion to start a survey for users who just created an account on Wikipedia. Our goal would be to get a better idea of why people signed up / of what they expected.

What are the questions we could ask? Please add yours to the subsection below… --Frank Schulenburg 23:38, 23 September 2010 (UTC)

Draft: Questionnaire[edit]

English
Why did you create your account?
  • I wanted to be part of Wikipedia [why?]
  • I was just curious about what would happen if I created an account
  • I thought an account is required to edit articles
  • I thought I would get some extra features [what features?]
  • I usually register for websites I like
  • I tried to start an article, but I couldn't do that without a user account
  • I wanted to upload images
  • Other: …
Deutsch
Warum hast Du / haben Sie gerade ein Benutzerkonto angelegt?
  • Ich möchte gerne ein Teil von Wikipedia sein
  • Ich war nur neugierig, was passieren würde, wenn ich ein Benutzerkonto anlege
  • Ich dachte, man braucht ein Benutzerkonto, um Artikel zu bearbeiten
  • Ich dachte, dass ich besondere Softwarefeatures nur nutzen kann, wenn ich ein Benutzerkonto anlege
  • Ich registriere mich üblicherweise auf Webseiten, die ich mag
  • Ich habe versucht, einen neuen Artikel anzulegen, aber das geht nicht ohne Benutzerkonto
  • Ich wollte ein Bild auf Wikipedia hochladen
  • Sonstiges: …
Spanish
¿Por qué te has registrado?
  • Quería formar parte de Wikipedia
  • Quería saber qué sucedería si me registraba
  • Creía que era necesario para editar artículos
  • Pensaba que tendría acceso a algunas opciones adicionales
  • Me suelo registrar en aquellos sitios web que me gustan
  • Intenté crear un artículo pero no podía hacerlo a menos que me registrase
  • Quería subir imágenes
  • Otro: ...
Bengali
আপনি কেন আপনার অ্যাকাউন্টটি তৈরি করেছেন?
  • আমি উইকিপিডিয়ায় অংশ নিতে চাই [কেন?]
  • অ্যাকাউন্ট তৈরি করলে কি ঘটে শুধুমাত্র তা জানতে আমি কৌতুহলী
  • আমি মনে করেছিলাম নিবন্ধ সম্পাদনা করার জন্য অ্যাকাউন্ট তৈরি করার প্রয়োজন
  • আমি মনে করেছিলাম আমি হয়তো কিছু অতিরিক্ত সুবিধা পাবো [কি সুবিধা?]
  • যে ওয়েবসাইটটি আমার পছন্দ হয় আমি সাধারণত সেখানে রেজিস্ট্রেশন করি
  • আমি একটি নিবন্ধ শুরু করার চেষ্টা করেছিলাম, কিন্তু ব্যবহারকারী অ্যাকাউন্ট না থাকায় আমি তা পারিনি
  • আমি ছবি আপলোড করতে চেয়েছিলাম
  • অন্যান্য: …
Bahasa Melayu
Mengapa anda membuka akaun?
  • Saya ingin menjadi sebahagian Wikipedia [Kenapa?]
  • Saya ingin tahu tentang apa yang akan berlaku jika saya membuka akaun
  • Saya anggap bahawa akaun diperlukan untuk menyunting rencana
  • Saya fikir saya akan mendapatkan ciri-ciri tambahan [Apa ciri-ciri tambahan?]
  • Saya biasanya membuka akaun di laman web yang saya suka
  • Saya cuba cipta artikel, tetapi tidak boleh tanpa akaun pengguna
  • Saya mahu muat naik fail
  • Lain-lain: ...
French
Pourquoi vous êtes-vous inscrit(e) ?
  • Je voulais faire partie de Wikipédia [pourquoi ?]
  • J'étais curieux(se) de savoir ce qui se passerait si je m'inscrivais
  • Je croyais que je devais m'inscrire pour éditer les articles
  • Je croyais que j'aurais accès à plus d'options [quelles options ?]
  • J'ai l'habitude de m'inscrire sur les sites que j'aime
  • Je voulais créer un article, mais je ne pouvais pas le faire sans être inscrit(e)
  • Je voulais télécharger des images sur Wikipédia
  • Autre : …


Polish
Dlaczego utworzyłeś(aś) własne konto?
  • Chciałem/am być częścią Wikipedii [dlaczego?]
  • Byłem/am ciekawy/a co się stanie, gdy założę konto.
  • Myślałem/am, że utworzenie konta jest potrzebne do edytowania haseł.
  • Myślałem/am, że dzięki temu będę dysponować dodatkowymi możliwościami [jakimi możliwościami?]
  • Zwykle rejestruje się na stronach, które lubię.
  • Starałem się utworzyć hasło, ale nie mogłem tego zrobić, nie posiadając konta.
  • Chciałem/am przesłać grafiki.
  • Z innych powodów: …

Discussion[edit]

The obvious difficulty being, of course, than once people have given up, it is very difficult to ask them why. And they're the once we're really interested in, I think. Sorry if I sound negative...it is a challenge, for sure. Surveying people that did stay is no doubt a laudable idea too, but I think we need to concentrate on people who don't. But as long as we bear in mind the inherent slant in responses - ie that the ones who respond are the type who will, whereas those who won't are prob the target we need need to address...then yes, by all means.

Should we have an option for 'I was blocked from editing, and so I created this new account'? </cynic> Chzz 13:20, 24 September 2010 (UTC)

Hey Chzz, actually, we have data about people who left. Some time ago, Howie conducted the "Former Contributor Survey". The goal of the above survey is to find out why people create a user account. What are their expectations? And it doesn't cover only the people who start editing, but everbody who successfully creates a new user account. --Frank Schulenburg 13:34, 24 September 2010 (UTC)
Perhaps we can introduce this question in the registration form, below the password textbox, so, it is mandatory to reply. We can use a drop-down list, similar to the "How did you know about us?" question in another websites. Regards. Emijrp 21:00, 24 September 2010 (UTC)
Yes, making it mandatory is a great idea. We are now working on it and I hope it'll be up and running by next week. --Frank Schulenburg 21:58, 24 September 2010 (UTC)
I think we can simplify and reduce the list to avoid random selections by lazy users (whoa! a huge list!). Also, it would be nice to have a blank option to write whatever you want. Emijrp 22:23, 24 September 2010 (UTC)

I think a large number of users join to create one specific article, about themselves/their company/their band. Can we somehow ask that? Maybe one of the 'reasons' choices - "I wanted to create a specific article about something I know about" or, IDK, split it to company/self/whatever. Just a thought. Chzz 01:24, 26 September 2010 (UTC)

  • I would be careful with this. What about if these questions get them off? And if they don't want to respond? . I would firstly like to see something like "Would you like to take a survey specially for our newest users" or something like that. --Diego Grez 23:29, 27 September 2010 (UTC)
  • How about "I was referred to by another wiki" as one of the questions? TeleComNasSprVen 09:46, 28 September 2010 (UTC)
Diego Grez raises a very valid point, above. Sometimes, I come across a website that requires registration to do what I want; if the sign-up process takes more than about 10 seconds, I don't bother. I mean...for example, I read a news article, and want to comment, so click 'sign up' Sometimes, it asks me to fill in a page full of questions, in which case, I give up on the idea. I think that is quite common. If the survey could stop people bothering to join, that is a big problem. Chzz 23:23, 28 September 2010 (UTC)
How about is we make the survey optional? For example, after a new user registers, a "banner" would appear on his talk page asking him to take the survey. That way, he doesn't have to take the survey. Or, there could be a bot sending messages to newly registered users asking them to participate in the survey. It'll give the newbies the option of whether to take the survey or not. Bejinhan 03:26, 29 September 2010 (UTC)
The banner idea sounds so Wikia-istic :-P I've seen that there, and it is really annoying. What I would like to see is: On the screen that says "Account was created successfully" or something like that (not sure where is the MediaWiki page for it, but I know it is editable), a link to the survey, with something that encourages the new editors to take it, like... "Want to help improve Wikipedia? Take this quick survey!". --Diego Grez 23:39, 29 September 2010 (UTC)
Found it: MediaWiki:Welcomecreation. The same box used for the new messages would be superb! --Diego Grez 23:40, 29 September 2010 (UTC)
I get your point. Although, it's only one question with a very limited number of options to answer (in order to not overwhelm the new user). How about if we implemented well visible "skip this survey" button? Would that work for you? --Frank Schulenburg 13:19, 29 September 2010 (UTC)
Works for me. Diego Grez 23:34, 29 September 2010 (UTC)

Posted French survey. —  Waterfox  14:54, 6 November 2010 (UTC)

Wikipedia only?[edit]

Why are only Wikipedia projects mentioned? Why not the other Wikimedia projects, which are too numerous to mention here? Is this an oversight or a deliberate decision? Thanks --AlexandrDmitri 13:16, 28 September 2010 (UTC)

Actually, it's not Wikipedia only. I've added data for the English Wikinews… and it's quite different. Do you have any idea why 100% of all new users on Wikinews start editing within the first 10 days? --Frank Schulenburg 18:48, 28 September 2010 (UTC)
One would expect 100% of all new users anywhere to start editing within the first day, let alone 10 days. Editing being the only reason one would need to create an account, after all...
I imagine the difference on en.wikipedia reflects those users who register an account, then can't use it because their username was overzealously blocked, or tried to edit a page but were denied by the abuse filter, or even just clicked on "edit", read all the crap about how their changes aren't even going to be visible until approved by a self-selected group of moderators and thought "sod that I can't be bothered". 81.158.249.29 16:16, 29 September 2010 (UTC)

wow[edit]

That's a hell of a lot of brainstorming and discussion to address a problem that could be solved in 3 easy steps:

(1) rewrite en.wikipedia's signup screen messages so there's 3 lines of text not 3 pages
(2) tweak the configuration of the dozen filter/blacklist/patrol extensions you have installed so that they are less restrictive
(3) generally stop treating new and unregistered users like shit
81.158.249.29 16:13, 29 September 2010 (UTC)

"Percentage of accounts that edit" metric is flawed[edit]

because of unified login. I have an account on the Hindi Wikipedia because I once followed a link there. The account was created for me without my permission, or even notification. Of course, I've never edited there, because I don't speak Hindi. Same goes for a lot of other wikis. Gurch 19:25, 29 September 2010 (UTC)

Yes, you're right. People who create accounts as part of the unified login are included in the statistics. --Frank Schulenburg 21:49, 29 September 2010 (UTC)

Is 30% good? A quick comparison to Facebook account creation numbers[edit]

I came across some numbers on Facebook account creation. As part of a Facebook co-authored paper studying the determinants of newbie contribution, they observed the following in establishing their dataset:

". . .we selected a cohort of all 254,603 users who registered for Facebook on a randomly chosen weekday in March 2008. The cohort includes members from 207 countries with 24% registering from the United States. We collected all model variables during the newcomers’ first two weeks and then predicted their content sharing in the subsequent three months. To mitigate the effects of fake accounts and members who never returned, any users who did not log in at least once during their third month were removed from the data, leaving a set of 140,292 newcomers."

So of the 254,603 users that created accounts during this random weekday, 140,292 came back at least once during the third month. I'm assuming that this means that if a user logs back in during the second month, but not in the third, they are not counted as part of the 140,292. That means that there is a 44.6% drop off between account creation and the third month for this cohort of users.

I'll try to get clarification on this number from the authors. I know this is not apples to apples, but it gives us something to compare.

Howief 23:24, 1 October 2010 (UTC)

Screencast video[edit]

I've started a scratchpad at w:Wikipedia:WikiProject Screencast/Scripts/Newuseraccount. Anyone is welcome to help script it.

I'm not sure which to concentrate on first, (or if we even need to make both? Possibly just having 1, is enough):

  1. From the angle of User has just registered: Show them what they can now do with their account, what new features are available. (Possibly to be embedded at w:MediaWiki:Welcomecreation.)
  2. From the angle of User is currently unregistered: Show them why they might want an account.

Feedback either here, or directly at the script, is welcome. Quiddity 04:32, 9 October 2010 (UTC)

I'm concerned about the new pages part in the "Draft" section. Many new users do not know much about creating new articles and tend to create articles leaning towards promo or non-notability. Perhaps before we talk to them about creating new articles, we should give them the basic rules about article creation. While wiki-formatting might also seem slightly minor, many new users who immediately jump into article creation tend not to know much about them. Maybe the basic of it could be included? Ex, use of headers, reference formatting, etc. Bejinhan 03:41, 18 October 2010 (UTC)
Hi, thanks for your thoughts. Essentially, I was trying to avoid duplicating the other screencasts (See w:Wikipedia:WikiProject Screencast/Screencast Gallery, particularly the newest batch at the top by Sage Ross). For all screencasts, it is consistently difficult to decide which elements need to be repeated, and which can simply be mentioned as "explained in our other video at ..." type-thing (E.g. See the File:Boldness and links tutorial.ogv video, which starts by pointing towards the "Help:Sandbox tutorial" page&video).
This particular screencast, is partially aiming to be a rapid overview (because we simply can't cover too many points in detail without overwhelming), and it will probably point towards the other videos, at the end. It's intended to whet the viewer's appetite, for learning more. More feedback welcome :) Quiddity 05:09, 22 October 2010 (UTC)

Related Work[edit]

Hi All, thought you'd be interested in a paper by the GroupLens research group called "Wikipedians Are Born, Not Made". Particularly in their sub-categorization (and classification) of editors into 'wikipedian editors' and 'non wikipedian editors' (read: casual editors). I feel like this work (as well as yours) suggests tons (!) of opportunities early in an new user/editor's participation that could potentially have high and lasting impact on their future behavior and interaction on the sites. Would love to discuss further, but in the mean time, link is here.

Looks very interesting. I'll put it on my reading list.//Hannibal 10:00, 22 October 2010 (UTC)

Design[edit]

Please look also how the different steps of the account creation path are designed. Are they full of warning signs? A lot to read? --92.204.92.215 22:34, 27 October 2010 (UTC)

Latest development?[edit]

Things seem to be pretty quiet in here lately. May I know the latest development? Is there a schedule for this initiative? Bejinhan 12:15, 13 November 2010 (UTC)

Hi Bejinhan, we are currently exploring the option of having someone working on this for half a year. I'll get back to you once we know further details. --Frank Schulenburg 14:39, 16 November 2010 (UTC)

Why do people log in or choose against it?[edit]

Another interesting question would be, if there are many users who created an account, but hardly ever use it. I'm one of those. Since years, I contribute on a minor scale, basically just when I find something to add or correct when I'm reading. I don't log in for those edits. However, a couple of years ago, I had the idea of starting acticles about a topic, which was missing in Wikipedia, and I made an account. These articles immediately got deleted (within one hour). I read the discussion board and found out, that some (not all) of the administrators believed, the topic was of no public relevance, although it clearly was. (Now, there's an acticle on the topic.) They also believed, the article was too short. (It was not even one hour old.) I tried to clarify my position on the discussion board, but all I got were uninformed, harsh, ignorant judgements. My desire to become a part of wikipedia immediately vanished. I guess, that's not an uncommon experience for a new user, and it would be worthy to investigate that, too! 23:43, 16 January 2011 (UTC)

Thanks for your input. From what I know, your experience is not unique. However, the problem is not easy to solve, since it entails changing the entire culture of Wikipedia. But we do think about this question, a lot. and we agree that is something that we will have to do. So if you have any suggestions, please make them here. Best wishes, Hannibal 15:09, 19 January 2011 (UTC)

First results from the Account creation survey[edit]

Final survey results - English Wikipedia
Final survey results - German Wikipedia
Preliminary survey results - Spanish Wikipedia

Hello,

A couple of days ago, I asked admins on 7 language versions of Wikipedia (English, German, French, Spanish, Swedish, Bengali and Bahasa Melayu) to put a link to a very short survey on the page that new users "land" after they have created their new accounts. The survey asks them about what they expected when they created the account, or in essence, why create an account on Wikipedia?.

We intend to keep this survey running until February 4th and then analyse the results. This will be a helpful step in understanding what we need to change during the upcoming change in the account creation process.

But let's not wait until the middle of February to look at the results. Here is a small sneak peak of what people have chosen (for the full list of alternatives, see above):

  • out of 592 respondents, 67,4% on English Wikipedia say that the reason they created their account was... "I wanted to be part of Wikipedia".
  • this is, at 69,2%, also the most common answer on German Wikipedia
  • ... and on Bengali Wikipedia (75,5%)
  • ... in fact it's the most common answer across the board
  • the other two common alternatives are "I usually register for websites I like" (24%) and "I thought an account is required to edit articles" (26%)
  • even on Wikipedias where you don't have to have accounts to create articles, people assume that they do need accounts to do this

Some people have posted things in the comments field. Here are a few comments that are typical:

  • "I don't want to edit articles anonymously."
  • "Class assignment"
  • "I LOVE WIKIPEDIA"
  • and...
  • "Dunno. I was bored."

Right now, we have 1024 responses to the survey. And now, 1028. And in a few minutes even more. With the service we use (SurveyMonkey), it's really easy to track what people have chosen and said. After the 4th, I will do a quick analysis of the results, and hopefully have the results ready on February 7th. By then we should have many more responses, which means that the answers we get will be a very helpful guide for us. If you want to help change the process without doing the survey, you can help out in several ways, including by doing research or suggesting features of the new process. See the main page for the project for more information.

NB: The survey has not yet been launched on French and Bahasa Melayu. If you are an admin on those language versions, please contact me. And we would like to do the survey in even more languages, so post your translations of the survey on this page, and I will put them up along the others.

Finally, please sign up here to become part of the team that will change the account creation process - and spread the word about this. Our goal is to have changed the process by June and during that time increase the numbers of new users that start and continue to edit Wikipedia.

Thank you for reading this.

Best wishes//Hannibal 15:36, 26 January 2011 (UTC)

New Userpages for Existing Editors[edit]

During the usability initiative tests, we found that user accounts would get repeatedly reverted for the edits these new users would make. We also found that they hardly noticed when they got messages informing them of these reversions. While this wasn't the aspect of wikipedia editing we were testing at the time, we all felt kind of bad that some of these users who made their first edits would go home later that day and find their edits were reverted, so I wrote a quick note on the test user's user pages explaining who they were and the nature of the account. The results were incredible; pretty much none of the edits were reverted afterwards, and, in fact, some editors took the initiative to take the intent of the edit and work it in to the page to meet policies. How useful would that be for all people? So, my proposal is the following changes:

  • Creating a default userpage for a user after they've made 10 or so edits (assuming they don't have a userpage yet) explaining that they are new, they've been a wikipedian since (whenever), made (n) edits, clearly have an intent on being a wikipedian but might need someone to show them the ropes.
    • Creating a default userpage for a user after they've made 10 or so edits" IMO not only "after - newbies are able to make own user page by inputbox in MediaWiki:Welcomecreation - an example in pl wiki: pl:Wikipedysta:Jankowski zbigniew + more here + welcoming by bots in other wikis Przykuta 20:27, 28 January 2011 (UTC)
  • Notifying users via email when they get new messages, something that they can turn off, but have this on for new accounts by default so they know that they have a message in the system.
    • Uhm, but it is closed in preferences of big wikis :/ Przykuta 20:27, 28 January 2011 (UTC)
  • Some version of the "don't bite the newbies" template that lets people know the user they're about to message has only made a low number of edits, they're new, and to be helpful if they can be. Also work in conjunction with developers of tools like Twinkle so that users have to take extra measures to revert a "first edit" or "first 5" edit.

I think we can test the top and bottom ideas independently.

Second survey: motivation for creating an account[edit]

Hi all,

We would like to start a second survey on all Wikipedia language versions, where the results of the first survey (see above) are stable. This time, we would like to learn more about what motivates people most to create an account. Here is what the second survey question and the answer choices look like:

Question Congratulations! You just created your own Wikipedia user account. If you had to create a user account again, which of the following statements would convince you more?

Answer choices

  • Having a user account under my real name means that everybody will know how I helped improve Wikipedia.
  • Having a user account will enable me to customize the appearance and behavior of Wikipedia to my own preference.
  • Having a user account means that I can follow all my favorite topics on Wikipedia.
  • Having a user account means that I will get my own user page on Wikipedia. Other users will be able to send me messages.
  • Having a user account means that my contributions to Wikipedia are not connected to an IP-address, but to a nickname that can't be tracked back to my physical address.
  • Other (please specify)

So far, we have only a German translation for this new survey. We started it this morning and got some first results: 65,4% of the respondents stated that Having a user account means that I can follow all my favorite topics on Wikipedia would motivate them most. Whereas only 30,8% state that Having a user account will enable me to customize the appearance and behavior of Wikipedia to my own preference is important to them.

Again: we won't start the new survey for language versions where only a few people filled out the first survey (e.g. only 35 people filled out the Swedish version so far). But you could help us a lot by translating the question and answer choices above.

Here are the numbers for the first survey (English and German being completed):

  • Spanish: 640 responses (stable)
  • Bengali: 86 responses (not stable yet)
  • Swedish: 35 responses (not stable yet)
  • Polish: 34 responses (not stable yet)
  • French: 0 responses (not started)
  • Bahasa Melayu: 0 responses (not started)

Thanks a lot in advance for your help --Frank Schulenburg 20:48, 29 January 2011 (UTC)

After consulting with Guillom, I shortened the answers a little. The present survey looks like this.//Hannibal 19:04, 31 January 2011 (UTC)

Spanish translation[edit]

Question - Pregunta ¡Felicitaciones! Acabas de crear tu propia cuenta de usuario de Wikipedia. Si tuvieras que crear una cuenta de usuario de nuevo, ¿cual de los siguientes argumentos te convencería más?

Answer choices - Opciones de respuesta

  • Tener una cuenta de usuario con mi nombre real significa que todos sabrán como ayudé a mejorar Wikipedia.
  • Tener una cuenta de usuario me permitirá personalizar la apariencia y comportamiento de Wikipedia a mi gusto.
  • Tener una cuenta de usuario significa que puedo seguir todos mis temas favoritos en Wikipedia.
  • Tener una cuenta de usuario significa que obtendré mi propia página de usuario en Wikipedia. Otros usuarios podrán enviarme mensajes.
  • Tener una cuenta de usuario significa que mis contribuciones a Wikipedia no serán conectadas a una dirección IP, sino a un apodo que no puede apuntar a mi dirección física.
  • Otra (por favor especifique)

Diego Grez 21:36, 29 January 2011 (UTC)

More survey results[edit]

After roughly 1200 responses

Hello,

Here are a few more results. This time it's the second survey on English Wikipedia (see above). Click on the image to make it bigger.//Hannibal 21:53, 2 February 2011 (UTC)

The results of the Account Creation surveys, part 3[edit]

Hello,

This is the third report on the surveys for new users that we put on a few language versions of Wikipedia. You can read more about the first one above. But here are the first analysis of what we have. Later on, we will let you know more in detail what people wrote in the freetext field (but that information will take a short while to analyse)

Before anything else, though, I want to thank everybody who have helped out. I hope that more people will join the Account Creation Improvement Project now that you see how much fun we have and what wonderful results we get.

Now, let's get down to the numbers of the surveys.

How many people responded to the surveys?[edit]

In order from the fewest to the most responses:

  • Bengali - 112
  • Swedish - 155
  • German version 1 - 255
  • Polish - 359
  • Spanish version 1 - 881
  • German version 2 - 932
  • Spanish version 2 - 1050
  • English version 1 - 1412
  • English version 2 - 1200 (roughly[1])
  • English version 3 - 3189

(The French and the Bahasa Melaya was not started despite the surveys having been translated.)

Grand total of respondents, all languages

9545

In other words, we have circa 9500 people responding to our surveys. This is a significant number of new users that took the time to answer surveys just because they thought it would improve Wikipedia.

What did people answer?[edit]

For the two surveys we have, all languages had the same statements and alternatives (see above). So for instance, answer 3 is always the same. Also, people could choose several answers and fill in a text field (except for a while with the English version of the survey).

Pretty soon it turned out that the results were stable, so the most common answers didn't change much across time or language versions.

For each cell there are two numbers, eg. 71.8% (79). The first number denotes the percentage of all who took the survey who chose that alternative. The other is the total number of people who chose that alternative.

Survey 1[edit]

Language version Alternative 1
I wanted to be part of Wikipedia
Alternative 2
I was just curious about what would happen if I created an account
Alternative 3
I thought an account is required to edit articles
Alternative 4
I thought I would get some extra features
Alternative 5
I usually register for websites I like
Alternative 6
I tried to start an article, but I couldn't do that without a user account[2]
Alternative 7
I wanted to upload images
Alternative 8: Other People responding
Bengali 71.8% (79) 24.5% (27) 40.0% (44) 25.5% (28) 30.9% (34) 24.5% (27) 19.1 (21) 15.5% (17) 112
Swedish 60.4% (93) 17.5% (27) 21.4% (33) 9.1% (14) 24.7% (38) 22.7% (35) 9.1% (14) 11.7% (18) 155
German 66.7% (168) 6.7% (17) 28.6% (72) 6.7% (17) 17.5% (41) 15.9% (40) 3.6% (9) 15.9% (40) 255
Polish 45.5% (161) 13.8% (49) 17.8% (63) 20.1% (58) 16.4% (58) 30.2% (107) 7.9% (28) 13.0% (46) 359
Spanish 71.6% (618) 12.2% (105) 28.3% (244) 21.8% (188) 35.5% (306) 15.8% (136) 9.0% (78) 11.0% (95) 881
English 68.1% (952) 18.0% (252) 27.1% (379) 22.8% (319) 23.9% (334) 18.9% (264) 15.6% (218) 16.1% (225) 1412
Mean
(rounded off to two decimals)
64,02% 15,45% 27,77% 17,67% 24,82% 21,33% 10,72% 13,87% Total: 3174

Survey 2[edit]

Language version Alternative 1
Everybody will know that I helped improve Wikipedia
Alternative 2
It will enable me to customize the appearence and behavior of Wikipedia to my own preferences
Alternative 3
I will get my own user page on Wikipedia. Other users will be able to send me messages.
Alternative 4
I can follow all my favorite topics on Wikipedia.
Alternative 5
My contributions to Wikipedia are not connected to an IP-address, but to a nickname that can't be tracked back to my physical address.
Other/Comments People responding
German 45.9% (406) 30.3% (268) 24.7% (219) 32.0% (283) 53.0% (469) (101)[3] 932
Spanish 34.3% (347) 31.8% (322) 60.8% (615) 36.5% (369) 27.7% (280) (129)[4] 1050
English variant 1[5] 21% (244) 13% (151) 13% (151) 36% (408) 15% (175) 1% (13) 1200
English variant 2[6] 45.16% (1459) 42.25% (1365) 41.91% (1354) 60.04% (1940) 38.13% (1232) 8.08% (261) 3189
Total 36,59% 29,34% 35,10% 41,13% 33,46% 6371

Interpreting the results[edit]

What conclusions can we draw from these survey results? I think there are a number of conclusions that are interesting. You are very welcome to add your own conclusions below. Be bold.

  • The number of participants over a short period (volunteered by clicking on an extra link after creating an account) makes the numbers relatively significant. Of course a longer survey would mean more certainty, but the stable results across the survey period suggest that we would get a variation of these results. But I am very happy that this many took the time to answer the questions we put to them. Perhaps the short surveys (only one question at a time) is key here. Also, we can very soon get big groups of people to answer our surveys, which I feel we should use in the future.
  • The most common answer across all language versions we surveyed was "I want to be part of Wikipedia". In fact, the stable results on that point was the reason why we did another survey to understand what "part of Wikipedia" meant.
  • "I thought an account is required to edit articles" and "I usually register for websites I like" are the next most common answers. "Usually register" probably means that they do not intend to edit very much, while the other is the opposite.
  • Not many create accounts because they want to upload pictures.
  • A fair share of people thought they needed accounts to create articles (which is true for some but not all Wikipedias). On German and Spanish Wikipedia only about 15% thought that.
  • About a sixth of the new users thought they would get some more features. Whether they know about the features or not is not certain. That's why we added the alternative in the second survey about the watchlists ("follow articles"). On Swedish and German Wikipedia, this was not a big deal.
  • Following favorite articles on Wikipedia was the most common answer in the second survey, with the notable exceptions of German Wikipedia where anonymity is a bigger reason, and Spanish Wikipedia where getting a user page and being able to send/receive messages are the most important features.
  • Not many Germans were curious about what would happen when they created their accounts. Maybe they had enough information already.
  • When forced to choose one alternative, newcomers to English Wikipedia think that being able to follow articles is the most important reason to create an account. That is also important when they can choose.
  • When not forced to choose one alternative, newcomers to English Wikipedia more than the others that we surveyed like to tinker with the look and feel of Wikipedia.

Notes[edit]

  1. Due to a mistake, the specific result is swimming around somewhere on the LimeSurvey, but we have a screenshot of the results, see above. We are trying to get it back, though)
  2. In many language versions you can create articles without accounts, so any answers here are significant of what the new users think they need.
  3. This survey does not have the alternative "other". This cell instead measures the number of people who have filled in the textbox "other"
  4. This survey does not have the alternative "other". This cell instead measures the number of people who have filled in the textbox "other"
  5. With only one answer possible
  6. With several answers possible

Final thoughts[edit]

As always, feel free to add your own thoughts and comments. We are always open to suggestions and ideas. If you think we have missed an important part (should we ask about their genders or ages? should we ask about what they would change in the account creation process?), please make sure we get it by adding it below.

Next steps in the Project are to a) create a technical solution for testing new ideas and b) creating the pages that we want to test. Want to be a part of this? Great! Sign up above or by emailing lennart(at).svgwikimedia.org - and we'll contact you with more details.

Thanks for your time. Best wishes, Hannibal (talk) 15:20, 9 February 2011 (UTC)

If someone tries to upload pictures on Swedish Wikipedia they get immediately redirected to Wikimedia Commons, which might disrupt the numbers. The Swedish users who first and foremost want to upload pictures will probably create their first account on Commons, hence the fairly low responses with that reason on this survey from the Swedish users. --Ainali (talk) 18:40, 9 February 2011 (UTC)
I like to share my thoughts about Bengali Wikipedia. Although the survey got 112 participants, and it's a low number, but it supposed to be more fewer than 112. Because the link to the survey was on the site notice (wrongfully) for few hours, and was also announced in the mailing list. So I guess, in overall, not all of them are new users of Bengali Wikipedia. Regards, — Tanvir • 06:10, 10 February 2011 (UTC)

I've added a pledge[edit]

Pledge has been deleted by Prodego. 146.201.160.104 00:35, 30 March 2011 (UTC)

A consideration specific to English Wikipedia - WikiProjects[edit]

The Welcome template on the English Wikipedia should encourage people to join the WikiProjects related to their interests. I think some other projects also have equivalent concepts. Kaldari (talk) 22:23, 16 February 2011 (UTC)

Picking up on Lennart's question about gender[edit]

Hi Lennart! Congratulations on making progress on this project: it's really important, and I'm glad it's moving forward.

I spent much of yesterday reading Recent Changes, New User Contributions and New Pages, and I found myself sympathizing with the editors who do Recent Changes patrol. I can understand why they are sometimes impatient with new people, and feel like they don't have time to properly welcome and support them --- especially when it's not always easy to distinguish who's interested in editing seriously, versus who's just playing around or has malicious intent. It's particularly hard to assess new editors' motivations and likelihood of sticking around productively when they don't have a userpage or their userpage doesn't have much information on it.

I think there are two related problems here: i) Recent Changes patrollers are working quickly, and don't necessarily have the time or inclination to carefully assess new editors' intent, particularly since it's quite hard to do. Therefore, they may bias towards deleting/reverting. And ii) editors who would be willing to help coach/support brand new editors, may have difficulty finding new editors who show signs of promise, and/or who the experienced editor is particularly suited to help. The gist of the problem overall: I think it is too hard, currently, for experienced editors to distinguish between new editors who show promise, and those who don't.

(And by the way: to the extent that what I've written here is just plain wrong ---like, I am missing some important piece of information--, anyone here should feel free to tell me that. I am myself not a vandalism patroller, nor am I particularly experienced with tools like Twinkle. So I may be misunderstanding things that are obvious to others who've got more direct experience.)

On the assumption that my analysis is at least partially correct, I'd say that it'd be a service to all editors, if we could persuade new editors to offer at least some basic information about themselves. Premise being, even the act of providing the information would help us understand the new editor is serious (probably not a vandal, probably not a prankster), and the information that they give us might help us coach/orient/support them.

To that end, I think it would be helpful for us to ask new editors to (optionally) tell us their gender, their age, the country they live in, and perhaps a little about their interests. We might also ask them to self-identify if they have expertise in a particular subject-matter. All that information would help Recent Changes Patrollers identify new editors who are non-vandals and non-pranksters, and it would also enable experienced editors find people they could usefully coach and support. For example, experienced female Wikipedians might want to coach other women; Indians might want to coach Indians, and so forth.

More speculatively: I have also been thinking about whether we might want to encourage the use of userboxes or templates for new editor userpages, identifying those editors as new (and therefore calling for special leniency and encouragement), and possibly identifying them as members of groups that are underrepresented on Wikipedia (ie., women, editors outside Europe and North America, older editors, etc.). Userboxes/templates like that might already exist, in which case I think it might be a service to automate placing 'new editor' labels on new userpages, and to make 'underrepresented group' labels easier for new people to find and put on their pages.

I'm writing this super-fast so forgive me if I've missed obvious stuff :-)

Thanks Sue Gardner (talk) 23:23, 20 February 2011 (UTC)

There are some very interesting points here, Sue. User pages, userboxes and templates, plus categories, which I think you missed, can all help identify good newcomers. But I believe that one of the things we will have to work on more (and note that I am in constant awe of people doing vandalism patrol) is to transform bad editors into good editors. If you would have looked at some of my first edits, there would have been no way to know if I were a bad editor or a good one :-) But some personal messages on my talk page served two functions: a) they let me know what I did was not cool at the same time as they taught me how to do it better, and b) let the others know who I was, which led me to... this... And that took time. But I am convinced that we need to take that time to nurture good editors. After all, have not my history after those first edits made up for those bad edits, and more?
However, to not get too far off the Account Creation stuff: let's start thinking of more technical ways to handhold the new editors through their first 100 edits, that we can start when they create their accounts.
One initiative that I like is to send newcomers emails, if they give us their email adresses. But what should the emails look like? Of course, we should be careful so they need to be a little personalized. Perhaps they are based on the same kind of algorithms that make up Suggestbot? Any thoughts?
Another thought is to have mentors that are more personalized than they are right now, so that you can pick someone who is more like you. Thoughts about that?//Hannibal (talk) 09:14, 21 February 2011 (UTC)

I am a newbie who joined because I've been working on a Polish cabaret project and noticed a lack of entries in this category. I found the structure of the site very intimidating and the information that could help me kind of spread out all over the place. I wish the template of wiki language was immediately obviously available to a newbie. I wish there was a "new article" template which didn't take me through the whole "is there an article already" "is this neutral" question sequence but which did have the coding already present in the editing box for a new article.

I can't figure out if I need a different login for this Wikimedia outreach page! Also, I can't figure out the best place to leave various queries. Things which are obvious to long-time people are obscurely located for new people. I would sign this with four tildes if I'd been able to log in. Jane Peppler

Clueless[edit]

Please forgive me for intruding. I received a message posted by Lennart Guldbrandsson on foundation-l@list and was totally amazed by the pages' titles. They all had in common «Landing page» and that's how Lennart called them but they all contained different versions of the "Welcome message". That is very fine and interesting work, but I thought your project was dealing, as the name says, with "Account Creation". I believe Frank Schulenburg present the case for improvement very clearly in his "Account creation on Wikipedia and Facebook" pdf. It is amazing how people are still finding and clicking on "Create one" on UserLogin, "crear una nueva cuenta." on Entrar, "Créer un compte (facultatif)" on Connexion, or "Pode criar uma." on Entrar. Is this one more case of a project with a clear, well defined and narrow objective going completely haywire?

To save us all some valuable time, if you don't like these kind of questions and only welcome praise and pats in the back go ahead and block me to infinite right away, that will be fine with me for all I care.

Best regards,

(Redacted)

Vapmachado (talk) 00:45, 21 February 2011 (UTC)

I agree that, from the looks of this documentation, the focus is on welcome messages, however, I can assure you that the project is looking at every aspect. It's simply that the welcome messages are easy to change and lacking in helpfulness right now; the actual interface and text boxes should be changed, but that requires developer intervention. We're always open to feedback, so thanks for bringing this up. --Mono (talk) 01:12, 21 February 2011 (UTC)
Thanks for your question. Here is a more full answer:
As you can see on this page we are going to test versions of three different pages: the page you mention and two more. We believe that the easiest page to change is the last one, the confirmation page or the landing page. It's the final page newcomers see before we let them do whatever they want. So by some small changes here we may get very different results. But that doesn't mean that this page is the only page we will change.
I hope this is a little clearer. Again, thank you for taking the time to ask the question.//Hannibal (talk) 08:47, 21 February 2011 (UTC)

Sorry. My mistake. Wrong project. Vapmachado (talk) 03:37, 1 March 2011 (UTC)

Starting the tests[edit]

Hello,

In a short while, we will start testing on English Wikipedia, with other language versions to follow as soon as we have translated the messages. I'm very excited to see which version gets the best results in terms of increased number of new editors who start and continue editing. If you have any ideas about this, please add your own version to the page listed above.

As always, I am available for questions and comments.//Hannibal (talk) 17:13, 22 February 2011 (UTC)

How can you measure the success of these messages? --Mono (talk) 00:22, 23 February 2011 (UTC)
In two ways. 1) By doing the same Toolserver things we've done previously (number of accounts created, number of account who became active on the first day, etc) and 2) in a lesser degree, by checking up on individual editors (as a sample). Sorry for not including that on the testing content page. I will add it soon.//Hannibal (talk) 13:58, 23 February 2011 (UTC)
For number of new accounts use these stats. Some links are here (blue and red user pages). Przykuta (talk) 15:47, 23 February 2011 (UTC)

Another user's numbers[edit]

I just saw this research, which I found interesting.//Hannibal (talk) 23:39, 4 March 2011 (UTC)

Suggestions[edit]

I posted these on the strategy discussion and it was suggested that I share them here, so here are some concrete suggestions based on what I read there:

  • On the create login form, ad a checkbox [] I am a new editor in Wikipedia (or similar). If it's checked, trigger an auto-welcome message on the talk page with some useful links.
  • In the welcome message, have a "tell me how I can help" link that goes to a page or triggers a reply message explaining what a WikiProject is and listing 10-12 currently active projects across several areas of interest, with links to "add your name as a project member" and "introduce yourself and your interests on the project talk page".
  • Ask Projects to volunteer to support care and feeding of new editors.
  • Automatically add a template message on new articles by new editors similar to "This is a new article by a new Wikipedian. Please help improve this article by adding relevant wikilinks, citations, and formatting changes, and make suggestions for general improvement of the article on its Talk page."

- PKM (talk) 01:47, 16 March 2011 (UTC)

Thanks for your suggestions. Some of these may involve tech support, but let's see what we can do.../Hannibal (talk) 14:28, 16 March 2011 (UTC)
Also, I think new editor's User pages should already have a sandbox/test/practice page, and a button or link to "create a new test page". Eventually, it would be great if the test/sandbox page had a button or link to "move this page to production Wikipedia" that would let the editor choose a page name (with reminders about not starting with "A" or "The" except in titles of books and films, capitalizing only the first word and proper words, etc.), then make the copy and prompt the user to add appropriate categories to the article. - 216.178.108.235 23:19, 16 March 2011 (UTC)

Ideas from the plwiki's "Newbie" Wikiproject[edit]

The section is devoted to the analysis of the impact of the account creation mechanism on the new users' activity. It's a part of the Account Creation Improvement Project. Przykuta (talk) 09:28, 16 March 2011 (UTC)

  1. Logging in separated from account creation.
    Explanation: Currently the users see a login page, where they need to find and click an account creation link. The minimum requirement is to include the message "Don't have an account? Get one." in a separate frame on the login page. It would be best, however, to separate login and account creation with a more definite separator on the page which is visible for unregistered users.
  2. Changes on the registration page - put the form a bit higher, more accessible username rules.
    Explanation: after clicking the account creation link the form appears too low on the screen. What should be visible underneath is at least a language selection box. SUL is active nowadays (but it is a minor point), so people do not need to register separately for different projects (but it is also a minor point).
    In order to get to know the username rules, a user needs to click a link, which makes it easy to abandon the form - a most inconvenient solution. It might be a lot wiser to copy the rules and put them in a form of a collapsible template in the default collapsed form (with a + mark to show its contents) on the account creation page itself. Also, after the link to the username rules is clicked, it ought to open a new window.
  3. Excluding inappropriate phrases in usernames on the software level.
    Explanation: usually a user gets to know the facts after an account has been registered. Phrases like xxx.pl ought to be able to be excluded at the software level.
  4. Hints from the software as to similar usernames which have not yet been reserved, if an already reserved username has been typed in.
    Explanation: There seem to be fewer and fewer unique "weak" usernames available; after the user types in an already reserved username, the system could suggest a free username, e.g. adding numbers. Not all users will be satisfied with such a hint, but one can get discouraged in the process of hunting for the perfect username.
  5. Improving upon this solution.
    Explanation: After the option of creating a userpage has been introduced, which includes a list of useful links and is delivered via an inputbox preload, more than 500 new users decided to create their userpages including the bunch of useful information. Earlier, users were less willing to do any edits, and before any edit was made, it was unusual to welcome newly registered users. After such a userpage has been created, there is no need to overload user talk pages with welcome templates repeating the same links, and instead they can be greeted more individually. Moreover, it is visible that after the semi-automatic creation of userpages, users decide to edit pages more willingly. It would be good, however, to personalise the preloaded template a little - depending on what the user wants to do at Wikipedia, the set of links could be more specific.
  6. Bot-greeting of users who do not create a userpage as described above.
    Explanation: If the user does not know the rules of the project, (s)he can start editing in an unwanted manner, which can result in a quick block or rather unpleasant "test" templates insterted into their usertalk page. If after a few minutes (which should allow for the user to create a userpage) a bot greeted those who did not create their userpage (not everyone decides on the full spectrum of links, but there is a choice), the ratio of "incidents" would fall down. Additionally, the Welcoming Commitee could turn into a full-fledged Assistance Commitee and provide actual help instead of a semi-annonymous welcome template.
  7. Adding an inputbox of this sort to MediaWiki:Newarticletext (User=).
    {{#ifeq:{{NAMESPACE}}|{{ns:0}}
    |
    .
    .
    .
    |User= text + inputbox here
    Explanation: If the user does not follow the suggestion to create a userpage found in the Welcomecreation message (immediately upon registering the user account), quoting Polimerek -> "people get a red link to their userpage somewhere in the corner of the screen (OK... so what? Even if they click it, it opens an empty edit window which they usually do not know what to do with) and a link to their user preferences with a cartload of weird options, which are additionally hard to grasp for beginners. And that's all folks." Exposing the preload link a bit more and maybe adding the substituted template to be inserted into the user page, perhaps a even set of templates (if you wish to... -> insert this code, if you are planning to... -> put this into the edit box) - could be a useful hint. Optionally, automatic preload after a user has clicked the red userpage link in the said corner of the screen.
  8. "Create your own userpage" instead of the red userpage link in the corner of the screen.
    Explanation: I don't know if it is feasible, but if the userpage is missing, it would be better to replace the red username link with a "Create your own userpage" message. Combined with the above ideas, it could mean bigger motivation to join the community, especially when the preloaded template puts links that point the user to different channels of communication with other users.
The bot idea is implemented on nlwiki (User:SieBot and User:FoxBot take care of this) and users can tell what they think of the message. Although I think it is a good idea, enwiki doesn't seem to like it - TBloemink (talk) 17:40, 8 May 2011 (UTC)

Click tracking[edit]

Is click tracking being done on the create an account page? It would be good to know how many people visit that page, per day, and perhaps be able to normalize the # of new accounts/day based on that. Also, how about the # that hit submit? Upon hitting submit, there could be warnings, like your selected username is taken or too similar to an existing one, then the user might give up. Aude (talk) 20:36, 31 March 2011 (UTC)

Hello,
Unfortunally, we have up until now been unable to track those types of things, but the engineering department of the WMF are working on an A/B testing, click tracking system. If you are interested, you can read more about it here. If you do make comments (and please do), let me know. Thanks//Hannibal (talk) 23:48, 31 March 2011 (UTC)

enabling email during account creation[edit]

One thing I haven't seen addressed so far during this project is enabling email. We're seeing from the Wiki Guides project that there's a lot of potential to do better retention of newcomers through email.

Do we know what portion of new accounts enable email? And can we test ways of getting that portion higher? I think it's going to be an increasingly important opportunity, reaching newcomers by email.--Sage Ross (talk) 20:29, 13 April 2011 (UTC)

Thanks for the suggestions, Sage. As I understand the upcoming tracking solution, it will enable us to track what people click, including if they enter an email adress, and it is one area where I also think we should focus more on. Perhaps get back to inactive users with suggestions, like Amazon does, or just saying "we miss you". The first step, though, is to rework this womderfully robot-created message: http://en.wikipedia.org/wiki/MediaWiki:Confirmemail_body. I will approach the community later this week to start the reworking of that.//Hannibal (talk) 20:42, 13 April 2011 (UTC)

Next step in Account Creation Improvement Project[edit]

I just posted a short review of the next steps in the project on English Wikipedia, namely here (diff).//Hannibal (talk) 14:31, 2 May 2011 (UTC)

I have to say, it looks very good - TBloemink (talk) 16:16, 8 May 2011 (UTC)

Bad idea[edit]

"We should include the option "I want to tell my friend about this" - I strongly dislike idea "make it like facebook" 149.156.96.15 13:43, 11 May 2011 (UTC)

Facebook is a good idea, as a very large amount of people are on facebook, worldwide. An email option is also nice, if you prefer this - TBloemink (talk) 14:57, 11 May 2011 (UTC)
Anything like this, we would test first. If it turns out that plenty of people would use it regularly, it would likely be kept, but otherwise it would be taken out.//Hannibal (talk) 19:26, 11 May 2011 (UTC)

I think Facebook is a good model for us in its simplicity (well, in signing up, at least), but I also agree that we should keep the social-network-like functions to a minimum. The last thing we want is Wikipedia turning into Wikia (no offense to Wikia). Fetchcomms (talk) 03:54, 12 May 2011 (UTC)

Wikiproject[edit]

Along with giving new users a big list of articles can there be a way for people to connect with wikiprojects that they may be interested in. While many projects aren't active *coughdeadcough* there is normally one or two active editors that watch the talk pages and may provide a more human touch. --Guerillero 06:23, 15 May 2011 (UTC)

Yes, that's what we have been thinking about too. So let's test this hypothesis. Go to Account Creation Improvement Project/Testing content to see what is going on.//Hannibal (talk) 15:53, 15 May 2011 (UTC)

I love these new pages[edit]

We at WikiProject medicine have this banner here En:Template:MedWelcome. Anything we should do to integrate it better with this? Maybe allow new editors to pick areas of interest which than displays specific data for that area if not done already? BTW it is a huge step that we encourage people to use there real identity. This will not only bring more legitimacy to our project but hopefully decrease vandalism --Jmh649 (talk) 04:56, 12 June 2011 (UTC)

Thanks. Interesting. I think you're on the right track here. Perhaps one way to lead new editors to the right tasks is to use the pages that we set up for one of the tests:
http://en.wikipedia.org/wiki/MediaWiki:Customusertemplate-ACP2-Which_skills_do_you_have_to_offer%3F_%28Biology%29
The challenge is to keep them updated.
There is an interesting project similar to this, that the Association for Psychological Science are working on. You should check it out.//Hannibal (talk) 11:13, 14 June 2011 (UTC)

The Wikipedia Adventure[edit]

I'd like to throw out an idea we've been kicking around on EN-WP for a few weeks. It's an interactive tutorial that new editors could take/play which would introduce them to the community, the interface, and the basics of navigating, interacting, and editing. The spirit of the project is to make learning Wikipedia engaging, integrated, non-threatening, and fun. There are several hands working on the concept, the script, and the game development, but we would love more help. Especially needed are coders and those who have ideas about how we could put this guide/game in the hands of new editors. This seems like the right place for that, so I'd love to hear any feedback. The game is at ENWP, but if there's interest I'll create a project page here, too.

The Wikipedia Adventure

A fun, interactive tutorial for new editors
Link: The Wikipeida Adventure Project Page

Please leave comments or notes if you're interested in any aspect of the project. Cheers, Ocaasi (talk) 23:27, 12 June 2011 (UTC)

Cool idea. I read through it quickly, but it seems like something that can be very worthwhile. You should check out this TED talk for some inspiration. I'll see if I can contribute anything to it.//Hannibal (talk) 11:17, 14 June 2011 (UTC)
Thanks for that link. I did like it! And I want to know what the other 3 game dynamics are... While we're at it: Jane McGonigal goes to bat for games and rips it: Gaming can make a better world. Cheers, 69.253.41.54 01:31, 16 June 2011 (UTC)--That's me, sorry, I wasn't logged in --Ocaasi