Saturday, May 14, 2011

An interview with Sue

Wikimedia Foundation Sue Gardner Red BG Sept 2010
When Sue gave the green light for me to ask her 10 questions, it was the perfect start of my birthday. I hope you will enjoy the questions and answers, they provide a nice insight in the current thinking and practices of the  WMF.

So far every WMF introduced novelties are focused on the English Wikipedia. can we have novelties that focus more on other projects, countries or languages
It’s actually not true that the Wikimedia Foundation staff works only on features for English Wikipedia --- for example, we just rolled out the Upload Wizard for Commons, which supports all the projects.

Typically what we try to do is develop features that will be relevant to many languages and possibly multiple projects. Sometimes we roll them out on a smaller project first, like the strategy wiki, or projects like Wikinews or Wikibooks, and then later release on enWP and other large projects. Sometimes we start by rolling something out on a small proportion of enWP articles, like we’re doing right now with the article feedback tool, or for a small proportion of users, like we did for the edit section link experiment.

It’s true that we wouldn’t typically develop a feature that can only be used on a small project or small language-version. I would have a hard time justifying dedicating our technical resources (which are pretty slender) to something that would only affect a small number of readers, when there is so much low-hanging fruit that affects lots of people. It makes sense to me that the Wikimedia Foundation would generally aim to do work that supports the maximum possible number of people.

I will also say though, that the Wikimedia Foundation needs to better support internationalization and localization work. We know that :-)

The top 10 Wikipedias in traffic generate more then 90% of our traffic. Would you mind if this was 80%
Of course not!, why would I? :-) I would love to see the smaller language versions flourish.

I believe the Wikimedia Foundation should be agnostic with regard to languages: it’s not our job to try to direct readers towards specific language-versions, or to try to persuade editors to edit specific language-versions. That’s their business, not ours.

So --- if it were a level playing field then everything would be fine. The problem is, it’s not. Currently as you know there are impediments to people editing in some languages, and those impediments restrict the growth of those language-versions, which in turn constrains readership. IMO therefore, the job of the Wikimedia Foundation is to eliminate the impediments that are within our control, so as to level the playing field as much as possible. After that, it’s up to individuals to decide for themselves what they want to read and edit.

Traffic for the Chinese Wikipedia is sky rocketing.. do you have any clue why 
Yes you're right --- in the last three months there’s been a significant bump in traffic to the Chinese Wikipedia, and the Chinese Wikipedia has the 4th highest reader growth of any language-version over the past year.  I haven’t heard anything that would seem to explain this from Chinese-speaking Wikipedians, nor have I read anything in the media that seems relevant. But eyeballing comScore, it looks like Baidu and Hudong both took an uptick in roughly the same timeframe. So whatever’s happening, it may be more about the internet audience in China, rather than something specific to us. But that’s a guess; I don’t actually know.

Traffic on mobile phones is sky rocketing ... but bandwidth is expensive to our users. what can we do to prevent costs to our readers
Yes --- data on mobile phones is expensive. Obviously this is an impediment for people who want to access Wikipedia via their phones, and it hits less-wealthy people disproportionately hard. That's a problem we would like to solve.

So in the coming year, the Wikimedia Foundation aims to create partnerships with mobile carriers that will enable people to access Wikipedia on their phones for free. There are obvious benefits to the carriers in doing this, because free access to Wikipedia would be very appealing for mobile companies’ customers and prospective customers, and because offering Wikipedia for free is a public service: it’s a really simple way in which mobile companies can do good in the world, and be seen to be doing good. And of course it’s great for readers.

Mobile is a priority for us, because mobile usage is growing so quickly. We all know the numbers. A few years from now, more people will be accessing the internet via mobile than laptops/desktops. There are going to be 6.5 billion mobile subscribers by 2015, many of them in the Global South. We want all those people to have easy access to Wikipedia.

There’s more on our mobile work here:
And for developers, the initial work on porting our mobile gateway is happening here:

    There is quality in intellectual content and quality in readability. is Wikipedia in danger of becoming too high brow?
    Mmm, I think I might reject that premise --- as a former journalist, I feel like intellectual substance and readability cannot possibly be diametrically opposed :-)

    When I worked in public broadcasting, we aimed to make our work accessible to anyone who was smart and curious, but who had no particular expertise in the subject-matter. At the same time, we were especially pleased if and when we also got positive feedback from experts.

    I think we should have essentially the same goal at Wikimedia. Foundational or introductory type topics should be accessible to anyone, and more specialized topics can appropriately be somewhat less accessible. My impression reading the English Wikipedia is we do a pretty good job of handling specialized topics. (By which I mean: I personally am never going to read the article about e.g., taq polymerase, and therefore it is totally fine that when I take a spin through it, I am baffled.)  But I think that when it comes to introductory material we are all over the map: some articles are very good, some are far too basic, and some too advanced. That’s not surprising: making things simple but not simplistic is a rare skill. But it can definitely be done :-)

    The appreciation of the WMF is much better then the appreciation of chapters. is this because the chapters still have to grow in their roles? 
    You’re referring to the results of the 2011 Editor Survey.  Let me recap: About two weeks ago, the Wikimedia Foundation ran a survey of roughly 5K editors in multiple languages. The results are in, but there are still data integrity checks and cleanup going on. I’ve been poking around the raw survey data in Limesurvey, so I have some early results.

    So back when we were designing the questionnaire, I'd wanted us to ask how editors would rate the performance of the Wikimedia Foundation, and I wanted to have some context for the answer. So, we asked an identical question ("on a scale of 1 to 10, how would you rate the performance of X") for four different groups: "your own performance," "Wikimedia volunteers overall," "the Wikimedia Foundation" and "chapters."  (We offered a Don't Know option, which didn't count
    towards the results.)

    Here’s the average weighted response for each:
    • "your own performance" 6.24
    • Volunteers overall 7.14
    • Wikimedia Foundation 7.32
    • Chapters 4.33
    You’re asking me, Gerard, why the appreciation for the Wikimedia Foundation is greater than for the chapters. The first and most obvious answer would be that there’s a methodological/data problem, and that’s definitely possible.

    But you implied that the answer might lie in chapter’s relative youth, and I think there’s truth to that. The chapters' value will become more obvious over time, as they are better understood (and better understand themselves). It's also true that chapters’ work is inherently fairly experimental --- that has no bearing on the actual value of the work, but experimental work by definition doesn’t get
    predictably get good results, and it‘s not always obvious what value is being achieved. So that could be a factor too.

    I also think it’s worth pointing out that if the Wikimedia Foundation disappeared unexpectedly tomorrow, the projects would have a huge problem, because the Wikimedia Foundation pays the bandwidth bill, it buys and installs the servers, defends against legal threats, and so forth. As long as it’s doing that, the Wikimedia Foundation is probably going to get X minimum level of appreciation from the community.

    So upshot: it’s inherently harder for the chapters, because the chapters aren’t essential to the immediate continued operations of the projects in that same way. Public outreach, editor recruitment, content partnerships, etc. -- all of that is IMO extremely valuable: it just doesn’t have the ‘duh, obvious’ factor that
    bandwidth-and-servers do.

    I want to note also in passing that I think it’s charming, and somehow characteristic, that Wikimedians tended to rate their own individual performance as lower than the performance of “volunteers overall.” Wikimedians, as we know from the article feedback tool, are tough graders :-)

    If Wikimedia is to be a movement, it must have room outside the WMF and still be part of "us". any ideas on this?
    Yes, I think that’s true, and I think it’s happening. I describe this as ‘porosity’ -- we need to be a porous movement, with people and ideas flowing easily throughout.

    I think it’s people’s natural instinct to want to put up barricades, and I think we need to be continually vigilant to not do it. When I first joined the Wikimedia Foundation I was disquieted to experience the movement as clubby and insular. There was a huge learning curve for outsiders, practically nothing was documented in an accessible way, there was important information that seemed to be available only through personal networks, tenure equaled status, and so forth. It felt clubby.

    I now believe that a lot of that was unintentional. It was a symptom, an unintended consequence of a system that simply hadn’t put enough emphasis on accessibility and openness.

    Over the last couple of years though, I believe we’ve begun to see the walls start to crumble. Not nearly enough, but it has started. Technically, this means ensuring speedy access for everyone, making the interface as user-friendly as possible, and continually developing features that support increased openness. Editorially, it means minimizing policy cruft, giving good-faith new editors the benefit of
    the doubt, resisting the urge to feel ownership over articles we’ve edited, and respecting the original injunction to be bold. This is an area where I think the chapters play an important role, because they’re uniquely situated to create partnerships in their geographies --- for example with schools, cultural institutions, media, and like-minded organizations who share our goals of freedom and openness.

    The WMF planning is really tight. how much room is there to deviate to allow for wishes from the community or new or altered opportunities
    Do you find the planning tight? I think it’s not bad. For example we’ve got the grant-making program, which in effect provides 600K worth of flexibility for chapters and individual Wikimedians. That’s not nothing :-)

    It’s true though that an annual planning cycle --which is what we have-- is inherently restrictive, particularly for organizations which are young and/or experimental. That doesn’t mean we don’t need an annual plan -- we absolutely do. I need to clearly convey the work of the year to the Wikimedia board of trustees, to donors, to editors and the staff. But I also need to build in an appropriate amount of responsiveness. So it’s a balancing act.

    News from the WMF is typically like "we have done this", how about more news about "the road we travelled".
    That’s really interesting. Let me give you a parallel, so I can see if I’m understanding you.

    When I first joined the Wikimedia Foundation, I spoke to the board like a typical CEO. “We did X,” “we are going to do Y,” “here are the results of Z.”  Over time I’ve found myself behaving quite differently, because I was finding that the succinct results-focused style didn’t suit us. Why? Because our board loves Wikimedia fiercely, deeply, and they are intellectually curious people. Which means they don't want just the upshot: they have the time and appetite to really dig in. And, we're experimenting together -- we're building something utterly new. Which means that sharing what we're thinking & learning is at least as important as sharing results.

    I think also that the Wikimedia culture places an unusually high value on intellectual honesty, which includes the idea that changing your mind is admirable not shameful. And so, I felt like it built trust between me and the board when I was able to share with them that we were course-correcting.

    Is that what you’re saying here, that you’d like the Wikimedia Foundation to show its thinking more? Because if so then yeah, that makes sense to me. If there are any specific stories / subject matter areas that you’re curious about, tell me and I’ll see if we can get them addressed.

    You moved to an Android phone, are you still a speed typist ?
    Sadly, no! The Droid is pretty much useless outside: the glare fromthe sun makes it impossible for me to read and type. That has killed my decade-long habit of multi-tasking by walking while e-mailing. ButI don't mind: the Droid has lots of benefits to offset that disadvantage :-) 

    PS you answering these questions is a nice birthday gift :)
    Happy birthday; I hope it is/was lovely. And I hope you are really, really well happy :-)


    Elitre said...

    This looks related:

    gifti said...

    I really liked this. It was different than usual and gave a nice insight.