I was in Japan November 3rd through November 10th. I went there to give two talks at the Kansai Open Forum conference in Osaka, and one talk at a MediaWiki developer’s meetup in Tokyo. I met a bunch of great fellow community members from the Wikimedia and open source movements at the conferences and on other days that they showed me around their cities.
A new experience for me
The talks were a new experience for me since I’m not used to talking to a Japanese audience, and I’m very much not used to having an interpreter. The experience of being interpreted is quite interesting, and it adds an extra layer of difficulty in speaking, and creating presentations.
Things that work well in a normal presentation don’t work as well when being interpreted. For instance, if you are used to putting jokes in your slides, the jokes aren’t as effective when they are based on the timing of the words and the slides in tandem. If you are saying the punchline as you show the slide for effect, you lose that punchline moment during the interpretation. I’m also used to practicing the timing of my talks, and this is something that is awkward when being interpreted. The rough estimate is that the total time of your talk will be twice as long if the interpretation is consecutive, instead of simultaneous. Of course that’s just a rough guideline. Some of the interpretation is going to be longer, and some will be shorter. The talk may be longer than you were expecting, or may be shorter. Your timing per slide will be much different than you expect.
Creating a presentation that is to be translated is also awkward. Certain phrases you take for granted don’t necessarily translate well. For instance, one well embedded mantra in open source is “Scratch your own itch”. It makes sense to many English speaking people, but this phrase doesn’t exist in all cultures. So, when writing the presentations, I had to pay special attention to try to rephrase any colloquialisms that I’m used to using. Cultural differences make other things awkward too.
An often used joke in the Wikimedia world is the “Every time you use meta-templates, Brion kills a server kitty.”, I like to use this slide too, but it’s kind of a hardcore image to send, so I decided I’d switch the slide to an icanhascheezburger image of a bunny crying, with the caption “You make bunny cry”. What I didn’t realize is something my second interpreter informed me of: In Japan there’s a saying, or a story, that basically says if you leave a bunny alone it will become very sad, and will die of loneliness. Funny enough, I still put forth a kind of hardcore image. Thankfully, I got smiles and chuckles from the audience, so I think it was OK.
Architecture talk – Kansai Open Forum keynote – Osaka
The first talk I gave was a keynote at the Kansai Open Forum conference. I’ve been told this is the largest open source conference in Japan. The energy was great. I unfortunately didn’t get much time to do much other than give the talks though. When I first arrived I spent a few hours going over the talks and presentations with Mr. Henri Hiraga, my interpreter. We did a run through of each talk, and he made some translation adjustments in the presentations. By the time we were finished we only had enough time to get some drinks from the vending machines, and head off to the presentation room.
The capacity of the room was 150 people, and it was full, plus some standing attendees. There was also a spill over room that had a video feed, which I was told had roughly 30 people. The talk was available online as a feed. I’m not sure how many people viewed it online.
I had a surprise when I started the talk. I incorrectly assumed I had a full hour for the presentation. I actually only had 50 minutes. Looking back at my emails, this was totally the case. I had prepared for an hour, so I had timed my presentation for roughly 50 minutes plus 10 minutes for questions. I finished the talk at exactly the 50 minute mark; no time for questions, which was unfortunate.
The talk was picked up by Gigazine, the largest blog in Japan. An audience member had live blogged it, and posted the finished product there. It’s pretty much a live translated version with pictures (which is totally awesome).
Community talk – Kansai Open Forum – Osaka
Immediately after the architecture talk, I gave a talk about how to be a part of the MediaWiki developer community. Like the architecture talk, this talk had quite a bit of interest. The room was smaller; the capacity was roughly 50 people. With standing room, there were roughly 60 people total. This talk wasn’t being recorded, and the room was small, so I was able to put the mic down, which was way more comfortable and intimate. Unfortunately, like the architecture talk, I thought I had an hour, and actually had 50 minutes. Again, this left no time for questions, as the talk was exactly 50 minutes.
Out of my two talks this is my favorite, and the one I’m more comfortable giving. I was a community member way before I started with WMF as an operations engineer. I’m always psyched about community incubation and I love talking to people about it. Our community rocks and I want everyone to be able to be a part of it. I especially want to ensure people feel welcomed no matter what culture they are from, or what language they speak. On the topic of culture, I learned an interesting thing about Japanese internet culture…
Before arriving at the conference, I went to lunch with the organizers and the interpreter. We had a really great lunch, and talked about a number of really interesting topics. One topic that was of special interest to me was one of the statistics for Japanese Wikipedia. Japanese Wikipedia has a lot of anonymous editors. It has way more than normal. What I learned is that, in general, it’s a cultural thing for Japanese people to not want to be known online using their real identities. This gives me mental images of all of Japan online being a giant 4chan. There were a number of possible reasons for this phenomenon discussed. For instance, in Japan it is very difficult to change your name. I’m very, very glad I had this discussion at lunch, because it pointed out a major flaw in my community talk when given to a Japanese audience.
In the “whys” of becoming a developer I stress a point about creating a very strong reputation, and a very public work record online. This is a big motivator in western culture. I think this is likely more of a demotivator in Japanese culture. Knowing what I learned from my lunch, I made sure to also stress the point that you can be pseudo-anonymous in our dev culture just like you can in our wiki culture. I stressed that while you can choose to have a public work record, if you want, that you can also be an anonymous contributor if you choose to do so as well. I think this was a good point to stress, as I could tell by the looks on people’s faces while talking about public work records that it was a stressful thought. I saw much more relaxed faces when I mentioned the idea of contributing in a pseudo-anonymous way.
Afterparty – Kansai Open Forum – Osaka
I didn’t get a chance to get any questions at the conference, but I had nearly two hours worth at the afterparty! The afterparty was great. It had roughly 200 people attend, and I got to meet people from Debian, OpenSUSE, OpenOffice, and a bunch of other open source projects. The most exciting people to meet were a couple of junior highschool students who are open source contributors. This is one of the things I love about the Internet, the open source movement, and Wikimedia Foundation (WMF) projects. You are solely judged based on the merits of your contributions and your communications.
The awkward thing about an afterparty like this, for those of us who don’t speak Japanese well, is that most of your conversations have to be interpreted. This is especially difficult after having a few drinks. I had a hard time remembering to pause for interpretation when I was really intensely answering a question.
I had a really good talk with an Opera employee who was interested in using the MediaWiki Selenium tests to test Opera builds against MediaWiki. They are, of course, very interested in ensuring their browser is working properly when using MediaWiki, as they want to make sure it works properly on WMF sites. This isn’t the original intention of our Selenium testing efforts, but it’s great that our tests for MediaWiki can also be used by others to test their own software.
Community talk – MediaWiki developer meetup – Tokyo
A couple of days after the Kansai Open Forum, I was to give a talk in Tokyo to Wikimedians interested in MediaWiki topics. After doing some exploring with a couple fellow Wikimedians kind enough take me around Osaka, I hopped on a Shinkansen and arrived in Tokyo. The next morning, my second interpreter, Ms. Tomomi Sasaki, brought me to lunch to discuss the talk. We had a good conversation, and also quickly went through the slides. When we arrived at the talk location, we did a quick run through of the talk, while she took notes, and made translation adjustments.
My talk at the dev meetup was scheduled for two full hours. My slides were written for roughly a 50 minute talk, so I added five more slides, and had Tomomi translate them for me. This added 15 minutes worth of time to my talk – not enough, but it would have to do. We didn’t have enough time to do a full run through of the talk before it was time, but it ended up going very well anyway.
The talk lasted an hour and ten minutes. I was kind of nervous at this point, because I needed to rely on getting questions. The good thing is, the crowd was very interested. I got nearly a full hour of questions! I didn’t realize that directly following my talk was a questions and answers panel for all of the presenters. I was asked another 10-15 minutes of questions during that too. Overall the level of interest was amazing. This really made me wish I had time for questions in the Osaka talks.
The panel questions were very interesting. The other presenters mostly discussed structured data. I was asked questions about structured data support in MediaWiki and especially WMF sites. Many questions were asked about Semantic MediaWiki (SMW) support. I don’t think my answers were necessarily what they wanted to hear, but they understood. Semantic MediaWiki really isn’t performant or scalable enough to work on WMF sites, and don’t totally solve the structured data problem anyway. It was nice to hear the discussion about this anyway. I did of course mention that I often use SMW, that I maintain a couple SMW extensions, and that I think SMW is a great technology.
I got a few questions asked from twitter, one of which was politically oriented and I decided to not answer. The question was: “How do you feel about China blocking Wikipedia access.”
Afterparty – MediaWiki developer meetup – Tokyo
The event staff, a number of the event attendees, and I went to a restaurant as an afterparty for the event. I got to meet a bunch of very interested and genuinely friendly community members. I answered about another hour of questions here, though many of the questions were about architecture, and not the dev community. There’s always the possibility we’ll get some operations contributions from the community talks though, so I think it’s still a strong positive. I always think sharing knowledge is a positive though, so if we don’t get any contributions, it’s still a win to me.
I had a great time in Japan, and think the talks went really well. The Japanese community rocks! I’m hoping we’ll get a number of contributors as a result of this trip. At the minimum, I hope we’ve given people an idea of how our community works, and people will be able to use the resources provided to produce something cool and useful. For sure, this was a great learning experience for me in a number of ways, especially regarding Wikimedia culture in Japan.
For those interested, I have the talks with notes here:
- Wikimedia Architecture (English)
- Wikimedia Architecture (Japanese)
- How to be a part of the MediaWiki developer community (English)
- How to be a part of the MediaWiki developer community (Japanese)
The notes sections in the Japanese versions of the slides are still in English (sorry!). When the notes sections are translated to Japanese, I’ll update this post, and the links.
I believe both the Architecture and Community talks were recorded; I’ll link to those and update the post when I get them.