[Outreachy] · March 2, 2018 · 7 min read

Documenting is the key to thrive

It’s no secret I am a fairly new member of the Wikimedia movement. As you can see here, I only started contributing on September 11, 2017, encouraged by Outreachy’s application process. Since then, I made a total of 2,254 edits including translations, daily notes about my internship and production of small videos to illustrate my Translation quick guide. I was completely immersed in one single aspect of the movement—technical translations on MediaWiki.org—for three full months, and I am aware I still have a lot to learn.

But I believe there is no need of having a great amount of experience with Wikimedia projects to point out that recruiting volunteers and organizing work on Wikipedia and doing that on MediaWiki.org requires different strategies—actually, if there is something that this internship proved right, it’s that sometimes you need novices that don’t quite know what to expect to help you find the most obvious points of failure. It’s like that QA joke:

QA Engineer walks into a bar. Orders a beer. Orders 0 beers. Orders 999999999 beers. Orders a lizard. Orders -1 beers. Orders a sfdeljknesv.

— Bill Sempf (@sempf) September 23, 2014

It makes me think about a recurrent trend in the software developement world: calling users “idiots” if they don’t behave like we predict. I disagree with this kind of mindset completely ¹: if so many people are having difficulties, maybe the problem isn’t them but the design²³. We cannot build systems and processes hoping that people will magically understand what we had in mind while developing them. We need to make them crystal clear—with no distraction whatsoever—so we can retain the most of the intended audience without running into perfectly avoidable problems.

---

“The Wikimedia movement is a volunteer movement: we edit and translate (and to a fairly large degree, do technical development and technical documentation) in our spare time, but the roads to becoming a translator are difficult to find, and it’s difficult to get engaged in the movement as a translator rather than as a an editor who sooner or later ends up helping out with translation.” — T158296

Here is an interesting fact: the Wikimedia movement isn’t a group of people deeply involved with free and open-source software, but a organization deeply driven by open knowledge that happens to develop free and open-source software to support their operations (and has a small, dedicated community with ties with FOSS as a result). This is an important distiction I suspect that makes all the difference when thinking about recruiting efforts for technical translation.

I was discussing with my mentors about how difficult is to a person that is not familiar with the Wikimedia movement to contribute as a technical translator. In Bringing documentation to light⁴, I point out that translation efforts are unorganized on MediaWiki.org, which kind of reflect the overall work culture within the movement: take a task you would like to spend time on (after all, you are doing this in your spare time); try to complete it as well as you can; contact us in case you run into any problems. We have chapters, the Wikimedia Foundation, user groups, thematic organizations, but my impression is that those things are there to support the movement, and generally every member not deeply associated with them works (or is encouraged to work) independently.

Although this strategy may work well with Wikipedia⁵, it has some downsides: you end up with a set of unwritten rules and conventions that are not easy to catch and usually are passed from a person to another informally (as organizing them takes a lot of effort and it’s easy to forget about this kind of task when you wish to dedicate your time to things you consider more important). Let’s face it: most people don’t like to document their work. It’s a boring, repetitive task, often not as exciting as coding or editing. And the way the Wikimedia movement works brings another challenge to the equation: even if they do document their work somehow, the lack of solid guidelines to help them complete this task (style guide, indication of where to make it available, how to publicize it) could result in not-so-useful documentation. The dimension of the movement itself make it difficult to find this kind of information. As a result, information is extremely fragmented.

Therefore, it’s not surprising that some of the most engaged editors will end up as occasional translators sooner or later. They belong to a group that needs to be fundamentally familiar with the way the movement works, the software that supports the operations of every single project. They have this willingness to help the movement because they know how important is to make content available in multiple languages. But here is something that often happens with people like me and them, people that aren’t “actual translators” but have the required fluency to do this kind of task: we are inexperienced (especially if this is our first contribution in the free and open-source software world and we are in charge of this role alone!) and frequently we don’t put much thought in long-term consequences. We just want to get things done⁶.

That is, of course, a recipe for disaster. Maybe someday you will become busier and won’t have time to contribute to that specific project. Consequently, you probably won’t have time to explain to someone else all the work you’ve done and that will require them to make some guesses. It won’t take long until inconsistencies begin to appear, making users extremely confused—particularly if said project follows the trend of only getting bigger and bigger everyday as we speak.

With that, we finally get to the reasons why I was so adamant and persistent about the idea of translation teams. After all, this is what most of the FOSS projects do, and being FOSS is one of the most fundamental traits of MediaWiki.

So I tried to create a Brazilian Portuguese team with new contributors only. I made a call on Twitter, being absolutely direct: I need some people to help me with this over the course of the next two weeks. I’ll issue a certificate of participation for your help. I had a better response than with my attempts to publicize the role of technical translators: with this approach, three people contacted me, two created MediaWiki.org accounts, one followed through—and they said they did this because they have always wanted to be a volunteer translator, but never heard back from projects they attempt to make part of.

Another surprise was my fiancé offering his help. He is a Systems Information undergrad and always wanted to contribute to free and open-source projects, but never did. He usually practices his English with Duolingo, but was growing tired of it. He saw that as an opportunity to combine business with pleasure.

And it worked. Both of my teammates have different schedules and sometimes they can’t contribute as much as they want (and I was busy doing other things during this period, so it happened with me as well), but together we increased the translation rate of selected pages from 24% to 65%. I am really happy with this!

So as you read this, I already interviewed both of them about their experiences and I am writing my final report. As my internship ends on March 5 and this is my second-to-last bi-weekly report, my next one will probably address my thoughts, conclusions and recommendations to the Wikimedia community. But this doesn’t mean it’s the end of my blog—I would say it’s only the beginning of it. I certainly enjoyed writing that much again and I still—and will—have a lot of things to share.