Now that we’re redesigning the website for Student Reps, I’ve learned about the tools that are used to localize a site. If you’re interested in localization, this post will provide a good primer so you can understand the technical side of an L10n project.

Using gettext as the localization format

There are a few localization formats from which to choose. Based on the scale of the project and our familiarity with it, we chose to utilize the gettext format for our localizations. Gettext allows us to easily add new languages to our project, and it also works well with Verbatim, the web interface our translators use.

In the first step, we use gettext to generate language-specific files. Gettext will give us a .po file for each language we need. In fact, this process can be easily automated by running a script. Once we have the files in the target languages, we compile these .po files into .mo binaries that will be used on the live site.

Managing Translations with Verbatim

Verbatim is a web localization tool for Mozilla projects that allows translators to localize the site without editing text files (such as .po files for gettext). Localizers create an account, choose a project and language, and then start translating.

The web interface is very simple and makes it easy to view the status of a project or particular language. On the left side of their display, localizers see the string that needs to be translated. On the right side, localizers enter suggested translations for each string. Those translations are then reviewed for correctness by a reviewer for that locale. Once a string’s translation is approved, it is committed using svn and shows up on the staged site.

• Wednesdays at 11am-12pm EST / 15:00-16:00 UTC
• Saturdays at 3-4pm EST / 19:00-20:00 UTC
• Sundays at 12-1 pm EST / 16:00-17:00 UTC

Getting Started

If this is your first time using IRC for chatting, I suggest you connect using the applet posted on our Student Reps IRC page. If you’re a more advanced user, stop by the Mozilla IRC Network for information on connecting with a different client. Once you are connected to irc.mozilla.org, type /join #studentreps to enter the channel.

Background

A few weeks ago we asked our Reps if they would be interested in having set times during the week when they knew they could chat with each other. Positive feedback rolled in! To further define the best way to work this, I created a short questionnaire. Sixty Reps responded. Here’s what they told us:

• Best days for Office Hours (in order of preference) were Saturday, Sunday, Friday and Wednesday
• Best time of day was evening, with an 88% interest level (28% liked afternoon, and 5% preferred morning)
• 85% want to both give and receive advice
• Most interest came from India (37 responses), followed by the US (7) and the Philippines (4)

To start, we’re choosing Saturday, Sunday, and Wednesday because they were the most popular days, and they are nicely spaced to provide coverage throughout the week. I’ll be in the channel myself on Wednesdays, and I’ll try my best to be there on weekends as well.

As we learn what works best for everyone, the hours for each day may change slightly. Keep in mind that these are just official times when you can expect me to be online along with Reps and Regional Leaders. You’re always welcome to be in the channel anytime, 24/7.

As Mary Colvig mentioned yesterday, we recently surveyed our community marketing team. And that of course includes our Campus Reps (soon to be known as Student Reps).

Thanks to all 367 Reps who completed the survey. That’s an impressive response and represents about 20% of our active Reps. With that in mind, let’s look at the findings.

Who took it?

• 367 Reps
• 44 countries
• Mostly new contributors

Reps' Years of Contribution

How active are they?

• 78% promote Mozilla constantly
• 17% promote during news announcements or marketing campaigns
• 5% promote every once in a while or rarely

What are the Reps interested in doing?

Across the board, Reps have the same interest in activities that the whole community marketing team has.  Reps showed slightly more interest in Public Relations, Hosting events, and Market research. Exact differences can be seen below.

Reps' Areas of Interest

Also, Reps selected 5.1 activities on average that interest them, compared to the 4.6 activities the whole team selected. This means that Reps are interested in participating in more types of activities.

What do Reps want to learn about?

We asked what types of workshops our team wants to see. Not surprisingly, workshops specific to Campus Reps were at the top of the list. Speaking/presenting and graphic/web design were also quite high with 57% and 56% of Reps interested in those workshops, respectively. Interest in the other workshops can be seen below.

Reps' Types of Workshops Wanted

So what does this mean?

• Reps are very active in promoting Mozilla, doing so either constantly or during announcements and campaigns.
• Student Reps have similar interests as the rest of the community marketing team. However, it’s important to note that Reps tend to be interested in more types of activities overall. Therefore, we should make sure that a variety of opportunities are available to them.
• Reps show slightly more interest in PR, hosting events, and market research relative to the whole community marketing team. We can use this insight to offer more opportunities to Reps in these areas.
• Workshops about Campus Reps, giving presentations, and graphic/web design are highly sought after by Reps. This finding can help us in planning future workshop topics.

Since many of us haven’t seen the members of UVA secret societies in person, I thought this would be neat to share. I recorded a short video clip of their members as they left the wreath this morning and snapped a couple pics during the daytime so you can see it better.

Wreath

Letter

We surveyed Reps to find out their thoughts on the program's name

A few weeks ago we heard from some Campus Reps in Europe that the name of the program wasn’t well understood at their schools. They said that the idea of a campus is not well known in Europe because most students commute to class and do not live at their school. After talking with our Regional Leaders, we decided to ask all the Reps by creating a simple survey. With just three questions, we’re able to gain some valuable insights from the 191 Reps who responded.

Questions

• On a scale from 1-5 (1=not understood, 5=well understood), how well is the “Campus Reps” name understood by students in your country?
• If “Campus Reps” was to be replaced with another name in your country, which of the following names would students most likely prefer? (choices: Student Ambassador, Student Representative, University Ambassador, University Representative, or suggest your own write-in)
• Which country do you live in?

Not surprisingly, the data shows that student opinions on the Campus Reps name vary by geography. Overall, 56% of Reps feel the current name is well understood. Once we break down the data by country and region, we notice some interesting trends.

Exploring alternative names

Now that we know the Campus Reps brand doesn’t relate well in some regions, let’s take a look at how Reps liked the other names we suggested. Overall, Student Ambassador and Student Representative were the top picks (receiving 47% and 25% of votes, respectively). Choosing a name that focuses on the Student rather than the University makes sense. Not all of our Reps are in universities or refer to their school as a university — we even have some Reps in high school.  Also, Representative is a more natural fit since Campus Reps already refers to representative by using its abbreviated form, Reps. Choosing a name that doesn’t include Rep or Representative could cause confusion as well, which is why I think its best to pick Student Representative.

Insights

• In the US, almost all of the Reps feel that the current name is appropriate
• In Europe, the Campus Reps name is not well understood.
• In Latin America, the name only relates somewhat well to students.
• Student Representative appears to be the best alternative name.

If you want to take a look at the survey data yourself, you can grab it in csv format here:

Campus Reps name survey data

Moving forward

The next step is to get feedback on this data and decide how we want to tweak the Campus Reps brand for different regions. Do we want one other name for Reps outside the US? Or multiple names? What are your thoughts about the Campus Reps name? Let me know what you think in the comments.

UVa uses Twitter to let fans interact during football games

With seven turnovers by the Virginia Cavaliers, the first game of the 2009 college football season was not fun for UVa fans, as they lost in an upset to William & Mary: 26-14. Ironically, the only thing that brought smiles to the fan’s faces was a turnover of another kind: the social media turnover of the Jumbotron display to the Cavalier fan base.

Early on, the football announcer invited fans to use Twitter to tweet text and pics during the game. He told the near-capacity crowd that tweets containing WMvUVA hashtags (#WMvUVA) would be displayed on the new Jumbotron screen that towers over the end zone. And they were displayed—at speeds approaching real time.

My friends and I had a great time posting school-spirited tweets and some fun pics to go with them.

Last year, a similar service allowed fans to text messages to a 5-digit number and see their messages show up on the screen. While that system worked well, the new Twitter approach kicks it up a notch. Now fans can include pictures with their tweets, enriching the message’s value. And more importantly, those messages are now public and searchable on Twitter. Even fans watching at home can get in on the action by sending hashtagged tweets.

There’s no question that this a great play for college sports. But the takeaway has broader implications: it highlights the pent-up demand for more fan interaction. Now that athletic departments can engage their fans during games with social media, why not continue that Game Day experience back on the web?

Virginia has already started this with their Facebook profile and Twitter account. However, this content is nowhere near what it could be because it’s only flowing one way. That sounds more like Web 1.0 than 2.0 to me. Fortunately, that’s easy for teams to fix.

Sports teams, your next step is to actively engage fans online. Here’s how. Use your Facebook status to post questions and start conversations. Seek feedback and ideas from fans. Don’t just tweet about upcoming games and scoreboard updates. Reply to tweets and get to know your fans better. Retweet fans’ comments to your followers. And offer them rich media. Your wallpapers and podcasts are a good start. But don’t stop there: why not let fans skin their browser with a Persona of their favorite teams, add a Twibbon of the mascot to their Twitter avatar, or send in fan videos. Utilize your media assets.

If you bring the Game Day experience to fans outside the stadium, they’ll get even more excited when the next game day arrives. It only takes a small time investment to connect with your fan community, and the potential payoff is huge—fans who are more excited and more loyal. And maybe, just maybe, fans that are more engaged will cheer the team to victory more often. Just sayin’.

How can your favorite teams better engage their fans? What teams are already leading the pack with social media?

Data

The data set includes the 779 Reps that registered for the 2009 – 2010 school year as of June 2009. It should also be mentioned that almost three times that number of people started the survey but did not complete it. In the future, we’ll want to look for ways to improve the survey so more people complete it.

Regions

We have Reps in 57 countries. The majority of Reps are located in India, the United States, and Mexico. Here’s a list of the top 10 represented countries: India, United States, Mexico, Canada, Brazil, United Kingdom, Sri Lanka, Pakistan, Philippines, and Australia. We have at least 8 Reps in each of these countries and that group accounts for 85% of our Reps.

US Schools

Our Reps are spread across 37 states (plus DC!). The unrepresented states are: Arkansas, Colorado, Connecticut, Delaware, Idaho, Montana, Nevada, New Hampshire, New Mexico, South Dakota, Vermont, West Virginia, and Wyoming.

School year

Most of our Reps are upperclassmen and a good number of them (52%) will be graduating by next year. To make sure campuses continue to be represented, we should encourage these graduating Reps to recruit other students to take their place when they graduate.

How they found out about Campus Reps

Not surprisingly, most students find out about the program on Spread Firefox. However, it’s also great to see that 1/4 of new Reps are joining based on referrals from friends and classmates. To me, that speaks highly of the program and what students are getting out of it.

Reasons for becoming a Rep

Again, the number one reason that Reps join is not surprising — they simply love Firefox. Interestingly, Reps see the program as a chance to give back to the community, learn more about marketing, meet new people, and become a leader. The newly created Campus Reps Network definitely provides an environment for Reps to achieve these goals.

That’s it for now. In the next post, we’ll look at the projects and campaigns that most interest our Reps.

It started just two weeks before Firefox 3.5 would be released. Prashanth, one of our Campus Reps in India, posted an idea on the Reps’ forum. He suggested a synchronized social media campaign to celebrate the new Firefox launch — think of it as the stadium wave meets Twitter. Other Reps quickly joined the discussion and it was soon clear that there was lots of excitement about the idea. We could get all the Reps involved and tell our schoolmates about it too. We’d have hundreds of students participating in the Shock.

But what about the rest of the Mozilla community? Would we only target this campaign  to students? This wasn’t an issue last year with Firefox 3 — Download Day was open to everyone. And that’s what made it so successful. With that in mind, Jay and I knew that this campaign should include the whole community.

What would we call this virtual stadium wave? Firefox 3.5′s code name “Shiretoko” was a natural fit. And we used “Shock” to capture the wave effect the campaign would create. Each wave of social media would be called a “shock.”

Next, we planned the logistics for how the Shock would take place. Because we were celebrating the release of Firefox 3.5, we wanted the Shock time to be 3:50pm. We debated whether to do one huge, global shock or 24 regional shocks — one for each timezone as we spread the Shock around the world.  A single global effort would be bigger in scale and more noticeable on social networks. However, the timing would be inconvenient for certain regions. The regional shocks were convenient in terms of timing (3:50pm in each time zone), and they would allow each region to get excited about their regional shock. Since both options had pros, we ultimately decided to allow people to participate twice.

With the strategy set, I went about designing the Campaign page on Spread Firefox. The layout of the content is straight forward — an introduction to the Shock idea, directions on how to get involved, and the reasoning behind the campaign. The most difficult part of this page is communicating when someone is supposed to participate because there are 25 (24 + 1) shocks taking place. The campaign page listed the first 5 timezones for the regional shocks and the time for the Super Shock. We also linked to a world clock page that let participants look up when the Super Shock was in their timezone. In retrospect, I think it would have been best to list all 24 regional shock times so it would be more clear.

Campaign page

Since the Campaign page was lengthy and got into the details of the Shiretoko Shock, we decided to create a Landing page as well. The purpose of this page is to promote Firefox 3.5, allow visitors to download it, and give users a chance to spread the Shock. If a user wants to learn more about the campaign, they could visit the Campaign page, but the details are not mentioned on the Landing page. Jeff, another marketing intern, helped me design the Landing page on the Friday afternoon before the release.

Landing page

To publicize the Shiretoko Shock, we messaged our marketing, launch team, and Campus Reps email lists a few days before release to let them know we’d be announcing a campaign soon. On Monday, the day before launch, we sent out the details and launched the Campaign page. Then early Tuesday morning we launched the Landing page.

How big was the Shiretoko Shock? I’ll post about the numbers and the impact of the Shock in a few days.