Category: Interviews

Some people other then me saying smart stuff.

12 Nov

Interview with David Tenser – Firefox Support Director

Hello David,
Thank you for agreeing to do this interview!
First things first: Can you tell me a bit about yourself and how you first got involved in Mozilla? I heard it was a long time ago 🙂

Yes, the years have definitely flied by. 🙂 It started around the Netscape 6 days around 2001 when I discovered that Mozilla was basically the same thing as Netscape, only with different branding — or, rather, a lack of branding. Being the geek that I am, I wanted to learn more about this new open source project, so I started to participate in the newsgroups discussions and made connections with other people online. Before I knew it, I was pretty heavily involved with both asking and answering questions about the use, design, and direction of projects like the Mozilla suite and this new browser called Phoenix (which later became more known as Firefox).

Although I majored in Computer Science and have a history of software engineering, my involvement with Mozilla has mostly been focused on helping users. Part of the reason for that focus was that the Mozilla code base felt overwhelming and wasn’t something I had a strong desire to master. It was more fun to just help others while at the same time learning more and providing feedback on product features and decisions — especially since I was already busy with writing software in the non-Mozilla world. Also, helping people with computer problems has always been something I’ve enjoyed doing, so that part of me isn’t restricted to Mozilla.

So, that’s a bit of my history with the project. If you want the longer story, I wrote a longer blog post last year about what made me discover, and ultimately stay with Mozilla throughout all these years.

Aside from Mozilla, my main hobby is photography. I’m constantly on the lookout for new camera equipment, but I must admit I spend almost more time reading about cameras than actually using them. I guess that comes with the package of being a nerd. 😉 A more recent branch of this hobby is post-processing of photography, which I started to do this year. Being the open source evangelist as I am, GIMP is of course my photo editor of choice — actually, I even find it enjoyable to use now that I’ve set it up to use my custom keyboard shortcuts. It’s really not as bad as people say, it just takes a bit of time getting used to it!

Lastly, I’m Swedish, I live in Sweden, and I look pretty much exactly like the typical Swedish stereotype. I’m not a chef, though.

What did the very first SUMO looked like, and what were its pros and cons compared to the time it’s developed in?

The first version of SUMO, released in the fall of 2007, was very crude, but it still represented an important step up from the old support content that had been hosted on mozilla.org since the Firefox 1.0 launch in 2004. There were a number of improvements, but perhaps the main difference was that the support content was now hosted on a wiki, making it possible for people in the community to participate in the effort to improve the quality and grow the number of support articles. Essentially, the Firefox support became much more dynamic than it had been before.

Another benefit from the previous support content was that we now had the ability to see which articles were most popular, providing us with important insights about which problems Firefox users most frequently encountered. We could also finally see the most common search terms, browser usage stats, and much more. This led to an even faster growth of support articles and the ability to feature the most common support issues prominently on the support start page.

How it actually looked like? Here’s an old screenshot. Of course, a lot of things have improved over the three years SUMO has existed — it’s hard to even know where to start! Perhaps the most notable change is the sheer size of the project — both in terms of number of Firefox issues we have solutions for, and the number of people who devote time and energy every week to helping Firefox users have a better experience on the web. Our community is definitely the biggest strength with SUMO.

What’s the magic that got a lot of people involved in SUMO and why do you think a one would like to be a SUMO localizer?

I see two parts of the magic. One is the inherent magic of Mozilla itself and its products. If you’re helping people on SUMO, you do it because you love Mozilla’s products and believe in our mission.

The other part is the impact your contributions have on SUMO. Just as an example, if you spend just one hour to translate a popular support article into your language, that one-time effort can end up helping thousands of people every week. Depending on the language, your work could impact millions of users every year!

That said, there’s definitely a lot of magic yet to happen with the SUMO community — I don’t think we have truly taken off yet, despite having 50-100 active contributors every week. There are many areas where we can improve, and will improve. For example, we recently launched the Army of Awesome project, which focuses on helping people who are tweeting about their Firefox problems. It’s a great, fun, and simple way for anyone to give Firefox users a helping hand by pointing them in the right direction if they need assistance with solving a problem.

Another example is our upcoming knowledge base back-end, which is going to be a huge improvement to content editors and localizers worldwide. You can read more about this gigantic project in Kadir’s summary blog post — or you can check it out live on the testing server: master.support.mozilla.com.

After a lot of thinking about the platform SUMO is using, you decided to make some key changes to it?

I think you’re referring to the decision to completely rewrite the SUMO platform. Yes, this has ended up being the major development focus of 2010, and the decision was based on the current state of the SUMO platform at the time. The list of wanted features and fixes was growing faster than our development team was able to keep up with, and the fact that Firefox was becoming so popular forced us to spend a lot of our time just patching things up while the website was slowly grinding to a halt. We simply had to do something radical if we wanted to ensure we could remain competitive and agile, so we decided to build something new from scratch. A more detailed summary of the history behind this decision can be found in a blog post from February this year. Today we’re finally beginning to see the fruit of our hard work over the year — and the result is Kitsune!

What would be the 3 most important new features of Kitsune?

There are so many improvements in our new SUMO web platform (internally more known as Kitsune), it’s really hard to rank them. Our amazing development team has really gone above and beyond, working around the clock to give us a support platform we can feel truly proud of. We already have a new support forum based on Kitsune that has been out in the wild for a few months now, so I’ll focus on the knowledge base here, since that’s our big focus right now. Here’s my personal top three list:

Third on my list is the new localization and editing interface that has been radically improved to streamline all common use cases — from editing an English support article, through reviewing an edit made by someone else, to translating an article into another language. Have a look at the typical contributor scenarios we’ve worked on improving. To say that we’ve spent a lot of thought about what goes into the platform would be an understatement… 🙂 Special thanks to Chris, Michael and Kadir for their hard work in the critical design phase of the project.

On second place is the source code of the platform. This might not seem like a big deal at first, but it really is: the entire platform has been rewritten from scratch using the modern, python-based Django framework. What this means — aside from getting rid of countless of bugs and other issues that have plagued us for many years — is that we’ll be able to develop new features and fix bugs much faster than we’ve done in the past. Our SUMO development lead James Socol summarizes many of the benefits of the new platform in his recent blog post.

The most amazing new feature of Kitsune? Performance! The new Firefox Help website will be so much faster, both for users and contributors. In some places, Kitsune is almost 30x faster than the current website. It’s like we’ve gone from an old Wartburg to a brand new Koeningsegg! And we really needed it — with 400 million Firefox users and a thriving community of localizers around the world, timeouts and other weird behavior when editing support articles have become far too common on the website today. This will all be history when we release the new SUMO at the end of this month. 🙂

Where do you see SUMO in a 3 years period?

That’s an interesting question. I think there’s a lot of room for SUMO to grow way beyond our current size and reach. The recent Army of Awesome project has helped us reach out to more users in need of help, but there are still lots of Firefox users out there that aren’t aware of us — they simply don’t know that there’s an entire community of people ready to help if they have a problem with Firefox.

Three years from now, I believe that everyone who uses Firefox will know about its support site — and our community will be strong enough to help all of them. As for the support web platform we’re shipping this year, I can definitely see it turn into an open source support platform of choice for other projects like SeaMonkey and even non-Mozilla products. SUMO is already being used by Firefox (including mobile versions and Firefox Home) and Thunderbird, and this is just the beginning.

What message you’d like to send to people thinking about if they should join your team of contributors?

Please do! Helping Firefox users with their problems is simple, fun and only takes as much time as you’re willing to put into it. There are so many different ways to help — you can translate a support article into your language, which thousands of people will benefit from. Or you can help us improve the English articles. Or you can answer individual questions in the forum. Or you can interact with people directly using live chat, which is extremely addictive and rewarding. If you’re good vith video or image editing, you can help improve support articles by adding screencasts and screenshots. And if you think you don’t have enough time or lack the expertise needed to be helpful to others, we’ve got the solution for you. It doesn’t have to take more than a couple of minutes per day — and really, anyone can do it!

One last reminder: we’re just about to switch to a brand new knowledge base system. Please help us test the new stuff on master.support.mozilla.com so we can ensure a successful release at the end of this month! You can always ping us in the irc.mozilla.org channel #sumo, or post in our SUMO community forum if you have any questions. Welcome to the SUMO community!

14 May

Interview with Seth Bindernagel – Director of Mozilla Localization

Hi Seth. Can you tell me a bit about yourself, how you got into working for Mozilla and what were your first tasks?

Sure. I first started working for Mozilla back in 2006 when I was asked to start Mozilla’s Community Giving and Empowerment program. With the help of Asa Dotzler, I was able to launch a program to help members of our community with reasonable levels of support that would both assist and amplify a volunteer’s or a community’s contribution to the Mozilla project. Because so many of our community members had their beginnings in the localization work, I learned a great deal about localization and the needs of the community. It was clear that Mozilla should formalize even further the localization-drivers team to increase our focus on the global distribution of Mozilla applications and websites. Because I had much exposure to our volunteers, I was asked to help lead the l10n-drivers team.

Can you compare the importance of localization against some other parts of a complex projects like Firefox?

From a technical standpoint, someone could argue that localization is not as difficult as hacking on the Mozilla platform code or doing php web development, and they would be correct!  However, that is a bit like comparing apples to oranges because localization is incredibly important to the release engineering process at Mozilla.  If we were not able to localize our code, our global audience would not have as rich a user experience as they do now with a localized product.  Localization touches many parts of the release process along the way to final release, including the user interface, QA, and build.  Therefore, our localizers often have to wear many hats, understanding how to translate the language of the user interface, how to access nightly builds, how to read html and php code, and how to test the versions ready for release.

How do you see the localization in the future, in terms of a popularity among localizers and its complexity?

I think the localization of Mozilla products has gotten easier in the past two years, with more tools, reporting, and documentation available for localizers to use. I also believe that it will continue to get easier. But, the important fact here is not to eliminate choice. Our volunteers should be able to choose how they want to localize products and websites, whether using a “slick” webtool or using a more technical code editor. The point is that we should make hard things easy and let everyone experiment and participate as they choose.

Can you tell me, based on your experience, can we expect more and more localizers, and what do Mozilla do to attract new contributors and to promote using the localized builds of its applications?

We work very hard to continue to attract new localizers. This takes a combination of steps. The first step is to work with local communities to help build contributors. If a local community wants to actively build new contributors, we need to work with them to push that authority to them and to the edges to make sure they are empowered to do so. We are in the midst of planning one such event now with Mozilla’s “Inter-Balkan Meetup”. The second step is to build tools for better localization. This includes improvements to our infrastructure, documents, and tools for translation. New localizers do not always know exactly how our process works. So, we need to create a clear path for them to engage and learn how to contribute. With those two steps in combination, I think we will always be adding new community.

What does Mozilla do to improve the ease of translation for its localizers?

One very specific step we have taken is to implement tools like Verbatim. If a localizer visits http://localize.mozilla.org, he or she will see all the open web projects needing translation for various languages. This brings the work flow and avenue for participation to the localizer in a fairly understandable and clear way. Secondly, we try hard to furnish up-to-date statistics about the state of product and web localization through our dashboards. Localization communities can always see the status of various projects (including how many strings are needing translation and where they are located in the code base) by vising this URL: https://l10n-stage-sj.mozilla.org/ If we continue to enhance our tools and streamline our infrastructure to provide to our localizers the most timely information and statistics about work needing to be done, we will continue to make the process easier.

What message would you send to all potential localizers reading this?

Give localization a try!

  • If you speak a particular language, check out what needs to be done at the Verbatim URL: http://localize.mozilla.org.
  • Contact the locale leader for your language. You can see who the locale leaders are for various language teams here: https://wiki.mozilla.org/L10n:Teams
  • Email me if you have any questions. You can find my contact information at my blog: http://blog.mozilla.com/seth

We always look for new contributors and welcome anyone with any level of experience to participate.