Hot on HuffPost Tech:

See More Stories
Engadget for the iPhone: download the app now
AOL Tech

Firefox Friday: invasive, all-inclusive interview with Asa Dotzler, Mozilla's community director

Good afternoon! This week we're fortunate enough to have the Mozilla Director of Community, Asa Dotzler, here with us in the Download Squad bunker. His yappy little dog doesn't like it much down here -- but with soundproofed cells, who cares?

After my last interview with his comrade Aza Raskin (what are the chances of having two people called Asa working in the same office?), I was keen to find out a little bit more about Mozilla itself. Raskin knows a lot about how Firefox works, but Dotzler is a cornerstone of the the framework that supports Firefox -- Mozilla. In many ways it is Dotzler's work that has shaped the direction of Firefox's development, from its early, messy roots at Netscape, through to today where each Firefox release is the cumulative effort of thousands.

In today's interview I cover a truly epic range of questions: from HTML5 support and Jaegermonkey, to the future of Firefox's community efforts; from Firefox's roots as an IE-with-add-ons-and-pop-up-blocking, to its future in the enterprise environment. Asa has tried his very best -- and succeeded! -- to respond with an unprecedented level of detail about the inner workings of Firefox and Mozilla

Read this interview and you'll walk away enlightened, entertained, and slightly in awe of the all-encompassing might of the Mozilla machine.

Download Squad --
Firefox has always been a 'community effort', funded by the goodwill of the people, and driven by the champions of open source. Now that the browser war has turned commercial and proprietary again -- Silverlight, H.264, Flash -- can Firefox keep pace? Google is pouring lots of time into Chrome, to build a platform; IE9 finally wants its larger slice of the pie back. Can the community-driven effort of Firefox win?

Asa Dotzler -- Absolutely Mozilla and the Firefox community can innovate and continue to deliver the best browser in the world. Just take a look at what our Mozilla Labs community has been doing over the last couple of years with the Concept Series and Design Challenges. Innovations coming out of that community are increasingly finding their way into Firefox, as we saw with Personas in Firefox 3.6 and we'll soon see with Firefox Sync, Test Pilot, and JetPack all coming to Firefox 4.

But, the proof of the pudding, as they say, is in the tasting and Firefox's continuing strong user growth after the release of IE 7 and IE 8 and after the release of Google's Chrome almost two years ago should be a pretty good demonstration of our strength. To be more precise, and to repeat something I noted recently at my blog: for every user that Chrome's added in the last year, Firefox has added two and a half. I have no doubt that Mozilla and Firefox are still the strongest player in this space and that's because we are a community effort with thousands of dedicated contributors and a full-time staff that's unmatched in browser development experience.

If this becomes a money war, with either IE or Google pushing in their chips, how -- and on which front -- will Firefox choose to compete?

Mozilla has been competing with Apple, Microsoft, and Google for many years now. This is not some new phenomenon. I'll remind folks, Chrome is nearly 2 years old already and Microsoft and Apple have been working on and shipping browsers for pretty much Firefox's entire lifetime.

How do we do it? By providing a better experience of the Web for more people. As long as we continue to do that people will use it and people will contribute by telling others about Firefox or joining the project to make the next version even better.

This one's about Jaegermonkey's progress. Firefox 3.7 (4.0) is about half the speed of developer versions of Chrome or Opera -- do you have any plans to rectify that? Is raw speed even an issue when bleeding-edge speed junkies have probably already jumped ship to Chrome? In actuality, does the bulk of your 370 million users actually care about JavaScript speed?

First, JavaScript isn't the only or even the best measure of performance on the Web. It is a very important measure for rich, interactive Web applications and we care very deeply about making those as fast as possible. More and more people are using sophisticated Web applications but for the overwhelming majority of content on the Web today, the millisecond differences in modern browser JS performance simply isn't the deciding factor.

That being said, we've got a plan and we're executing on that plan to have a wicked fast JS engine for our next release. Firefox 4 will see the debut of Jaegermonkey, a new version of our JavaScript engine which adds what's called a method JIT -- a performance optimization that the other browsers have already shipped. We're adding that to the an optimization we're already shipping called tracing. When we've fully integrated and optimized the method JIT, we should be neck and neck with the other modern browsers and then our tracing optimizations should be able to give even more horsepower on top of that.

But as I said, JS isn't the only performance measurement. Firefox 4 will see several huge improvements to performance that are completely unrelated to JavaScript. First, we've got hardware accelerated text and graphics coming. Being able to offload rendering of text and graphics to the computer's GPU (a piece of silicon specifically designed and optimized to this set of tasks) will mean much faster rendering and less tax on the CPU. We've also added a brand new HTML5 parser that will help in both raw speed and responsiveness. We're doing Video decoding and display through OpenGL and Direct3D so that's going to just scream as well. And, debuting in Firefox 3.6.4 and included in the upcoming Firefox 4 as well, we've got plug-ins running in a separate process so that they can't slow down Firefox's responsiveness like they could in the past.

So, as you can see, there's lots happening on the performance front for Firefox 4, not just Jaegermonkey. Put that all together, and you've got one blazing fast browser. I really can't wait to get this next release into the hands of hundreds of millions of users.

In terms of manpower, are you directing more effort towards 'old' features like Jaegermonkey, or new ideas like Contacts and Account Manager?

We have the resources and the mandate to do it all. This isn't really zero sum because it's a completely different set of skills and people that work in these different areas. People who work on deep technical issues like compilers and rendering and scripting engines are few and far between and we have the best teams in the world doing these things, bar none.

We've also got a fantastic group of people in Mozilla Labs who are, along with a large community of volunteers, working to understand and design a better experience for, among other things, what I call "people in the browser", including Contacts and Account Manager.

As community director, are there things you can do with your 370 million users (and academic ties?) that Google cannot? What is the next step, after Test Pilot, Mozilla Labs Design Challenges and the Concept Series?

To answer that question, I have to describe a bit what the Mozilla universe looks like. I like to think of it as a series of concentric circles (or spheres if you've got good 3D rendering ;-) Rounding the numbers a bit to simplify, here's what it looks like. The biggest circle is our users, nearly 400 million strong today. These people decided that the default browser that came with their computer was insufficient and took steps to download and install a new browser.

Some of those users took the added step of becoming beta testers, about 4 million of them. This circle of people try out updates to our software about once a month and give us various types of high level feedback.

Inside that circle are the group I call "advocates". This is probably about 400 thousand people and includes the 100 thousand Spread Firefox members that have put up Get Firefox buttons at their sites, the fans that are regularly spreading the word on Facebook and Twitter, and people helping other Firefox users with support issues.

The next circle in is our daily testers and Test Pilots. These people, about 40 thousand strong, test and report feedback on the changes that are happening to Firefox every single day as well as offer structured feedback on particular features or design changes through our Test Pilot program.

Inside that circle are the 4,000 or so dedicated contributors who are doing everything from writing patches to helping to translate Firefox and Mozilla websites to participating in user engagement campaigns, to submitting feedback on Design Challenges, etc. A stand-out group here is the nearly 80 localizations of Firefox which are developed by an army of volunteers across more than 100 countries with only a small handful of full-time staff doing coordination activities.

Finally, there are about 1,000 people at the core who are either full-time contributors or critical volunteers who we depend on every day to lead the project forward.

We are able and willing to engage the smartest and most passionate people from all over the world, from casual involvement to the deepest levels of participation, because we're willing to give them real responsibilities in the project that can have a meaningful impact on the lives of hundreds of millions of people using the Web.
Upgrade the Web

Your first and perhaps memorable project was Spread Firefox. Is it still going? Where are your day to day efforts directed now? Will we ever see Firefox CDs through our letter boxes? How about a little Thunderbird lovin'?

Yes, Spread Firefox is still going strong. It's a gathering place for hundreds of thousands of Firefox advocates and the hub for long standing programs like Firefox Affiliates with its tens of thousands of members and Mozilla Campus reps with over 4,000 representatives at over 200 schools in more than 60 countries. It's also the home of several hundred community marketing projects, and our global events planning and community calendar. Finally, it's the launch point for major engagement campaigns like Download Day 2008 where more than 8 million Firefox users helped us set the Guinness World Record for most software downloads in a single day and Five Years of Firefox in 2009 where we celebrated Firefox's anniversary with hundreds of parties all around the world.

As for me, I'm only peripherally involved in Spread Firefox these days. I spend most of my time working with the Developer Engagement team and also picking up random but important tasks that otherwise wouldn't get done. As part of Developer Engagement, I work with a phenomenal group of people who are helping to build awareness of the Open Web platform as well as connecting Web developers to Mozilla engineers so that we can improve that platform in ways that the Web developers most need and want. When I'm not doing that, I'm doing things like helping Mitchell and Harvey with the Microsoft EU ballot thing, trying to better measure and understand our community health, or blogging my thoughts on the competitive landscape.

Netscape. No doubt many users of the Internet have never even heard of it, yet that's where Firefox begun! Do you have any fun stories to tell from the downfall of Netscape and the spinning-off of Mozilla? Is there a code fragment from Netscape that lives on to this day?

You're right about lots of folks never having heard of Netscape. The Web has literally doubled in users since Netscape disappeared. Firefox today has about 10 times more users than Netscape did at its height and Firefox accounts for somewhere between one quarter and one third of all Web users. It's a completely different world.

As for fun stories, there weren't many of them back in those days. The Web was in bad shape, users were hurting from pop-ups and spyware and viruses. Microsoft had dissolved most of the IE team after shipping 6.0. And most people didn't know what a browser was and weren't willing to download software from the Web.

I will offer this one story. It's not really fun, but it's informative and I think it offers a lesson that's probably even more relevant today than it was back then.

In 2002, Netscape released Netscape 7.0, a browser that was essentially Mozilla 1.0 with a Netscape-branded theme. But they went a step further than just altering the theme. They disabled Mozilla 1.0's very popular pop-up blocking feature. The reason they did this is because Netscape, as a business, was no longer a browser company. They were an Web content company that depended on advertising revenue from their then-popular portal, netscape.com as well as dozens of other AOL Time Warner web properties. The pop-up blocker threatened that revenue so before they shipped a re-branded Mozilla 1.0 as Netscape 7.0, they removed that feature.

Now, the tech press knew that Netscape 7.0 was just Mozilla 1.0 with a different theme and missing Mozilla's pop-up blocker, so they excoriated Netscape. They were brutal. And they, to the very last one, recommended Netscape users just go get Mozilla 1.0.

Netscape saw the reviews and quickly revved a new version, Netscape 7.0.1, "now with pop-up blocking". But they couldn't just do that. They went into the settings and turned on exceptions, allowing pop-ups for netscape.com and all of the other top AOL Time Warner websites. Users downloaded the new Netscape 7.0.1, "now with pop-up blocking" and fired it up and guess what. The first page to load was netscape.com with its pop-ups.

Now Netscape wasn't stupid. Certainly none of the people I knew there liked this at all. But Netscape's business model depended on a bad user experience, pop-ups, and so they had no choice but to make those decisions and to ship the product they did. It's really quite sad, and at the same time maddening.

And this is a problem I see in the browser world today. Most of the big companies making browsers are going to find themselves confronted with situations like Netscape faced in 2002. Only back then the concern was just something minor like the annoyance of pop-up advertising. Today, the concerns are much more important and include things like user data and privacy.

Mozilla is a global non-profit whose mission is to build a better Internet. There are no conflicts between providing the best possible experience to our users and our other businesses because we have no other businesses.

OK. That wasn't a particularly funny story. But it's highly illustrative why I and many others come to work on Mozilla each day with smiling faces and why Mozilla is more than just a browser. We're a movement to promote openness, innovation, and opportunity on the web and we're dedicated to the best possible online experience for everyone, everywhere. It's a great feeling to know that your are charged with doing the right thing to the best of your abilities and that you won't ever be asked to compromise that to satisfy some unrelated business interest.

I remember Firefox 1.0 being touted as the 'lean and mean' version of the Mozilla browser. Do you intend to remain the 'feature rich' browser, leaving the light-and-fast crown with Chrome?

We designed Firefox 1.0 to have the right set of features to provide a compelling and empowering experience for the largest possible audience. I think we still believe that today.

Enterprise. Back in 2005, just after Firefox 1.0, there was a lot of talk about whether Firefox could replace (or supplant) Internet Explorer on an enterprise level. Fast forward to today: do you see multi-national businesses with Firefox installed? With increased competition at the consumer level from Chrome, are you going to try and get Firefox into the workplace before Google and its Chrome (or Chrome OS!) can steal your thunder?

I know of a number of Fortune 100 companies that support Firefox and a Forrester study released just this month puts Firefox at about 20% usage in the enterprise. It's not a specific target for Mozilla, but it's tracking our consumer growth pretty closely and with announcements like the one from IBM just this week, I think we'll continue to see growth there.

This one's about Firefox as a platform for Web apps. The recent Identity Concept Series definitely hints at Mozilla moving Firefox in the direction of a 'platform' rather than dumb browser with add-ons. How do you feel about Google's Native Client (NaCl)? Will Firefox embrace this move to give the browser more control, more low-level access?

Well, I think Firefox, or more specifically, Gecko (the part of Firefox that shows you the Web) and all of the other modern browsers are already quite sophisticated platforms. Gecko, our take on the Web Platform, already supports everything from the traditional HTML, CSS, JavaScript and the DOM, to newer advances like audio and video; WebGL, SVG and Canvas; device access for orientation and geolocation; downloadable fonts and CSS3 awesomeness like gradients, box and text shadow, and transforms; drag and drop and the file API; Cross-domain XHR and send and read as binary; online/offline events, local storage and app cache; native JSON, web workers, and script defer and async; and much much more. That's a pretty rich platform already and we're not even close to finished.

I'm not a big fan of Native Client. Rather than replacing the Web Platform with a new native code platform running inside a browser, which I think is a big step backwards, I think the Web Platform should continue evolve to that rich experience that users get with desktop applications and want in their Web applications. And that's happening and Mozilla is helping to define that future through the appropriate standards bodies and through innovations in our Gecko Web Platform.

With Firefox 4, we'll support even more of the next generation Web platform, often referred to as "HTML5", and with our screaming fast Jaegermonkey JavaScript engine, and by hardware accelerating the hell out of pretty much all the graphics, I think we're going to give developers an awesome platform for their applications.

So I'm not a fan of returning to the old world of compiled code. I say, why go "native" when you can go "Web" :-)

Now let's talk about you. You don't have much of a marketing background, yet you spearheaded one of the most successful technological marketing movements ever. Have you just learnt as you've gone along? How did you end up being the Director of Community?

My background, before Mozilla, was actually architecture and historic preservation/restoration. I really had no formal education in technology or in marketing. But I grew up on computers and I've always loved their amazing power to connect people, whether it was sharing C64 games on floppy disks with friends or today's Internet.

I got involved with Mozilla in 1998, the only way a non-programmer could, by testing and reporting bugs. That evolved quickly and I found myself leading a testing community of thousands of volunteers. That led to a full-time job with Mozilla in 2000 organizing and leading that community of testing volunteers.

Over time, and as Mozilla shifted from being a technology producer into an organization that was building products for consumers, we saw the need for additional non-code communities. While still doing testing and other work, I helped build our project management group, drivers@mozilla.org and the community of people participating in determining what it was we wanted to our browser to be. After that, and while doing project management for Phoenix, the browser that would become Firefox, we realized the need for a marketing community so I helped launch Spread Firefox to do for marketing what we'd done already for coding, testing, and project management.

Since then, I've been involved in helping to build and support additional communities including our community of Mozilla-connected Web developers, end user support, legal, and various others.

We've been pioneers in just about all of this so we've had to learn as we go. There was no program in school that I could have followed that would have fully prepared me to do any of this. So, yes, I've just learned as I've gone along. It does help to be surrounded by the smartest people in the world.

Do you have any tips for other open source projects that are trying to attract mainstream attention? How did you go from 'open source code monkey' to where you are today?

Lots of other open source projects have followed in our footsteps and have had good success with some of our methods and process, but none that I know of have achieved the same connection with end users.

I think there are a few key things that do stand out for me though. First, we designed Firefox for other people. It wasn't designed for the people making it. It was missing lots of the geekier features I and many many others on the Mozilla team loved in the older Mozilla application suite. But it was the right set of features for our target audience at that time -- Internet Explorer users.

(Extensions [add-ons] were the compromise that made it work for regular folks and for power users and even for the developers building it, all of whom have different requirements.)

Second, Firefox improved on the experience IE users were accustomed to rather than totally replacing it with something foreign and confusing. We adopted as much of their experience as we could and then extended it to give users more control and more fun.

Third, and I think this one was crucial since we were trying to replace an entrenched product, we migrated all of their data over to Firefox so they weren't starting fresh. There's nothing worse than starting up a great new app and not being able to find your data, things like your favorites and logins.

Finally, I think it's really important to be engaged with your users for the long haul and to give them the opportunity to help shape your product as it goes forward. It really is a partnership we've developed with our users. They trust us to keep them safe and secure and to provide the best browsing experience and we trust enough to invite them to participate in the process of making the next versions of Firefox even better.

Finally, a bit about Mozilla itself -- I've always pictured Mozilla as somewhat of a 'new age' beardy operation, an incubator for the Richard Stallmans of tomorrow. Do you guys just sit around an open-plan office and throw ideas at each other, or are things a lot more formal now? Is there a fixed path for new ideas from inception to inclusion in a final release of Firefox, Thunderbird, etc.? Do you actually have beakers of fuming, bright-coloured, chemicals in Mozilla Labs?


Mozilla isn't random and informal, neither are we strict and inflexible. There are solid policies around certain things like code review but we're pretty loose about other things like work attire. We don't make a lot of rules except to codify best practices so that we don't have to repeat ourselves so often and so that the project can outlive any one person. We do have a hierarchical structure, but rank is determined by merit and nothing else.

We're also quite distributed. Mozilla employs about 250 people (at my last count) but depends on the contributions of thousands of volunteers for every Firefox release. We have offices in the US, Canada, France, China, Japan, and New Zealand and we have people contributing in nearly 80 languages from more than 100 countries. So what ever you might see if you visited, say our office in Toronto, is just a snapshot of one place and one time and probably won't fairly represent "Mozilla" as a whole.

That's all to say that it's not particularly easy to characterize what Mozilla looks like. We're simply too diverse for that.

As to how things get into the product, there are at least a few common paths for new ideas to make it into a Firefox release. One path is through add-ons where some great idea gets its start as an extension, becomes popular and demonstrates it's a value to a very large number of users, and is eventually migrated into Firefox. This was actually the case for several amazing features including our first Tabbed browsing implementation. The AwesomeBar and Session Restore also came to Firefox down this path.

Another path is a designer or a developer has an idea, files a bug, creates a patch, gets reviews and lands it in our development tree and it's then evaluated by the product leads for inclusion in an future release. An example of this was the Find Toolbar, which started out as whiteboard idea between a couple of Firefox developers, and was accepted as a late addition to Firefox 1.0. Many features make it into Firefox this way.

A third path is through Mozilla Labs where an idea can be carefully incubated, iterated on, and then, if appropriate, shepherded into Firefox. This was the case for Personas and will be the case for some version of Test Pilot, JetPack, Weave, and Contacts in future Firefox releases.

If you had to guess (or if you have a metric?), what percentage of Firefox features begin their life in in the Mozilla offices -- and what percentage do you think your community has been responsible for?

It's hard to come up with a percentage because it's hard to define a "feature". So, there are a few ways to look at this. One way is simply by code. Historically between 30% and 40% of the code that makes up Firefox has come from volunteers. But that's a pretty narrow view that misses things like 100% of the non-English versions of Firefox which were developed by volunteers or some very well known features like Tabbed Browsing and Session Restore that were developed first as volunteer-created extensions. There's no doubt in my mind that Mozilla would not be here today without the amazing contributions of our volunteer community.

* * *
Many thanks to Asa for taking the time to really dig into the fleshy bits of Mozilla and Firefox. If you have a question that hasn't been asked here, leave a comment -- there will certainly be more interviews with Mozilla staff in the future!
Share submit to reddit StumbleUpon.com

Tags: asa dotzler, AsaDotzler, browser, community, director, features, firefox, html5, interview, jaegermonkey, javascript, localization, mozilla, netscape, open source, OpenSource, security, translation

Comments

3