Monthly Archives: December 2005

Interview with Mozilla’s User Experience Lead Mike Beltzner

As promised, here’s the interview with Mike Beltzner. Although the bulk of the interview was done using instant messaging sessions between the 22nd and 25th of November, the final editing touches were delayed for a few weeks. Here’s what Mike had to say. (Frequent visitors of my blog may want to reload to get the proper formatting, as Firefox usually doesn’t check for updated CSS every page visit.)

First of all, thanks for taking the time to do this interview. Tell us a bit about yourself!

I’m a 29 year old Canadian; grew up in Queen’s University in Kingston, Ontario, and am now living in Toronto. At Queen’s I took two degrees, one in Cognitive Science (Faculty of Arts & Science, graduated in 1999 with a BScH) and another in Education (BEd).

Why two degrees?

It was a “concurrent degree program”, and I was interested in the mix of subjects. When applying to schools, I applied to totally different programs at three different universities. A multidisciplinary program at McMaster University, computer engineering at the University of Waterloo, and Concurrent Education at Queen’s. I eventually decided that I liked the idea of exploring the realm of human information processing from both sides: Cognitive Science examined the question of “how do humans process information?” and Education examined the question of “how do we teach humans information?” For a while I thought I was planning to be a high school teacher.

But then you changed your mind?

Well, I enjoyed teaching, and by all reports I didn’t suck at it. The reason I decided to look into other options was that the teachers’ unions in Ontario were constantly fighting with the Ministry of Education, and I worried about how I’d have to pick sides in that debate. I felt that students were the ones who were eventually suffering, and I didn’t know if I could support a strike action if I was a teacher. It’s somewhat sad, but based on that factor I ended up with an education degree and very little desire to put it to work.

So what happened then, after you graduated?

After I’d decided not to become a teacher, I started looking for IT jobs in Toronto. My cognitive science program had talked about the field of Human Computer Interaction, but there was little focus on careers and professional options, so I wasn’t really aware that there were jobs in that domain. I applied to several companies for positions in knowledge management, software design, technology communications … things like that. At a job fair on campus, I told an IBM representative the sorts of things I was interested in, and he passed my resume along to the User Centered Design group located at the IBM Software Lab in Toronto. I was interviewed for a type of job I didn’t even know existed! When I was offered the job, I jumped at the chance.

I spent 5 years working on IBM software development tools: the first Java Development Tools for Eclipse, as well as the product that is now sold as WebSphere Integration Developer, a suite of Service Oriented Architecture development tools for business process based application development. My job there included working with the user interface development teams, the graphic designers and the architects to ensure that our products met the needs of our users. I was able to get a lot of experience with the User Centered Design methodology, and conducted many focus groups, user feedback sessions and customer visits. It was a really great job.

What brought you to Mozilla? What made you leave IBM?

At IBM I became good friends with several of my colleagues, one of whom was also a good friend of Mike Shaver’s (one of life’s ironies: Shaver was living in Kingston when I was at Queen’s, and despite having intersecting social circles, we didn’t meet until years later when we were both in Toronto.) Mike and I became friends as well, and I was always interested to find out what he was up to, and to learn more about the open source software community.

One day I got a call from Mike Shaver (who was working for Oracle at that time) saying that the Foundation was looking for someone to fill a full time position to focus on user experience issues. He suggested that if I were interested, I should forward a resume. I was indeed interested, and being a bit of an overachiever, I put together three resumes that night and sent them the next morning. Before I knew it, I was on the phone with Brendan Eich talking about what sort of things someone in a User Experience Lead role might do for the Foundation. There was an interview process, and when the offer came, I faced a pretty big decision: continue on with the very promising (and yes, interesting!) job I had at IBM, or make a jump over to Mozilla.

Now, let me be clear that I think that IBM is a great place to work. I really enjoyed my time there, and got the opportunity to work with some incredibly smart people there. At the end of the day, though, working with Mozilla was too tempting an opportunity to pass up. I love the idea that I’m helping to refine an entirely new model of software and business development. I love the fact that our single mission is to make things better for the user. How could a usability guy not love that?

Indeed. What do you think will be your contribution to Mozilla?

There’s a bunch of things that I would really like to do, some of which I mentioned on my weblog when I first started. Obviously I want to improve our existing user experience in Firefox and Thunderbird. Don’t get me wrong, it’s good now! But there are lots of little things that can make it better. A longer term goal of mine is to integrate design and usability methodologies into the open source development process. Ben Goodger and the Firefox 1.0 project drivers took some great steps towards this goal, but I want to keep going and ensure that Mozilla offerings are always driven by user needs. So we should start by building a clear picture of who our users are (there’s likely more than one “type” of user, of course) and what they want to do with our technology, our websites and our software. That will help us prioritize features, ensure that we’re innovating in the right directions, and also help us with the design of those features. If we end up hitting questions about the design, I’d like to continue to build partnerships with organizations that can help us run usability studies so that we can get real data from user feedback instead of making assumptions about how people want to use their software.

And that’s where Google enters the picture?

Amongst others, yes. Google is obviously an excellent partner with whom we have very strong relations, but they’re not the only ones with usability research facilities who’ve offered their services. Novell, for instance, was one of the first companies to get in touch with me when I started at Mozilla, asking about what areas of the browser we had the least user data on so that they could contribute.

There is a great difference between a person capable of using Firefox, and a person capable of actually downloading and installing it. Depending on the way Firefox is distributed (e.g. OEM deals with vendors), isn’t everyone a potential user?

Definitely. Traditionally a Firefox user was someone who was familiar and comfortable enough with computers to find, download and install the browser. We’re actively investigating other means of distribution, and community efforts like and have also expanded our reach to the desktops of people who might not be comfortable installing software on their own. So now, anyone who uses the web is a potential user of Firefox. That makes for a pretty wide-ranging group of users, though, and so it’s important to segment that space into sensible “personae” that represent the needs, tasks and goals of different types of users. Socioeconomic factors, individual differences, general computing familiarity, goals for using the internet … these are all factors that could vary wildly within our user group. I really want to get a better handle on who those different personae are and what it is that they’re trying to do with our products, then make it easier for them to do so.

So, how long have you been working for Mozilla?

My start date at Mozilla was July 11th, 2005. My contract says I’m a “User Experience Lead”, but my business card says I’m a “Phenomenologist”, and really, we try not to get boxed in by titles. We’re still trying to completely figure out the reporting structure, but I spend a lot of my time talking with Chris Beard (Product Manager), Mike Schroepfer (Product Engineering) and John Lilly (Business Development). Mike Shaver (Technology Strategy) sometimes asks me to look at a topic or emerging technology.

What was your initial experience when starting to work for Mozilla?

It was very different going from an office building with 2500 employees to my home office with an IRC client. I’d met a lot of the other Mozilla Foundation employees during interviews, and I’d been getting more and more involved with the community for so I was able to hop on and get started pretty quickly. Mike Connor (mconnor), especially, was very welcoming, as were a lot of people on IRC like Benjamin Smedberg (bsmedberg), Asaf Romano (Mano), Gavin Sharp (gavin), Adam Guthrie (ispiked).

Within the first few days I’d been pointed to some bugs that could use my attention, and was spending a lot of my time getting caught up to date on issues regarding the Firefox 1.5 release. I was also getting used to managing the new information streams: I’d built up 5 years of organizational knowledge at IBM, and was starting entirely from scratch again. In many ways Mozilla is both a smaller and larger organization than what I was used to. At IBM I was working with a development team of about 75-100 developers, but there was a pretty clear structure of who did what, who reported to whom, and what everyone was working on. Needless to say, it’s harder to build that picture in an open source environment. There are a few key names within our community, but our developer network is vast and determining who has which interests and responsibilities continues to be a challenge.

Since I started, things have jumped around a lot. My first project was helping Deb Richardson with the design of the Mozilla Developer Center, which is simply a stellar resource for people who want to work with web technologies. Then I did some work reviewing bugs and helping guide UI development for the 1.8 branch, and continued to work with drivers in determining relative priorities for the 1.5 release. For the past month I’ve been focusing less on bugs and more on whole experience things surrounding our launch efforts (web page design, messaging for release, keeping momentum, working with partners on extensions) and most recently have been working on Firefox 2 planning and road mapping with Mike Connor and Mike Schroepfer.

Do you still work from home?

Mike Shaver, Mike Connor and I all share a small office in downtown Toronto. Shaver travels a lot, and Mike Connor only comes down from his home office once or twice a week, so sometimes it’s just me here, but it’s really nice to get that collegial atmosphere going once in a while. It’s also nice because my wife is a law student at Osgoode Law School at York University, so she gets full use of our home office now.

What are the basic plans for Firefox 2?

If you haven’t already, you should be sure to read over the various roadmaps that have been posted to Planet Mozilla and on the wiki. A good place to start is with Chris Beard’s post about our new strategy for product and platform releases, and from there take a look at the draft platform roadmap document. The question is about Firefox 2, though, so you should also be sure to stop in at our draft plan for Firefox 2 and give us feedback on the discussion page. Even if an idea for a feature doesn’t make it into the next release, we’re still interested in collecting and brainstorming new and exciting features.

As we looked at where today’s web was, and how more and more services are emerging that leverage web technologies, Mike Connor and I became convinced that the theme for Firefox 2 should be helping the user manage the vast amounts of information that their browser connects them to. This would leverage a lot of the in-progress work on an update to our bookmarking infrastructure; the idea being that we’d enhance the user interface to make it easier to find places you’ve been before, updates to things you’re interested in, and even places you might not yet now about but would probably be interested in. Also, there’s a lot of “fit and finish” work that I’ve personally got my eye on with respect to standardising terminology in our strings and layouts in our dialogs.

How would this new bookmark system actually work?

That’s a very good question. In my opinion, this is an excellent example of an area in which Firefox could benefit from some user centered design activities. Robert Accettura’s blog post about intelligent bookmarking makes some very good points about the limitations of our current system and potential solutions. I’d like to take that further, and actually find out how people are using their bookmarks. How do those usage patterns differ between people who are more and less comfortable on the web. How about people who use the web for work as opposed to only in their spare time. I’d like to gather that information and make sure that the resulting design is based on the needs and requirements of our users.

I read “tag-based system” instead of “heirarchical”. Does that mean no bookmark folders by default?

Axel Hecht recently wrote a blog entry lamenting the potential death of hierarchical bookmarks. I’m sure that he’s not alone. We’ve spent the past two decades promoting a hierarchical (or spatial containment) desktop metaphor for computer filing systems. It would be a disservice to many of our users to replace it completely. There are, however, significant advantages to tagging systems, especially in terms of building a system which defies classical ontology (for more on those advantages, see Shirky: Ontology is Overrated — Categories, Links, and Tags.) Adding tagging capabilities to bookmarks can be done in a way that is based in the existing user base’s conceptual (hierarchical) model, yet extends it to add richer interaction possibilities. That, I think, should be our goal.

Note: After this interview was conducted, Ben Goodger published a detailed feature plan for Places, which encompasses all of what we’re talking about.

Looks to me Firefox 2 will be a very interesting release. Any other cool ideas worth mentioning?

As I said before, we’re doing all the planning for Firefox 2 in the open so you can really see for yourself. I’m personally excited about fixing the way we communicate all forms of status with the user, and hoping we can find some ambient ways of passing that information along. My hope is that the browser continues to innovate quickly (did I mention we’re aiming at a July 2006 release date?) but carefully, remembering that there’s a large user audience out there who simply wants to browse the web. Some of the really cool cutting edge stuff out there is coming in the form of extensions. We’re starting to see some interesting ideas coming from the community as well as partners like LinkedIn and With the new Canvas feature, in Firefox 1.5, I’m really quite excited to see what our developer community comes up with.

What’s your favorite operating system?

I’ve been a daily Windows user for the past five years, but after three weeks of using a PowerBook, I’ve fallen in love with Apple’s OS X. The reason is that for common tasks, OS X just works. It makes common tasks a snap, and it’s incredibly easy to tinker with it. For example, Mike Shaver and I were looking for a way to let anyone in our office control my iTunes, which is hooked up to our speakers. A little bit of Google, a little bit of Applescript and a little bit of Python, and he’s well on his way to an AJAX web client that can be used as a web-based remote control.

With your experience in user interface design, the answer isn’t surprising. However, you’d expect an open-source developer to prefer open-source operating systems like Linux?

Actually, I would expect someone to pick the tool that best fits their needs. Obviously I endorse the open source development model, and believe in the advantages that can be obtained through that system. That’s not the only criteria for picking an operating system (or other piece of software), though. I think it’s clear that users want software that makes them more productive, and right now, the majority of the world (myself included!) can’t be as productive as they’d like to be with the open source desktop, for a variety of reasons. If there was an open-source operating system that “just worked” as well as OS X or Windows, I’d be happy to use it.

Finally, do you think Firefox will eventually become the most popular web browser, putting Internet Explorer on second place?

I think that if we can get people to realize that they have a choice in web browsers, and that the choice is worth making, then we have a chance of significantly increasing our overall popularity. The goal of Firefox — indeed of Mozilla — is to promote innovation and choice on the Internet. Firefox succeeded in providing that choice and innovation, and our current position reflects the fact that there’s a core group of users out there who are actively choosing us as their browser. Our next challenge is to get more people to realize that there is a choice in web browsers, and get them making that choice based on the criteria that matters to them. At that point, I’m pretty confident that our “popularity” will continue to climb, as there are some very compelling reasons to choose us. Of course, I need to keep working to make sure that one of those reasons is a better user experience.

Thanks, Mike, for taking the time to do this interview! It’s been fun.

My pleasure, and thanks for getting in touch with me! I hang around on (beltzner) and can also always be reached through email at beltzner at