Internetling

If it were not for Mozilla and its numerous open source projects that saved us from the IE Dark Ages, the browser market today would definitely be very different. In 2010, the Firefox browser is facing some pretty tough competition from the likes of Google Chrome, Internet Explorer 8 (v9 is actually good), Safari and Opera, all modern and feature-packed web browsers. With Windows’ new ballot screen, things might become even more interesting.

The Web Browser is evolving at a mind-bogglingly rapid pace, and the changes it went through in 2009 only are incredible. That is why I was really happy for a chance to talk to Asa Dotzler, community coordinator for Firefox marketing projects, who has been with Mozilla for 12 years. This is my first interview and I hope you’ll enjoy reading Mr Dotzler’s answers as much as I did. We cover Firefox, Mozilla and the new Web. Don’t forget to check out Asa’s blog for even more on the topic.

The interview:

In your personal opinion, what is Firefox’s flagship feature?

Probably not the kind of specific answer that you’re looking for, but I think that customizability is Firefox’s flagship feature. This feature takes many forms in the product but add-ons are clearly at the top of that list.

I don’t believe that any single browser feature set can satisfy everyone. I don’t believe that there’s even a “80%” case to strive for. Everyone engages with the Web in their own way and a browser that individual human beings can bend and mold to their own, very particular, use cases is a powerful tool for making the Web better for people and that’s what Mozilla’s all about.

In addition to customizability, I think another very powerful “feature” of Firefox is why and how it’s made. Firefox is not here to further some business interest. We have Firefox because Mozilla’s public-benefit mission is to promote choice, innovation, and participation on the Web. Firefox is a browser you can trust because of who makes it, why they make it, and how they make it. The public-benefit goals, combined with the open source development process, managed by the not-for-profit Mozilla Foundation, means that users never have to worry about ulterior motives. Firefox is a browser you can trust.

Benchmarks aside, Firefox still feels much slower than other browsers, except IE. Do you think Firefox’s slowness hinders wider adoption? Is Gecko the problem?

I don’t believe that Firefox feels slower than other browsers. When I’m in Web apps, Firefox works as fast or faster than all of the other browsers I use. It also includes features and offers extensions that let me accomplish my online tasks much faster. One example that you can try yourself is scrolling a long spreadsheet in Google Docs. Firefox destroys the other browsers in that use case, even Chrome is awful there.

So, performance is also a very personal measure. If Firefox is really fast compared to other browsers at something you do often, like working in Google Docs, then you’ll think Firefox is great. If Firefox is slower than other browsers at something you do often, like starting up the browser, then Firefox might not seem so great. We have to get better at some things like startup speed, but we’re leading the pack at other things.

Also, some people who are on very old hardware or who have installed certain add-ons can find Firefox slow. My advice there is to try to move to more modern hardware and to be careful about which add-ons are used. It’s also important to keep those add-ons up to date — especially plug-ins like Flash which cause significant slow-downs and crashes in Firefox. Fortunately, the next version of Firefox won’t crash when Flash or other plug-ins do. You can even see this in action today in the Firefox developer preview builds.

Benchmarks of rendering time and JavaScript execution put most browsers all very close to each other and for most use cases on the web the difference isn’t really perceptible. We’re much faster at some things and not as fast at others, but we’re always working improve and Firefox has gotten consistently faster with every release.

Will we ever see a WebKit distribution of Firefox?

WebKit could never replace Gecko. One thing that I think people don’t understand about browser platforms is that there are actually a lot of pieces. WebKit in Safari is not the same as WebKit in Chrome for example. Both of those browsers have their own JavaScript engines, their own rendering and graphics layer, their own networking code, their own security services, etc., etc. Chrome on Linux actually uses Mozilla’s NSS security module because that OS, unlike Windows and Mac, didn’t provide them with one.

Mozilla’s Gecko also provides lots of features not available in WebKit, like XUL which is the foundation for most of Firefox’s extremely popular extensions, and a cross platform networking stack and security module that we can improve as we need without waiting on the OS vendor to release new versions. In addition to those long-established features, we’re leading the way on many Web standards where Firefox’s implementation is actually informing the standards and helping to work out cross-platform and cross-browser compatibility issues. The flexibility and feature set we have with Gecko makes it an ideal platform for Firefox and I can’t imagine WebKit offering that any time soon.

That being said, just as Google is borrowing from Mozilla’s code, Mozilla is and should be borrowing good ideas from the WebKit project. Having several different implementations of an agreed upon set of standards is a good thing and I’m thrilled that Apple, Google, Microsoft, Opera, and Mozilla are all competing to have the best implementations. With competition like this, we can insure that the industry never stagnates like it did when Microsoft stopped development after the release of IE6.

Web video: HTML5 with proprietary codecs or Flash? Do you think Ogg Theora has a chance, despite its poor performance in comparison to H.264?

There are really two discreet issues here. The first is the HTML <video> tag. The <video> tag is a first-class Web citizen that integrates video with the rest of a Web page because it is scriptable with JavaScript, stylable with CSS, and fits in the standard web page as a normally functioning block-level HTML element.

Video has become an integral part of the Web, like images and text did 15 years ago. It deserves to have the same status on the Web as images and text. The <video> tag in HTML5 and the DOM media API give us that. This is a much better long-term solution for video on the Web than proprietary plug-ins which, in addition to performance and stability problems, suffer many compatibility issues because they are not “native” to the Web.

The second issue is codecs. I think the way to think about this is to consider the <IMG> element in HTML. You can put a jpeg image in the <IMG> element. You can put a png image in the <IMAGE> element. The tag doesn’t care about which of the several image codecs you might be using. Different codecs are good for different things. A jpeg, for example, is much more efficient than a png when you’re putting photographs on the Web. A png is much more efficient when you’re posting line art or other hard-edged graphics. A png can do transparency and alpha-blending so if you need that, you wouldn’t use a jpg.

Different video codecs have different strengths. (As I mentioned earlier, irrespective of codec, the <VIDEO> tag is a win for the Web.) The most commonly used video codec on the Web today, H.264, is a very efficient codec for all kinds of video and there are lots of tools available for creating and playing H.264 video. But it’s got some down-sides too. The most important negative, as I see it, is that H.264 is patent encumbered and requires a license and a fee for commercial use. The terms of that license are also subject to change at the whim of the group that administers the license. That makes it not well suited to the Web because participation on the Web shouldn’t be restricted to just those with money and lawyers.

Theora is another video codec that’s growing in popularity on the Web, thanks to support in Mozilla’s Firefox, Google’s Chrome, and the Opera browser (which combined make up about 30% of all Web browsers.) Theora is also an efficient codec that produces good quality video all the way up to HD. It is not patent encumbered and requires no licensing fee to use for producing content or consuming content. It’s a simpler codec than H.264 so it doesn’t match H.264 in quality under all circumstances, but it’s “good enough” for almost any Web use case and because of its simplicity, it requires less processing power. The downsides to Theora are that it doesn’t yet have support in IE or in Safari, except though third-party codecs or plug-in.

So, these two issues are closely related, but not inextricably bound and the <video> tag, regardless of the codec issue, is a very good thing for the Web. We do still have to sort out the codec issue and I expect we’ll see the landscape changing pretty dramatically this year so keep your eye on this one.

Browser ballot screen… good or bad?

I don’t know yet whether it will turn out to be a good thing or a bad thing. The European Commission was quite keen on this but Mozilla never advocated specific remedies. In our communication with the EC, we focused on the principles that the Mozilla community believed were at stake, the issues that any remedy should address. The first two, and most important to us were 1) Microsoft must respect choices users have already made, and 2) Windows should not give any technical advantages to IE.

These two principles seem to be well reflected in the final settlement between Microsoft and the EC and I think that’s great. How the ballot, and the many other terms of the settlement, turns out is something we’ll need more time to learn.

Fennec, the Mozilla mobile browser, supports Windows Mobile and Maemo. What is going to happen to the Maemo version now that it is essentially a dead platform because of the MeeGo merge?

Maemo isn’t dead. All of the work that we put into Firefox Mobile on Maemo will easily translate to MeeGo. The two platforms were very, very similar and that’s why they could merge so easily.

Firefox Mobile is also running on Android now, though not in a release version yet. (And, while we’re on the topic, Firefox on Android handily beats the Android browser in JavaScript benchmarks.) As with the desktop version, we’re going to target all of the major open platforms. On the desktop, that’s Windows, Mac, and Linux. On mobile, that’s Meego, Android, and Windows.

Why did the community embrace Firefox rather than the (at that time more stable) Mozilla Suite (SeaMonkey)?

The community embraced the Suite enthusiastically from 1998 through late 2002 when Firefox (then Phoenix) started to gain traction. I believe that the Suite was chasing Netscape Communicator users (compare the feature set) and there just weren’t that many of those people left in 2002. Firefox, on the other hand, targeted IE users and there were a lot of those in 2002.

Firefox took the best features of the Suite and simplified them and made them more accessible to “regular people”, many of whom came online long after Netscape was a distant memory. With Tabbed browsing, pop-up blocking, and extensions — all taken from the Suite but improved dramatically in usability, Firefox had what IE didn’t at a time when IE was suffering a major security compromise seemingly every week.

By 2003 and 2004, we had also managed to gain some major wins in Website compatibility. Gecko, Firefox’s rendering engine, was more compatible with Websites and, thanks to the efforts of Mozilla’s outreach programs, most major Websites had been upgraded to support Web standards and Firefox.

That’s a really interesting point, I think. Today, if you decided to build a browser, you don’t really have to worry much about Website compatibility. Thanks to the successes of Firefox in gaining market share and to Mozilla’s standards evangelism, most major Websites (with exceptions for a couple of Asian countries) work quite well with a standards-based browser. You might remember that when Apple launched Safari, it impersonated Firefox when identifying itself to Websites so that it could get the standards-based pages rather than the IE-only content. Now website compatibility, once a huge barrier to alternative browser adoption, is mostly a non-issue.

Let’s talk about Thunderbird. Is webmail killing the desktop client?

I don’t think it’s killing the desktop client in the general sense. It certainly has dealt serious blows to several desktop clients as they exist today, but that doesn’t mean there’s not a place for an innovative desktop mail client and that’s what the Mozilla Messaging community is working on.

Messaging is evolving really quickly, think Facebook and Twitter, and I think any desktop client of the future needs to do a lot more than the nearly 40 year email technology.

What do you think about browser-based operating systems?

For many of us, it’s already a reality. I spend 90% of my computer time in a Web browser. I rarely interact directly with the operating system and desktop applications. So, does removing everything but the browser make sense? For some, maybe. For me, I still do 10% of my computing in programs that haven’t made it to the Web yet or if they have don’t have as rich and powerful an experience there yet.

At Mozilla, we’ve been working on many new Web features that will help to close the gap for that remaining 10% of my usage. We’ve got geolocation support, experimental multi-touch input, a rich file API, Web protocol handlers, native audio and video, and several other capabilities that used to be considered the domain of the operating system. The more of this we add, the more Web developers will be able to build the kinds of experiences we used to get from the OS.

Your main workstation, OS, mobile device?

I have a Lenovo Thinkpad that defaults to Windows 7 but can also boot to Ubuntu Linux. I also have an Apple MacBook Pro that defaults to Snow Leopard and also boots to Windows XP when I need it for testing something. Finally, I have a Nokia N900 where I test Mobile Firefox.

One more question: Foxes or dinosaurs?

Well, Firefox isn’t really a fox. Nor is it precisely a Red Panda. Firefox is a mythical creature we invented to represent our product.

While I love the Firefox name and our Firefox logo, my history with Mozilla goes back to 1998 and I think I will always be partial to the red T-rex.

—
I’d like to thank Asa for his time!

Expecting a good discussion in the comments section