Archive for July, 2007

Some people have wondered if revenue is the reason we’re looking at making some changes in our approach to Thunderbird. The answer is no. No, no, no. The reasons for looking at a change are articulated in the last 5 or 6 posts:

• The impact of browsing and the Web, as delivered through Firefox, dwarfs Thunderbird
• Thunderbird is a different enough product and audience that the focus on browsing and the web doesn’t automatically bring Thunderbird what it needs
• Thunderbird — both its strengths and its weaknesses — are overshadowed by the giant footprint of Firefox.

These issues would concern us whether or not Thunderbird generates revenue. Mozilla is not aimed at maximizing revenue. And Firefox revenue is funding a range of activities beyond Firefox, from infrastructure to Thunderbird employees, to documentation for web services.

Once we figure out what kind of organization makes sense then we can look at what funds it would need. Then we can figure out what combination of money from Firefox and what from Thunderbird activities would make sense. We don’t plan to leave Thunderbird high and dry without funds.

I know from previous comments that some people will never believe this. But repeating myself won’t change that, so I’ll stop here.

In my last post I described the profound difference in impact between the Thunderbird and Firefox projects. This goes beyond the 10 or 20 to 1 difference in size of userbase. It also includes Firefox’s effect on openness and innovation in general. I described how this causes the relative prioritization between Thunderbird and Firefox to be severely skewed towards Firefox and why I believe it will remain that way.

This could perhaps be OK if the two products were very similar, so that work on one was intimately related to the other. We have found this not to be the case.

The two are complementary products for a set of users, but much less so in development. There are a number of reasons. The products are different, the userbase is different, the international aspects are different. Tristan described this nicely already, so I’ll be brief.

The products have large areas that are not as similar as one might think. Thunderbird uses the underlying Mozilla platform of course. So do many other products. But Thunderbird is intimately tied to IMAP and POP, specialized areas fundamentally different from the core of the web. So the technical relationship between Thunderbird and browsing is actually less obvious than the overlap between Firefox and all sorts of other products that are fundamentally about the web — video viewers, web based music programs, to name a few. Testing and QA are also different.

Perhaps even more fundamental, the world is still moving new things into the browser platform, but many consumers are moving away, have already moved away or may never use stand-alone desktop email.

• Web mail usage grows. Younger generations in particular use other techniques.
• There are many parts of the world where email is less common than in the US, Western Europe or Japan. For example, in parts of the world where Internet cafés are a major way of accessing the Internet desktop email is not the norm. There is a serious question of whether these folks will ever move to mail, or if other options, either web and / or mobile based will always supplant email as we know it.
• Thunderbird is much closer to an enterprise product. Development may still focus on what’s useful to an individual. But given the consumer adoption of webmail, enterprises are a significant source of interest in desktop email. Yet there remains debate about the Thunderbird roadmap, which does not include calendar as the key feature. I’d like to see a structure that promotes maximal feedback between the Thunderbird team and the userbase, and believe a focused organization focused is a better place.

This doesn’t make mail unimportant. It does reduce the degree to which the same development organization can excel at both products.

One large them of responses to the Thunderbird post is the question: Why can’t Thunderbird and Firefox both prosper in the same development organization? Since there is money, what’s the problem?

The problem is trying to do two different types of things exceptionally well at the same time. This is extraordinarily difficult. (I’ll describe why we’ve found Thunderbird and Firefox are different enough to make this so in a separate post.) Trying to do two different things requires a constant balancing of the needs of each. In many cases it results in an inability to optimize for either one and both projects suffering. In our case it also results in a constant need to prioritize between the two. And in this prioritization Firefox is getting and will continue to get the vast bulk of resources.

This is because the impact of the two products is wildly different. Thunderbird is a solid product that provides an open source alternative in an important area for a set of users. That’s important and worthy of attention.

Firefox is important in moving an entire segment of the Internet industry towards a more open, more innovative place. We’re not the only factor of course, there are lots of other critical people and organizations involved. But modern, innovative browsing and web development as displayed through Firefox is part of what moves the Web landscape now.

Firefox effects are felt by people who use other products. Its effects are felt by people using other browsers (even IE has a development team again!). Its effects are felt in the standards world, where Firefox’s footprint strengthens our efforts to move web standards forward. Its effects are felt by web developers who have made Firefox their development tool of choice. The effects of Firefox go far beyond the daily user experience of its userbase.

This difference between Thunderbird and Firefox is profound. Each day the development team can work on Thunderbird, which serves its users well. Or they can work on Firefox, which affects a giant swath of the web industry, and serves a userbase that is at least an order of magnitude larger.

In this setting it does not make sense for a development group to give Thunderbird equal focus. Counting just the userbase, Firefox is 10 or 20 times bigger, so maybe one would say it should get 10 or 20 times the attention. If one adds in Firefox’s effect on the web industry the focus on browsing related activities goes way, way up until it dwarfs Thunderbird even more completely. One might then adjust these numbers in favor of Thunderbird because of a desire for diversity, a desire to continue to serve Thunderbird users, because we’ve always had mail, or for any number of other reasons.

But the point is that this is not a good setting for Thunderbird to get sustained attention and focus. Hiring more people doesn’t solve the problem. It doesn’t change the equation for determining relative attention.

This is our setting. This is why I say that I do not see the existing Mozilla development organization increasing its focus on Thunderbird in the forseeable future. Every time we look at it we are convinced that the current prioritization is correct.

We want Thunderbird to thrive as an open alternative for email. Thus the current effort to find a structure where Thunderbird and email can be the focus. We can imagine this happening within the Mozilla umbrella if there are separate development organizations — thus Option 2. We can imagine other possibilities — thus the other options. But the current setting needs change.

Many of the comments to the Thunderbird / mail post include suggestions about features for Thunderbird. These range from topics like adding visualization to adding calendaring functionality.

I encourage people to take these topic to the Thunderbird development and discussion areas. I would say the same thing if the topic was Firefox. Feature and product planning happens within the project group. For example, there are many comments about calendar functionality, and there is an active calendar project underway. The Thunderbird discussion areas are the right place to ask about the roadmap for integrating Lightning into Thunderbird, to note the degree of need and to look for a timeframe. Better yet, that’s the place to get involved in making things happen. Mozilla is successful when we are rooted in active, distributed involvement and contribution.

If it turns out that there is some barrier to getting involved, or if there is some other problem for contributors, then we’re in a different place and I definitely want to know about that.

I suspect most of us agree there is a lot of exciting potential improvements for Thunderbird and for mail in general. The point is how best to get sustained, focused attention and real movement to addressing these.

There has been some interpretation of my Thunderbird /mail post as saying that Thunderbird doesn’t fit into the mission of the Mozilla Foundation. This is not my view at all. If Thunderbird didn’t fit into the mission, then we would be having a very different discussion. For example, Option 2 of my original post — a separate organization inside the Mozilla Foundation — would not be possible if Thunderbird didn’t fit within the mission.

The Mozilla mission is very broad; broad enough to encompass many aspects of human interaction with the Internet. This includes the work we’ve always done and the long-time Mozilla projects and products. It also includes a range of new activities that could be undertaken.

The question is: How does the Mozilla Foundation best serve its mission? Where does it focus? How do we develop maximum participation in our software development and in improving the quality of Internet life?

Until now the guidance of both browsing and desktop email related work has been in a single organization which today is the Mozilla Corporation. Most of those closely involved in the daily development believe this is not the best solution. (The question of why we believe this is another broad theme, which I’ll address in a separate post.)

The “Call to Action” is a public call to help figure out a good solution. It might well be that we end up with Option 2, in which Thunderbird efforts remain completely within the Mozilla world. Currently the two lead developers are currently leaning toward Option 3 which I believe is due to the desire for simplicity in getting started. The other options are on the table precisely because Thunderbird is indeed within the Mozilla mission and none of us wants to see it decline.

The Call to Action is also a request that those people interested in committing time and energy to developing Thunderbird step forward. The Thunderbird community is currently small. We’re trying to find out who is ready and able to contribute. Community vitality is a fundamental part of Mozilla’s mission. So please do speak up if you are interested and able to be a part of Thunderbird development going forward.

Similarly, the Call to Action for a broader mail initiative is to see if there are people interested, competent and ready to get involved in a serious way. This is not walking away from mail, it is an invitation for those interested in such a project to step forward so we can evaluate what is possible.

Mozilla has resources to apply to these efforts, including funds. We are trying to figure out the best way to use resources effectively. This includes Thunderbird, this includes a potential new approach to mail, and it includes new ideas that may emerge over time.