Earlier this week Bob Sutton came to talk with Mozilla Corporation employees. Bob is a professor in the School of Engineering and in Organizational Behavior at Stanford, part of the founding team of the Stanford Design School, and the author of a number of books about organizations and their behavior.
Archive for June, 2006
June 29th, 2006
June 26th, 2006
Here is the first half of Mike Shaver’s marvelous summary of the discussions relating to Mozilla project goals held at the Mozilla Corporation in the spring of 2006. (More info about these discussions can be found in my summary discussionof this topic.) Mike outdid himself in capturing the themes of these discussions, and deserves greats credit for taking this on. I’m of course responsible for any problems 🙂
1. Our Work and Legacy Transcends the Browser
This was a nearly universal sentiment (possibly, in fact, universal, but my notes don’t permit me to be certain): what we are doing with the browser is vitally important to the future of the web, but what we are doing as a project and the technology that this project is building has ramifications that go well beyond that. Exactly where the project and corporation should invest to extend that influence is not a well-agreed point (see below), but that the browser is not the whole of what we do was not controversial.
Of the areas in which we hope to have a lasting effect on the industry and the world, some representative sentiments were:
- “Kept the web open, making the Next Thing possible”
- “Showed that open source can product great products, not just great technology”
- “Built and sustained a culture of software development focused relentlessly on real people”
- “Showed that open source and business can be mixed to the ‘benefit’ of a public-good mission”
- “Grew, nurtured, and deserved a strong and energetic community of supporters and contributors.”
The openness of our communication and, of course, software licensing were seen by many as critical to our ability to achieve a lasting legacy. That the artifacts we produce — applications, platform software, records of decisions, organizational processes, business models — will remain for others to build upon even if our project or Corporation should cease to exist is itself a key part of what we do and how we do it, and not just a side benefit or tactical decision.
(An interesting intellectual excursion briefly explored if it was even *possible* for the Mozilla Project to cease to exist, while people continued to use our products and technology and such. Ended in a draw, I believe.)
2. The Need to Appropriately Balance Focus and Diversification
Unsurprisingly, a frequent topic of discussion was how we should balance efforts directed towards leveraging Firefox’s current opportunities and success with investment in other applications (such as Thunderbird, Minimo, or Sunbird/Lightning), or “generalized” technology (such as XULRunner or embedding support). While there is not widespread agreement about the relative value of different tradeoffs, it seems that we largely agree that the managing of such tradeoffs is an important factor in the direction of the project and Corporation both. Also, it was frequently expressed that guiding people in these sorts of tradeoffs should be an important goal for the mission
Discussion and positions on the issue itself were predictably wide-ranging:
- How many different things can we (project, or Corp) manage successfully?
- We should be ready to undertake things that are not as clear to us as “the browser” is today.
- Related: if we were to start today without the history of Netscape’s source release, would we still be building a browser?
- Related: is it easier to “convert” success with Firefox into opportunities in other areas, or vice versa?
- Do we want to encourage more services to be brought into “the content area” via web services, discourage that trend, or remain neutral?
3. “Choice and Innovation” Is At Least As Hard As It Sounds
Whenever the conversation turned to our motto of “choice and innovation,” there seemed to be agreement that it was not concrete enough to be a useful guide in many areas. In particular, the opportunity cost element of choice, as reflected in the previous section, was felt to be especially troublesome. But for all that “C&I” is in sufficient to guide the bulk of our work, most felt that it was a good starting point for determining what the project could and should undertake.
As noted in the overview of the “legacy” threads, the innovation element here was widely and strongly believed to extend beyond “mere” product and technology choices. The organizational, fiscal, and social innovations that the project produces are clearly an important element of our work as a project and as a Corporation, and areas in which people wish to see us continue to invest effort.
Another common thread was that of “indirect innovation.” Whether through packaging our technology for others to use, improving the viability of standards on the web, or demonstrating a “third way” open source/business model, many of us believe that work to help *others* innovate was as important as innovation that we perform directly. (Several people mentioned that the changes in IE7 were an example of this indirect innovation, albeit where we helped create demand for improvment rather than the improvement itself.)
“Choice” is of course no crisper than “innovation” in its impact on specific decisions, but virtually everyone agreed that “choice of browser” is not the only interesting choice for us to work towards. Whether choice of operating system, choice of device, choice of language, or of interaction independent of disability, though, most expressed that the user’s choice was paramount. One person remarked specifically that “putting it in the hands of the user makes [choice and innovation] real”.
Our success in providing the user with meaningful choice is something that’s hard to assess (let alone measure quantitatively), and several people asked what we could use other than browser market share to track our progress there.
Finally, how would a future in which Firefox’s success brought it to 50 or even 90 percent market share affect our pursuit of choice? Would a Firefox hegemony be better than the IE one we’ve been working to break? Would we be as happy with a browser market (for instance; as always, the discussion was frequently about non-browser domains) carved into ten equal slices as into three [or four]?
June 25th, 2006
A while back the Mozilla Corporation held a series of small-group discussions on the topic of the mission of the Mozilla project, and the characteristics of the Mozilla Corporation, eight in total. I decided to start this time with Mozilla Corporation employees, since some of the questions related to why people choose to work for the Mozilla Corporation and also due to some procedural issues that I’ll discuss in a later post. We had a series of eight discussions, including almost all employees. (Shaver and I participated in each group. Otherwise I excluded the management group — schrep, cbeard, John Lilly, Brendan — to help ensure people said whatever was on their mind.) These discussions were intended as a first step and to try out a structure for a broader discussion within the Mozilla community as a whole.
Mike Shaver was the moderator for these discussions, took notes and created a summary document. The language that follows in quotes is Mike’s work, with some minor edits on my part.
The conversations were seeded with four questions:
- What do we hope to accomplish, in the next year and in the longer run? What do we want our “legacy” to be?
- What do we hope our products and technologies do? How do we make choice and innovation concrete?
- What motivates us?
- What excites us about getting up and coming to work each day?
The intent of these conversations was not to establish any wide-ranging consensus, nor to democratically determine the direction of the Corporation (or, of course, the project). Instead, we sought to create a forum in which people could candidly and genuinely express their feelings on these topics, and hear those held by their co-workers, without some of the logistical or communication issues that have been seen in “all-hands” discussions of these subject areas.
The only topic that was declared explicitly as “out of scope” was that of the Foundation/Corporation division of responsibility, simply because that issue alone could easily have consumed all available time and energy.”
Mike identified a set of major themes from these discussions which are below. The overall summary is a bit long, so I’ll break down the contents of the major themes into subsequent posts of a more digestible length. Then we can look to move this discussion to a broader set of people.
- Our Work and Legacy Transcends the Browser
- Need to Appropriately Balance Focus and Diversification (balancing Firefox as our flagship product and our other efforts)
- “Choice and Innovation” Is At Least As Hard As It Sounds (is “choice and innovation” crisp enough to guide our actions)
- Communication and Openness (how to maximize these)
- Why This? Why Here? (why do people choose to work at the Mozilla Foundation / Corporation?
June 21st, 2006
In my last post I identified three major areas of importance to the Mozilla Foundation. These were: project governance, promoting open source and Mozilla software, and development of the Internet as an open, accessible platform. Here I take a stab at translating these broad areas into specific activities that the Foundation should consider. I would say “should undertake” but all things require prioritization and I haven’t tried to rank activities here.
A. The Mozilla Foundation should cause useful open source internet software to be delivered to human beings. The Mozilla Foundation has delegated a piece of this work related to technical and product development to the Mozilla Corporation. So the Foundation should exercise oversight.
June 20th, 2006
What kinds of activities should the Mozilla Foundation undertake? When I think of this I look at three major areas that provide input. The Mozilla Foundation could take on additional tasks as well, but there are three that seem fundamental to me. I’ll describe each area in some detail below, but the summary is:
- Project governance and community dynamics — keeping the project healthy
- Promoting open source software and Mozilla software; and
- Promoting development of the Internet as an innovative, accessible universal platform
A. First, the Mozilla Foundation is the home of the Mozilla project. The Mozilla project existed long before the Foundation. The project was founded in 1998, and developed a community, a governance and implementation model, a set of projects and a great deal of software, expertise and best practices. This was done through informal arrangements based on community norms and participation, without the assistance of a legal home for the project. Those active in project governance had long wanted a legal organization for the project and that goal was achieved in July of 2003 with the creation of the Mozilla Foundation. The Mozilla Foundation enjoyed a high degree of continuity with the organizational structure. Key leaders remained in essentially the same roles. Key processes remained in place, particularly the ways in which developers interact with each other and create software. The Mozilla Foundation became the natural and long-awaited official home of the Mozilla project. Some things fall out pretty clearly from this role as home of the project.
One critical piece of the identity of the Mozilla project is tied to how we build software — an open source, distributed model with delegated authority. So a critical aspect of what the Foundation needs to do is to maintain healthy project dynamics relating to how we build software. Historically this was done by mozilla.org staff; it’s time to integrate the Mozilla Foundation and community-based leadership.
B. Second, because it is a public benefit corporation the Mozilla Foundation has a specific legal reason for existing which is set out in its Articles of Incorporation “The specific purpose of the Corporation [here meaning the Foundation] is to promote the development of, public access to and adoption of the open source Mozilla web browsing and Internet application software.
C. Third, the Mozilla Foundation’s tax-exampt status is governed by the exempt purpose approved by the IRS and the State of California. This purpose in the IRS application is: “The exempt purpose of the Foundation is to serve the general public by undertaking activities to (1) keep the Internet a universal platform that is accessible by anyone from anywhere, using any computer, and (2) promote the continuation of the innovation on the Internet (which as already affected the lives of more than 500 million Internet users). Specifically, the Foundation’s exempt purpose is to develop (a) open source, standards-compliant, free Internet applications that will be usable by (and made available free-of-charge to) tens of millions of users, and (b) foundational technologies that will be used by content developers and software developers to develop standards-compliant online content and open source Internet software.”
In my next post I’ll translate this into a set of more specific activities.
June 13th, 2006
There have been some questions about the RSS icon and trademark protection. The Mozilla Foundation filed a trademark application a while back during the process of evaluating what course of action to take. Same with creating a trademark license to see what it would look like.
After looking at this, we all believe that a community driven approach is best and that trademark licensing don’t make sense for a number of reasons. So several things could happen with the existing trademark application. The Mozilla Foundation could abandon the trademark application. The Mozilla Foundation could commit to managing the trademark only as determined by the community process.
The one thing that I don’t know how to do effectively is to give the mark to some other organization for it to manage. That’s because I don’t know of any generally known and accepted organization for dealing with trademarks. Creative Common is copyright only; the standards bodies aren’t trademark organizations. I’m pretty sure there are organizations that enforce particular marks, but I don’t know of anything analogous to a standards body for enforcing community-governed icons and marks. In the long run having such an organization could be very valuable and I think we’re going to see an increasing need for this. In the meantime, I think the options are as noted above: either the Mozilla Foundation has a trademark and manages it based on community process or we rely entirely on community norms.
My belief is that trying to have the Mozilla Foundation officially responsible for an icon that is already in use is going to cause angst as well as make it harder to have calm discussions about the RSS icon and the proper role for trademark in community-driven activities. This is counterproductive. If abandoning the trademark application (a) makes people feel safer using the mark and/or (b) allows us to have the community-driven trademark discussion in a calmer setting and without suspicion, then withdrawing the trademark application makes sense.
Look for more on the RSS topic from Frank Hecker shortly.
June 11th, 2006
Last week I wrote a bit about leadership and there was a brief discussion of the degree to which a leader helps the community decide and how much a leader actually make decisions.
The coordination function in the Mozilla project is huge. We are a large group of people with many different perspectives. Just developing good communication channels is a challenge. Coordinating the responses, repercussions and interactions of our activities is an even bigger task. This involves constant communication, gathering input, making sure relevant ideas are heard and understood, and juggling some set of activities to get to an answer. At heart, coordination is a process. It’s a critical process, particularly in an open source world where so many people can easily go elsewhere if they don’t like the results.
But coordination is not leadership; leadership is much more. Leadership involves taking all that one knows and setting direction. Open source projects are different from many other organizations in how one achieves a leadership role and how the scope of that role is determined. But the fundamental need for some person or group to make decisions, articulate a direction and lead forward motion remains.
One classic form of leadership in the Mozilla project is the module owner system. A good module owner listens to peers and contributors and in most cases makes decisions that set the direction for the code under his or her stewardship. And yet a module owner’s authority is ultimately determined by the expertise and contributions of the people s/he leads and the degree of confidence of other project leaders. Another classic form of leadership in our world is the person who decides it’s necessary to do something and does it. New projects, new ideas; new technology. Often people lead by doing, and seeing what happens.
Coordination is critical. Good listening is critical. Good leadership is more.
June 9th, 2006
My last post ended with the question: What is the best method of encouraging the use of the RSS icon in Firefox today by many players and still have the icon mean something clear and accurate? There are a few possibilities.
Option 1: The Mozilla Foundation allows the icon to be modified as people want, and to be used on whatever products, services or applications people want to use it on. No one has any particular ability to cause consistency. In other words, we treat the icon like code — use an open source license and turn the icon loose.
This allows maximum flexibility for software vendors and website operators — each can take a recognized image and use it for whatever one want. The flip side is that consumers won’t be able to look at the icon and know what it means. Many consumers are intimidated by the Internet as it is, and struggle to understand the difference between software, search, “the Internet” and data provided by websites. So to my mind, making things clear to the consumer is a high value. The purpose of the icon is to indicate something specific to consumers.
And we know that successful open development efforts have leadership to guide development, not just intellectual property assets floating in the wild. So even this route requires some attention and structure from someone as to how the icon is used, how it develops and how people might want to use the icon.
So to my mind attaching an open source copyright license to the icon and turning it loose seems like a bad plan; one likely to lead to maximum confusion for consumers, and minimum assistance for the groups wanting to create a useful marker for consumers.
Option 2: The other extreme is a classic trademark licensing program. In this setting the Mozilla Foundation licenses the icon as a trademark to everyone on the same terms and has a formal process for managing the evolution and use of the mark. That process might be community focused, and the Mozilla Foundation would be the ultimate decision-maker as the owner of the mark.
The advantage to this system include:
- A level of clarity over the degree to which the icon can be assured to mean something to consumers;
- Having a known “home” for the icon, a known place to bring issues and work out discussions
- Not all organizations show much respect for community norms. Living within a known legal framework provides another tool for responding to bad actors.
- Use of trademark law in open and community processes is new and sometimes not well liked.
- I believe the Free and Open Source Software world is due for a long discussion of trademarks, how we use them, what their value is and so on. Ultimately I’d like to see some Creative Commons type options available for trademark-type purposes. (Creative Commons licenses are all copyright licenses, and do not purport to address the trademark-like issues of providing clarity to consumers about what consumers are getting.) We haven’t had this discussion yet.
- A formal trademark process can be a significant amount of work. The Mozilla Foundation will take that work on for key marks, like its product marks. The Mozilla Foundation should also be willing to take on leadership, organization and work for other marks that are important to the industry, including the RSS icon. However, taking on these activities before we’ve had the trademark discussion could be counter-productive. If we use the RSS icon to start talking about what industry wide icons can and should mean we should come to some shared understanding (or at least shared vocabulary, if not understanding) of the value of an icon and how best to make this visual clues mean something real for consumers.
Option 3: Another option is to try a less formal process with more authority resided in community norms and seeing how that works. To do this the Mozilla Foundation would:
- develop a clear set of community norms and usage guidelines;
- lead a community process for the evolution of the mark and the guidelines; and
- identify a mechanism where companies using the icon (particularly in software products) publicly pledge to the implementation of the guidelines and complying with the results of the community process.
I believe Option 3 is the best approach for the RSS icon at this point in time. People are interested in the icon, people are already using the icon, we can learn how precise a meaning we believe it should have and how community norms function in this setting.
I’m recommending that the Mozilla Foundation adopt Option 3, publish a set of usage guidelines and proposed community norms (potential examples include: public discussion and evaluation before using the icons for new versions of file formats, proposed modifications of the icon, statement of intent that the icon continue to mean something precise to consumers, etc.) and provide a forum for discussion as soon as possible.
June 8th, 2006
Here’s a question with both philosophical underpinnings and a concrete set of product implications. The philosophical side of this question includes questions such as:
- how does a consumer know what web services he or she is requesting?
- How important is consumer clarity?
- How do we encourage clarity across a networked base of products?
- If consumer clarity is important, how do we get this?
- Is it possible to provide consumers this level of clarity without some organized control mechanism?
- Can these issues be managed through a community-based process? Or is a more formal legal structure useful?
- Is community leadership enough, or is an ultimate decision-maker necessary?
The concrete example here is the RSS icon that’s been in Firefox for some time now. Mozilla Firefox includes an icon that represents the availability of an XML based RSS feed. When you click on the icon in the location bar of Firefox you are able to add an RSS feed just as one adds a bookmark. When you click on a bookmark identified with the RSS icon (a “live bookmark” in our parlance) you see the recent entries from that RSS feed. A number of web sites use the same icon to help consumers recognize the presence of an RSS feed that can be added to Firefox or an alternative piece of software that understands RSS feeds (such as a “feed reader”).
A while back Microsoft approached us about using the Mozilla RSS icon in the upcoming version of its browser. We thought that having multiple software products use the same icon to represent the presence of an RSS feed would be helpful to consumers. It would provide a standard visual clue for consumers and provide clarity. Of course, this only works if the icon actually represents something reasonably crisply and accurately. If a single icon comes to be used for many different things then there’s not much benefit to consumers, and might even be some disadvantages. (For example, it would not help consumers to click on the so-called RSS icon and end up downloading some wildly different format.)
So what is the best method of promoting the use of this icon by many players and still have the icon mean something clear and accurate to consumers? In an effort to keep my posts to a digestible length I’ll describe the methods we’ve thought though in the next post.
June 6th, 2006
Periodically I hear people say things like “the community will decide.” Or “let the community handle it.” This has always sounded amorphous to me and I’ve finally realized why. This way of looking at things assumes there is little clear leadership and no actual decision-maker. This is a misunderstanding of many open source projects, which have real leadership and ultimate decision makers. For example, Linus Torvalds makes decisions for the kernel, Apache has a committee system, Brendan is responsible for technical decisions in the Mozilla project. In the Mozilla project we try to delegate authority so that:
- many decision-makers are involved in their areas of expertise;
- an ultimate decision-maker exists but is involved only when necessary; and
- decision-makers are chosen and evaluated by their ability to provide leadership that draws in other contributors.
This last point is extremely important. Decision-makers who don’t provide leadership make it hard to build a good project. Decision-makers who people don’t want to follow will struggle. Open source software always gives people a choice. Participants are always free to try something different using the same underlying code. So the decision-maker’s role has many facets. Basically it boils down to making enough decisions correctly enough that people continue to choose to participate. In other words, leadership.
Other projects may invoke the authority of the ultimate decision maker more or less often. The important point is that many open source projects incorporate clear decision-making into their processes.