Issue #5 March 2005

Tiemann named president of OSI

This month we talk with Michael Tiemann, Vice President of Open Source Affairs at Red Hat. Michael has recently been appointed as the new president of the Open Source Institute (OSI). We asked Michael to tell us a little bit about his new role.

Red Hat Magazine: How long have you been a member of OSI?

Michael Tiemann: I was actually at the seminal meeting at Christine Peterson's (Founder of the Foresight Institute) house where Eric Raymond, Bruce Perens, and others brainstormed the term "open source." But I did not join the OSI until January 2001.

RHM: How did you become a member and why?

MT: I have been on a number of small boards, all of which were companies or causes I believed in. The first such board was Cygnus, which comprised three very charismatic hacker-types interested in proving (and profiting from) the fact that there was more value in delivering services around free software than in trying to build a better bunch of bits. I think that was a success. Another such board was IUMA, the Internet Underground Music Archive. IUMA was a groundbreaking best-of-the-web site back in the day—attempting to be the Apple Music Store for unsigned, independent bands. That was less of a success, in part because the music industry is so cutthroat that they'd rather cut their own throats than to let a new voice be heard.

When Eric Raymond approached me to join the OSI board, I saw a great opportunity in that I think the OSI's mission is vitally important as well as a challenge. I had a lot of experience in working with individuals who lead the vanguard, founder-board-executive issues, etc. I had no illusions that being on the OSI board would be an easy job, but then again, anything meaningful takes effort.

I joined expecting that I could help with fund raising, founder transition issues, and helping the OSI to mature into the organization the world needs it to be.

RHM: What duties are performed by the president of OSI?

MT: Eric Raymond was the first president of the OSI, and his style was very effective at helping the OSI be heard. As President Emeritus, Eric will still be a major voice both inside the OSI and to our constituents.

The role, as I see it, for me as president is to make sure the OSI remains focused on and completes its highest priority tasks:

First, expanding the board to be more representative of the global open source community. Today, the board is populated 100% by Americans living in the US. My hope is that when the next election is held for president at least 35% of the votes come from members living or working outside the US and that over time we have even greater diversity of participation.

Second, continuing to be a strong and proactive steward of the Open Source Definition. The ACLU is an organization that sees themselves as advocates of the freedoms that the US Constitution grants to its citizens. Sometimes these freedoms lead to uncomfortable situations, but the ACLU takes the position that the principle of freedom trumps the convenience of comfort. The Open Source Definition defines the guiding principles for open source license writers. Sometimes these writers create licenses that the community won't accept, but is that a problem with the principles or a consequence of freedom?

And then a third task that I take on along with my fellow board members: to be a positive spokesperson for the organization, the community, and to speak truth to power.

Of course, the president is also responsible for managing the staff of the OSI, which has grown recently with the addition of a Director of Legal Affairs and a Director of Public Relations.

RHM: The number of OSI licenses seems to be growing to an unwieldy number. Do the folks at OSI agree or disagree with this assessment?

MT: Well, from my perspective this question is the question that the OSI and the community should be asking. And yes, I do have some opinions about where this question may lead.

Let me take a step back and start with the license that put me on my path to success: the GNU GPL. For me, the GNU C compiler was the code I most wanted to hack, and the GPL was the license that gave me the strongest promise that whatever else happened in the future, nobody could deny me the fruits of my labor. With good code and a good license, there was nothing stopping me from working 16-18 hours a day, 6-7 days a week. That the GPL offered me the fruits of others who valued a similar promise (and who often wrote better code than I could) was all upside. The GPL has another great property: its share-and-share alike nature has protected GPL-covered code from the kind of forking that fragmented Unix. I'm not saying that all forking is bad—it's necessary for innovation—but when forks cannot be reintegrated by willing developers, that's bad.

There was no license proliferation under the GPL—one either used it or one did not. When the LGPL was created, there was a license choice, but no license incompatibility. Great work if you can get it, but in 1998, scarce few companies were willing to use GPL code, let alone author code licensed under the GPL.

The OSI took the approach that instead of getting people to adopt principles encoded in a license (which is precisely what the GPL is—it even says so in the preamble, the OSI would author a set of principles that people could agree to independent of the license. This is a bit like the difference that standards bodies make between a specification and an implementation. Sure, one could take the approach that "TCP/IP is whatever Berkeley says it is," but the approach the market and the industry favored was to write specifications of TCP/IP (the principles) and encourage competition at the implementation level. So what if it took a few Connectathons before real interoperability began to emerge

When I look at Fedora Core 3 and see that over 70% of the code is covered by the GNU GPL, it tells me that today, the market favors the GPL as the best open source license. But I could not imagine using Fedora Core 3 without software covered by non-GPL licenses. There's value in competition, even at the licensing level.

I have no doubt that the discussions about license proliferation are going to force the OSI and the communities it serves—hackers, the private and public sectors, educational institutions, non-governmental organizations, etc.—to take a hard look at the principles and the consequences and decide where we should take open source for the next 25 years. I believe that in the end we will find that license proliferation is a symptom, not a problem, and that the symptom lies not with the OSD but those who try to bend the OSD to their own private agendas.

About the author

Michael Tiemann founded the world's first company exclusively devoted to providing commercial support for free (and later open source) software. He is Vice President of Open Source Affairs at Red Hat, President of the Open Source Initiative, and provides financial and other support to organizations that promote software, civil, and artistic freedom, including the Free Software Foundation, the Electronic Frontier Foundation, and the Creative Commons.