Off the Beat: Bruce Byfield's Blog

The headlines about Wikileaks and its founder Julian Assange are bringing countless issues to the attention of the general public. For instance, what privacy, if any, is possible with modern technology? How does technology change the relationship between those who govern and the governed? And -- on a more personal level -- does being at best a selfish lover and at worst a possible rapist invalidate a man's leadership or the ideals he claims to espouse? But what occupies most of my thoughts is the basic question about the desirability of openness.

In theory, I am all for openness and transparent decision-making. I have a lifelong dislike of hierarchy to begin with, and I could not have knocked about the free software community for twelve years without having thoroughly absorbed those values. Anyone who didn't believe in openness and transparency would have fled screaming after a month.

Consequently, when I hear diplomats bemoaning the gossip that forms the bulk of the released cables and insisting that their work requires secrecy, my response is skeptical.

"Yeah, right," I mutter smugly to the page or the screen. "Better get used to the new world and stop complaining, because there's no going back."

The only trouble is, once the issue moves from the abstract to a concrete example, I am not so sure. What if an organization and people you respected refrained from publicizing an issue, citing the public good, then announced what they had done after achieving a satisfactory conclusion for everybody? What would I think then?

The relicensing of OpenGL

Specifically, I am thinking of a situation that happened a little over two years ago. OpenGL, the specification used for 3-D acceleration in free software, was discovered to be using non-free licenses. The matter was brought to the attention of the Free Software Foundation (FSF) in January 2008, and the FSF began negotiating with SGI, the creators of OpenGL, to change the licenses. Although the situation was an open secret in some circles, the FSF did not announce it until after the licenses were rewritten to make them free.

At the time, Peter Brown, the FSF's executive director, told me, "We didn't want to draw attention to it because we didn't want to pressurize SGI. We wanted to get their cooperation." The alternative, he said, would have been "strains in community relations," and "a huge amount of time and energy and resources that we would have to pump in to replace this code. We started figuring the funding needed to do it, and it was scary, frankly. That is why it was so worthwhile spending all this time speaking to SGI, because the alternative was really frightening."

Nor was this a needless concern. Brett Smith, the FSF compliance engineer told me that news of the situation was getting out, and people were wondering what to do.

"Someone came to me on IRC and asked if people should start sending angry faxes to SGI, telling them to please clean up their licenses," Smith said. "And I was like, 'No, that's not the right message right now.' We were trying to avoid that kind of reaction, because among the people in the GNewSense community, there was a visceral reaction initially, and it took some time for people to realize that we needed to give [SGI] a chance."

However, instead of such confrontation, the atmosphere of the discussions between SGI and the FSF "was friendly throughout the process," according to Brown. "It was friendly throughout the process," Brown stresses. "It was a case of education, then of working together to find an optimal solution" -- which was then implemented and only announced after the fact.

Openness vs. secrecy

At the time, the idealist in me was disturbed by this course of action -- enough so that I remembered the situation when I started thinking of the issues involved. I have never questioned that the FSF was acting in the best interest of its supporters, and doing what it thought best. After all, the FSF is hardly a corporate or government elite. It is a non-profit organization whose employees could probably find positions of comparable responsibility in the corporate world that paid at least half as much again. Much of what they get out of their actions is some brief recognition in the community and personal satisfaction.

Moreover, choosing secrecy worked. And, to the FSF's credit, it announced what was going on as soon as the problem was solved, which suggests that it always viewed secrecy as a temporary measure.

All the same, part of me continues to wonder if a more open approach wouldn't have been more in keeping with the spirit of openness and transparency. For instance, perhaps SGI could have issued a temporary license, or a public declaration that it would not enforce its non-free licenses?

Possibly, though, SGI would not have wanted to spend more time on what for it was a relatively minor matter. Perhaps, too, hostility from the community might have annoyed SGI so much that its executives became uncooperative, and a satisfactory ending became impossible. At the very least, finding a solution might have been more difficult. So perhaps secrecy may have been the most suitable approach after all.

Living with the ambiguity

I do not think that this example is unique, or that the FSF is the only organization that sometimes has to navigate between openness and secrecy. Very likely, there are dozens, if not hundreds of comparable dilemmas each day around the world. However, I have given the example because it illustrates, more clearly than a any hypothetical sitaution, that choosing between openness and secrecy is not always as easy in practice as it may appear in theory.

To this day, I am still not sure whether the FSF acted properly -- and I am extremely glad that I did not have to make the decision myself. Perhaps I simply have to accept that moments of similar ambiguity appear from time to time, in which no position has a monopoly on truth. Or perhaps I should simply decide from expediency, accepting what happened because the resolution was satisfactory? I am comfortable with neither alternative, since both seem a slippery slope that could quickly lead away from ethical behavior.

My only conclusion is that the choice between openness and secrecy is not as easy as I -- and many other people -- often assume. Against all expectations, those who argue that secrecy is sometimes needed may have a valid point after all. If so, then Wikileaks may in the end be less important for the information it reveals than in the way that the controversy focuses our attention on this basic issue of the digital age.

Comments