LJ Archive


What's the Kernel Space of Democracy?

Issue #264, April 2016

No one pretends that democracy is perfect or all-wise. Indeed, it has been said that democracy is the worst form of government except all those other forms that have been tried from time to time.—Winston Churchill

Might the same be said of operating systems and Linux? In both cases people live atop, and depend on, a deeper protected and enabling structure. In democracies, it's government. In Linux, it's the kernel. There are resemblances. For example, both need debugging, no matter how much bigger and better they get. A difference (one of too many, I'll admit) is that Linux gets debugged by contributors and maintainers, while in democracies, users are free to mess with the government, and vice versa, by many means.

Still, although government space bears little resemblance to kernel space, the question haunts me, because I can't help thinking democracies might learn something from Linux.

It can't be easy. Politics never is. Although government runs on the form of code we call law, those that operate on it are not compelled to obey, or even to agree. Craig Butron highlights the difference when he says “all technical problems are technical and political—and you can always solve the technical ones.”

The contradictions of democracy are old news. Otto Von Bismarck said “Politics is the art of the possible—the art of the next best.” Yet he also said, “Not by speeches and votes of the majority, are the great questions of the time decided...but by iron and blood.” While democracy might be as old as Greece, war is as old as our species, and perhaps even its evolutionary antecedents.

“Compared to war, all other forms of human endeavor shrink to insignificance”, said George S. Patton. Among those forms of human endeavor are democracy and governance. Even the rule of law, which all democracies strive to maintain, is suspended gladly when battle flags are raised. In War Is a Force That Gives Us Meaning, Christopher Hedges writes:

War makes the world understandable, a black and white tableau of them and us. It suspends thought, especially self-critical thought. All bow before the supreme effort. We are one. Most of us willingly accept war as long as we can fold it into a belief system that paints the ensuing suffering as necessary for a higher good, for human beings seek not only happiness but also meaning. And tragically war is sometimes the most powerful way in human society to achieve meaning.

Of course, meaning to one faction is not the same as to another.

Many years ago, on a Linux Journal Geek Cruise (man, those were fun), I gave a talk about political differences—for example, how one side tends to see the market as a problem and government as a solution, while the other side tends to see government as a problem and the market as a solution. My main source for that talk was George Lakoff's 1996 book, Moral Politics: What Conservatives Know that Liberals Don't (re-subtitled in 2002 as How Liberals and Conservatives Think). Even if you disagree with George's conclusions (or his politics), he does an outstanding job of laying out how liberals and conservatives talk past each other while constantly characterizing each other (sometimes correctly), and how hard it is for both sides to see truth in the others' framings. (Since the turn of the Millennium, George has written many other books, mostly to give Democrats new ways to frame debates. I believe Barack Obama would not have been elected, or re-elected, to the Presidency in 2008 and 2012, without those books. They mattered enormously to Obama's campaigns.)

After my talk, Andrew Morton said, “That's why the left thinks the right is evil, and the right thinks the left is stupid.” I still can't think of a better summary statement.

I was given a similar hunk of wisdom by David Hodskins, my business partner for two decades and one of the smartest businessmen I have ever known: “Republicans are the party of wealth creation and Democrats are the party of wealth redistribution.”

Consistent with that, I know Republicans who think all the good in the world is produced by business, while I know Democrats who think nature put a sum of money in the world and that it's the job of government to make sure those who have too much yield some of it to those who have too little. At one extreme, we have people who don't see how government produces any good (besides, as Ronald Reagan put it, “defending the borders”), and at the other extreme, we have people who don't respect much of what business does, besides providing jobs.

Around both poles gathers homophily: “the tendency of individuals to associate and bond with similar others”. Those associations and bonds also comprise echo chambers within which even friendly and helpful voices tend to go unheard or misunderstood.

For example, a few years back, I suggested at a university meeting that we ought to define what the Internet actually is, because there is no common definitional ground between net-heads and bell-heads (those who see the Internet as a transcendent entity and those who see it as just a service offered by phone and cable companies). I said the university would do the world a service by coming up with a canonical definition. A professor replied, “I didn't think we were going to work on policy this year.” I replied, “Whoa! This isn't about policy. It's about language. Linguistics. Dictionary stuff.” But the discussion drifted toward policy anyway, and the desire by nearly all present for net neutrality regulation, which made sense. After all, this was an echo chamber for netheads, including me (example: Cluetrain's New Clues). But I'm also a business guy, and at the time had at least some hope that a university could think in more, um, universal ways. But universities tend to be castles, and this was an example of the one on the left in Figure 1.

Figure 1. Two Castles

On the right were phone and cable companies who talk about “saving the marketplace” and “keeping government out of business”, even though they are regulatory zoo animals holding their keepers so captive that they have successfully lobbied legislation restricting business to themselves in all but a handful of states (www.muninetworks.org/communitymap) and wouldn't know what to do in a truly open marketplace.

Speaking of netheads and bellheads, it's interesting to see how net neutrality has gone since I wrote about it in the July 2006 issue of Linux Journal (www.linuxjournal.com/article/8979). In it, I sourced Larry Lessig, Vint Cerf and Michael Powell, who was then a recently retired FCC chairman and is now cable's top lobbyist. Here are a few one-liners from a talk he gave at Freedom to Connect that year:

...be careful of inviting the legislative process when they have a very bad understanding of the technical underpinnings. Because the secondary consequences of their errors can be enormous.

I would rather try constantly to position my industry where I succeed if government does nothing, versus positioning it in a way where I need them to do something or I'm dead.

I can tell you right now, very few people in Washington (understand net neutrality). This means you're going to get a potentially very ambiguous, subject to massive variations in interpretation, pile of law.

...government has a way of turning on people. Ask Bill Gates. It may be about networks today, but those same principles can be used against innovative business models and applications in other contexts. And, I submit to you they would be.

Be careful because you're playing their game....The average one of these incumbents, whether a cable or a phone company, have 40 lawyers in Washington dedicated to this work. Resources. Ability. One hundred years of skill....Then, let me add the judicial process. Every decision you get from the Congress and the FCC will spend the next three and a half to four years in court.

Since then, no net neutrality legislation proposed in Congress has moved past the committee stage. But on the regulatory side, the FCC in 2010 issued the pro-neutrality Open Internet Order. In September 2011, the FCC added a final rule titled Preserving the Open Internet. Naturally, Verizon sued, eventually winning Verizon v. FCC in the DC Circuit Court of Appeals, which vacated two of the three main portions of the FCC's Order. That happened in May 2014—right in line with Powell's estimate for time in court.

But the battle is hardly over. In February 2015, the FCC issued “strong, sustainable rules to protect the open Internet” by re-classifying “broadband Internet access service” under Title II of the Telecommunications Act. Like everything Federal, it's complicated, which means it met mostly with approval by netheads and disapproval with the bellheads. Expect some of this to be worked out (and not) over the next few years.

Meanwhile, Linux won without going to war, even though it had lots of opposition from the start. Microsoft hated it, as did competing UNIX and UNIX-ish operating system camps.

I showed up when Linux itself was to the right of the decimal point, in the early 1990s. I was invited by Phil Hughes, who included me in an e-mail conversation about starting a free software magazine. This became Linux Journal in April 1994, when Linux arrived at v.1.0. (Little known item: the first editor was Bob Young, who left to start Red Hat.) In those days, and to a large degree ever since, the Linux homophily has been largely a libertarian one. Perhaps this is a manifestation of the distinction between kernel space and userspace. The perspective of the kernel on userspace is like that of the Earth's core toward what happens on its surface. As Linus often says about userspace, “I don't care.”

Yet all of us live in userspace, and we have politics here, especially between now and November, when the US holds its presidential election. While that's happening, I suggest we technical folks look at the process as one that requires debugging. Here are my suggested few:

  1. Get money as far as possible out of politics. Larry Lessig has required reading on this (republic.lessig.org).

  2. Amend the Constitution to eliminate the Electoral College and allow direct election of the President.

  3. Eliminate gerrymandering, as far as possible.

  4. Expand voting rights to all citizens.

All those are simply for restoring democracy and nothing more. Like many in our community, I not only want better government, but less of it, especially as we work out what it means to be digital as well as analog creatures. For particulars on that, check out what I said about Ralph Nader here in 2000 (www.linuxjournal.com/article/4316). A sample:

Consumers and workers are rhetorical relics. The Net is uniting both, and they're throwing off their chains. Industrial communism and capitalism are both terminal. They can't survive in a networked marketplace, where “We the People” means exactly what it says.

We have real challenges in this marketplace, not the least of which is understanding how it's going to work, now that “the People” are working both sides of the supply-and-demand handshake. What are the new social contracts? What laws do we really need, and why are we keeping the ones we don't? How can we truly help those who can't help themselves? What are our agreements about privacy and anonymity, and how do we organize them? How do we build the needed infrastructure around our new Commons, and how do we keep those stuck with dead industrial market models from wasting our time?

Those questions still stand. So does the one in the headline above.

Doc Searls is Senior Editor of Linux Journal. He is also a fellow with the Berkman Center for Internet and Society at Harvard University and the Center for Information Technology and Society at UC Santa Barbara.

LJ Archive