Mimsy Were the Borogoves

Mimsy Were the Technocrats: As long as we keep talking about it, it’s technology.

Open source value shift at OSCON?

Jerry Stratton, July 25, 2013

I’m at OSCON right now in Portland, Oregon. The last time I went to OSCON was when it was in San Diego, probably in 2002.

If, at OSCON 2002, a White House official had said that they were going to take visitor statistics from WhiteHouse.gov and use them to strongarm local legislatures with “localized data… to present the mood of their constituents”, there would have been an audible outcry. Here, there was nothing—the statement by Leigh Heyman (Director of New Media Technologies, Executive Office of the President) wasn’t applauded, it just went completely unnoticed. He was talking about the future of the White House petitioning system, We The People.

We The People was created by the White House “so that government can allow citizens to petition them.”1 Before We The People, he said, writing to the White House was “a bit like writing a letter to Santa Claus.”

He didn’t say this, obviously, but with today it’s a whole lot more like writing a letter to Santa Claus: writing a non-functioning third party to get free stuff from functioning second parties.

He had some interesting petition signing stats. There are currently (as of the time he made the slide, presumably) 9,660,791 members; he called that individual users. Ten million people is approximately 3% of the United States population. If you’re not a stats person, that might seem small. To me, it seems very large. I suspect that there are a lot of duplicate signups in the system—people who sign up under multiple email addresses to skew the petitions.

It’s likely to get worse. There’s a new API in the works to make We The People work more on the community organizing model: instead of having a place to go for votes (which will still exist to provide an illusion of voting), the new API will allow advocacy groups to bundle signatures into the system.

If, at the 2002 OSCON a speaker had said, as Jared Smith (Director of Open Source Outreach at Bluehost) did, that government is “too big to be efficient”, it would have fit right in: the government model of top-down one-size-fits-all programs is antithetical to open source’s focus on the end user doing the work themselves to create what they need and want.

The second half of that statement would have been more divisive, rather than ignored, as it was here: “it also doesn’t have the resources to do what we want it to do”. This was his paradox of government: it’s inefficient and doesn’t have enough resources. But it’s not a paradox. Any inefficient-by-nature organization will never have enough resources. That’s one of the tenets of open source, you don’t fix monolithic systems by feeding them more resources. They will always eat those resources and still be inefficient monoliths.

The way to fix that is to break down the monoliths—reduce the reach of government. The solution is not to, as Tim O’Reilly said in the next section “to change the way we feel about government”.

This, he said, was the goal of Code for America, who among other things organizes a collection of “brigades“ around the country writing software to help governments. Not necessarily a bad idea, but the terminology is eerie.

Laura Weidman Powers of CODE2040 talked about convincing people to take up coding as a career path. She said that the core value of the open source community is universal access, which she understood to mean that open source is universally accessed. That’s not quite true, in fact it’s kind of the opposite of what open source is all about.

Open source is about DIY: if you want to make it, you should be able to make it: the tools are there and available to you. If you don’t want to make it, well, there are a lot of people in the world and likely someone else has already created what you want, and that’s available to you as well.

If you don’t want to make it, and no one else wants to make it, then everyone has decided it isn’t important. It’s about freedom, the freedom to create as well as to access, not a top-down you must access and you must create.

She added that “No one likes to be told that they have a problem.” She said this to an audience of people who love to be told that they have a problem—as long as it’s true. She was speaking to an entire audience of people who hear “problem” and think “opportunity”.

What they don’t like is being told that someone else’s problem is their fault, and being told what they should do to fix it.

For example, she runs programs that try to bring people into coding who are not interested in coding. That’s great, because the state of public education today is so bad that people who might love coding aren’t going to find out that they love coding because they’re never exposed to it.

However, even after being taught to code, some of her students still had problems getting jobs. Because no one told them that having side projects was necessary to get coding jobs.

This is one of those “wet streets cause rain” logics loved by government programs. Having side projects isn’t necessary to get a job coding. Having side projects is something that people who code well want to do. If you enjoy golfing you don’t need to be told to golf on the weekends, you’ll just do it. And you’ll get better at golfing as the result of doing something you want to do. If you don’t enjoy coding, it’s going to be a very frustrating career path regardless of your side projects—which will also be frustrating.

This does not mean that basic education can’t be improved. Back in the early days of personal computers, high schools thought they needed to teach programming, and it’s not a bad idea. But high schools are run by that inefficient government, and so they don’t often have useful teaching paradigms. I was talking to my girlfriend about this a few days ago: she was “taught” to program by the teacher giving them computer code to type in—and that was it. They were never taught that they could change the code to customize the results.

This is like teaching someone history by giving students the answers to a test and asking them to simply copy the answers to the test. Without teaching students where they can research the answers, how they can logically extrapolate answers, and why answers are not the right paradigm to begin with: questions and problems are.

Which is, in fact, how many high schools (and increasingly colleges) teach history and everything else: memorize facts (answers) with the goal being nothing more than remembering them long enough to place them in the correct location on a test. That method won’t teach a love of history, nor will it teach a love of coding.

But once you know that you can change the code to create new results, if you don’t find it interesting enough to want to change it, forcing yourself to have a side project you don’t care about is not going to work well for you.

Finally, there appears to be a campaign to denigrate copyleft. Copyleft is a form of open source license makes source code available so that other people can modify it and make the resulting software available again—but if they do make their software available, they also need to release the changes they made to the source code. The idea is to ensure that code improvements are shared, and even the improvements can be improved upon.

The other style of open source licensing is the “permissive” license, which does not require making changes available. People can take the source code, build software on it, and then keep their changes secret.

Both license types are used by major open source projects. Apache, for example, the major web server today, uses a permissive license; Linux (on which Android is based) uses a copyleft license.

Yesterday, someone spoke on the benefits of cathedrals inside the bazaar. At least two speakers this morning pushed permissive licenses such as the MIT license over copyleft licenses such as the GPL.

Tom Werner of GitHub simply chose MIT as the best license for freedom; his argument against the GPL was that it’s bigger than the MIT license.2

Eileen Evans from Hewlett Packward spoke specifically in favor of permissive licenses; earlier, she says, she believed that “copyleft was necessary for a vibrant open source community.” But it’s more complicated now. How complicated? As complicated as it needs to be, I think. I am currently reading Thomas Sowell’s Vision of the Anointed and she sounded a lot like the person who argues their case up to the level of complexity that proves their point, and no further.

Which isn’t to say that she’s wrong, but it seemed an odd argument to make: that you guys out in the audience have already decided it’s okay to use permissive licenses, and so therefor you should find it okay to use permissive licenses. If her statistics were correct, there was no need for what she said. It was an odd juxtaposition of talks to have so much in favor of government control and permissive licenses in one morning’s presentation.

In response to Portland Open Source Convention 2013: O’Reilly’s conferences rarely fail to disappoint. We’ll see how OSCON 2013 goes!

  1. Yes, the right of the people to petition their government has degraded to the government allowing their people to vote in online polls.

  2. I’m not averse to this as being one argument—I’ve been waiting futilely for the shorter Free Simpler Documentation License for too long to count now—but it shouldn’t be the major reason, presented just for a laugh.