I spent two days in Boston this year, as the FSF had promised a bigger, more organized conference, with more speakers than would fit a one-day meeting like last year's. I'm a lousy trip-planner -- Nina knows it, you all know it. The process makes me anxious, and so I defer it, which means that when I do get down to the business of, say, figuring out where to stay over a weekend, there are fewer options and they're more expensive. So it was that I neglected to discover that good ol' Greg had returned from China and was living back near the vicinity of MIT again until it was too late to ask him to couch-surf, gift of whiskey or no. Instead I booked myself a room at the Hampton Inn and shelled out for Amtrak tickets there and back. I woke up at 5:00 AM in order to make my 7:00 AM train, attempting to reach Penn Station via the F train, then falling back to trusty old Carecibo when the F hadn't come after 25 minutes. To get to the actual Acela platform at Penn Station you have to go down this escalator tube from the waiting area. Even though the first time I went to Boston for the conference I went via Chinatown bus, looking that tube hole always reminds me of that time, when I didn't really know where I was going or what it would be like when I got there. We stopped at Roy Rogers. I think I had french fries for breakfast.
The conference has moved around a bit over the years. The first few years I went to it, it was at MIT, which was neat, 'cuz that was probably the only way I was ever going to see MIT. And then they moved it to Harvard after MIT stopped cutting them a good enough deal on space and catering, and that was neat, too, because I'd been curious about Harvard. Last year they moved to Bunker Hill Community College, which was somewhat less neat, because small and drab and out in the sticks (oh god the Orange line) with only strip mall amenities. This year they moved yet again, to the UMass Boston campus, which was not super convenient to get to (T to a 20 minute ride on a shuttle bus) but which is right on the harbor, and so the high windows in its great big meeting rooms provide a wide-angle view of Dorchester Bay, cold and still, like a gray mirror. It's very New England, you see.
David Sugar was wrapping up his talk on the GNU Telephony project as I arrived. He fielded some questions about SIP support and various types of audio codecs, and then yielded the stage to Michael Flickinger from GNU Savannah. Michael gave a run-down of the services offered by Savannah and how they distinguish it from other established software forges like SourceForge as well as some of the flavor-of-the-month ones like gitorious. In particular, every project submission on Savannah is human-reviewed to ensure license consistency and avoid tricky legal situations further on down the line. (This is also a resource squeeze for them, as the process can be time-consuming.) He also explained some of the current plans for improving -- or rather, rewriting from scratch -- the software that drives Savannah, Savane. The audience was nonplussed, and he seemed nervous, so I raised my hand and gave a little sales pitch for Savannah, which I genuinely love, despite its warts. I don't know if it worked.
We broke for lunch after that. They'd arranged for these little bag lunches to be delivered, and I found an empty table in the adjacent room. Some people travel to these events with friends, but I'm kind of a unicorn among my local peer group when it comes to this particular interest, and so I'm always there alone. Not knowing anybody used to make me really uncomfortable, but over the years I've gotten used to doing this stuff on my own, gotten used to being and feeling weird. So while I was prepared to eat by myself, it was nice to be joined by people who introduced themselves Alison Chaiken and Tom Marble. We talked about data serialization frameworks (something that's been a focus of mine recently) and swapped FSF gossip.
Alison's talk was directly after lunch. She's working on establishing a foothold for Free software in cars, as part of her involvement in the more general "right to repair" movement. She pointed out that since the first software systems embedded in cars was for purposes of "info-tainment" -- DVD players and video games for the back seat -- there's an established tendency towards complacency around the next generation of car "apps," which will likely focus on safety and driver informatics. Some manufacturers have even begun to release SDKs of a sort, such as the Cadillac User Experience framework, which is built on top of X11. But without unrestricted access to source code, users will just have to trust manufacturers and their partners to deliver secure, bug-free software, which is by no means a safe bet: She referenced a study done by researchers at UCSD and the University of Washington that produced an exploit capable of disabling the brakes and steering console of a moving car by hijacking a wireless tire pressure sensor. Wowza! And then there is also, of course, the perennial motivation of being able to inspect and modify the operations of a device that you've, you know, bought.
Brett Smith gave an update on the work he's been doing in the licensing lab. He works on two fronts towards a single goal: The lab strives to help Free software developers do their work without excessive interference from the law; and it helps legislators understand what Free software developers do so that they don't interfere. Toward the first half of that agenda, he presented some new licensing resources for developers. The FSF's guide How to choose a Free software license for your own work aims to provide a criteria-based approach to license selection. There's a new paper about what he referred to as "Javascript labels," a technique for providing formal descriptions of license characteristics for Javascript source files, reflecting the current trend toward client-side processing for interactions with web applications. (He noted that the current proposal was unlikely to "take over the world," but said the FSF was very interested in feedback.) And the new version of the Mozilla Public License, MPL 2.0, features compatibility by default with the GPL; the MPL 2.0 process, in fact, was inspired by the highly interactive one that produced GPLv3.
"So that's our friends," he said. "Now let's talk about our enemies." The FSF has apparently been sending him around the world to participate in panel discussions for international trade agreements and bits of legislation. In the U.S., the Library of Congress is about to go through an every-three-years mandated review of the "chilling effects" of the Digital Millenium Copyright Act. The last time around, they resolved that "jailbreaking" devices like cell phones ought to be permitted. Brett said that this time the Software Freedom Law Center is going to be pushing for a provision to allow people to install their own software on any computing device that they own. And he's going to sit in on the negotations for the Trans-Pacific Partnership, actually at the request of the treaty authors, who are hoping to avoid a post-ratification debacle of the scale of the one resulting from ACTA, which left even some of the representatives who voted for it wishing that they hadn't.
ginger coons (she spells it lower-case) is the E.I.C. of LibreGraphics Magazine, a paper-and-ink publication whose aim is to promote the use of Free graphics tools, with a focus on creators and how they use software tools. She repeated a question that she's often asked about Free graphics software: It's good, but is it print good? As proof she explained how the sausage gets made. The whole magazine is produced in Scribus, Inkscape, and GIMP, among other applications, and using Free fonts (Univers Else, Prop Courier Sans). They print on recycled paper with soy inks, and license the whole thing CC-BY-SA. She had several issues for demonstration and sale. They looked good! The print design straddles the line between trade publication and art magazine, kind of the same way that SEED, the magazine Nina used to work for, did. I wanted to buy her a copy of issue 1.3, which featured a striking pattern of opening and closing eyes -- ginger claimed they'd figured out a way to randomize the colors on a per-physical issue basis -- but it sold out quickly.
The "keynote" of Day 1 was Evan Prodromou's presentation on freedom for the social web. Since the launch of StatusNet a few years ago and with his commitment to engaging with its users for the purposes of technical support and philosophical debate, I feel like he's developed a pretty high profile in the Free world. So his talk was a celebrity appearance of sorts. It didn't hurt that he's a very engaging speaker, equal parts nerdy intensity and practiced charm. He opened with some historical perspective on social networking, detailing how providers have evolved from an application-per-media-type model to social platforms like Facebook -- a model he refers to as the "imperial network." These platforms have the primary benefit of mapping independent streams of shared social data onto the same unified social graph. This has provides the benefit of one-stop shopping for marketers, but doesn't do much for users -- in addition to the fact that it's, you know, evil and non-Free, it doesn't accomodate users' disparate interests very well, unless those interests are confined to finding one's friends on Facebook.
Evan proposes / predicts what he calls the federated social network, which is really a network of social networks with interoperability at the borders. He cited some examples of work in this direction: OStatus; activitystrea.ms; pubsubhubbub; Webfinger; salmon. The question always gets asked: Why aren't we there yet? Where's my Free Facebook? These things take time, he said, and people often focus on distractions like novel architectures and cryptotopian fantasy.
Evan wrapped up, and was quickly mobbed by inquisitve software developers. Matt got the room under control and dispersed everyone with instructions to reconvene at JJ Foley's on Berkeley St. I rode the shuttle back to JFK/UMass chatting with Deb Nicholson about Occupy Wall Street. At JJ Foley's I found myself at a table with Tom and two guys having a heated debate about the right business model to use for running a Free hardware mail-order business. Josh Gay stopped by and gave a engaging if somewhat manic explanation of his "theory of change." I can't claim to have understood him fully. I was looking for a way to engineer an outcome; he seemed to take a descriptive rather than a prescriptive view of things: When we're successful the associated circumstances will be such and such. But maybe that's the a more sophisticated way to think about it. Have any of you read Anathem? The senior FSF strategists remind me of the monks that live in the center of the monastery and only come out once every thousand years. The hipster monks might find them frustratingly impractical, but they've got powers, babies.
I was exhausted when I checked into my hotel. The room was clean, featured a fancy writing desk (who uses those?), and the huge bed had an embarrassment of pillows, big and hard like breast implants. I showered, shat, and flipped through a few local channels on the TV before calling it a night. Some people were having a party in another room on my floor. I could hear it but it didn't keep me up. I started the next morning with a breakfast of eggs and hastily-scarfed spicy potato cubes in the hotel lobby, CNN Headline News playing on a flat-screen TV levitating above some ficuses, and then struck out for the university.
Eben Moglen was in the middle of his yearly update on the legal battlegrounds for Free software. As ever, much of his focus was on software patents. The patent wars continue, and his most recent conclusions were more pessimistic than in years past: Organizations have found that software patents have become more worthwhile to trade than they are to hold, which has turned the legal brinksmanship over software patents into a multi-billion dollar game and thus prolonged the existence of patents themselves. "We can't stop the patent war," he said, "and even if we were participants we couldn't stop it." And he said that the Free software movement would not be able to achieve its social and political goals until the war is over.
Mike Linksvayer and Chris Webber from Creative Commons were up next, with a somewhat sunnier presentation on the progress of CCv4. There were a couple of pleasant diversions from the charted course of their talk in the form of plugs for MediaGoblin and the Liberated Pixel Cup (relevant to my interests!) both of which are side projects of Chris'. But they managed to get across the important data about the license revision process; to wit, its goals:
- Internationalization: Beyond translation, the licenses need to be "ported" to the legal jurisdiction of other countries; in contrast with Free software licenses, a lot of this porting has already been done, but they're looking to do more
- Interoperability: ...with older versions of the Creative Commons licenses and with other types of software licenses
- Readability: For the lay public and lawyers alike
- Rights for additional media: Such as databases of indexed content (this is apparently more of a thing in Europe)
Yukihiro "matz" Matsumoto, the creator of Ruby, gave a charming autobiographical talk about how Emacs changed his life, beginning with his interest in hobby programming (the Ruby compiler) during the economic depression in Japan in the 1990s and how that led him to using Emacs, and then, as this type of thing often leads, to wanting to extend Emacs in the direction of better support for Ruby syntax (i.e., `ruby-mode'). And it was this activity that really got him deep into languages and software development. His argument, as best as I could transcribe it, in list form:
- Emacs taught me freedom for software
- Emacs taught me how to read code
- Emacs taught me the power of Lisp
- Emacs taught me how to implement a garbage collector
- Emacs helped me to code and debug
- Emacs helped me to write an edit text/email/documents
- Emacs helped me to be an effective programmer
- Emacs made me a hacker
- Emacs has changed my life (forever)
Matthew Garrett, now at Red Hat, talked about some hardware concerns for Free software developers, specfically the various implementations of "secure boot" and how they interact with non-vendor-approved (i.e., Free) software. Summary, from what I could understand: It's all bullshit predicated on some flimsy separation of hardware and software near the BIOS.
Karen Sandler, the executive director of the GNOME Foundation, and Joanmarie Diggs, who does accessibility development gave a talk on the status and importance of accessibility support in Free software, using the GNOME 3.0 development lifecycle as a miniature case study of sorts. One point they made stuck with me in an uncomfty way (I assume this was the desired effect): We're all going to need accessibility technologies to continue to use software systems as we get older. We are merely "temporarily able-bodied." Jonathan Nadeau, an FSF campaigns intern who also happens to be a blind GNU/Linux user followed up with a first-hand account of the state of accessibility software. He's a big fan of Orca, a screen reader that's part of the GNOME project. In fact, he was using it to read some of the slides in his presentation back to him during his talk. I'd never seen a system like that in action before; it was impressive.
Jeremy Allison delivered the keynote for the second day. Like Evan Prodromou, he's become sort of a household name in the Free software world, and he also turned out to be a fun guy to listen to. His talk was less structured than Evan's; he presented the history of the relationship between his project, Samba, and the GPL. He'd chosen it initially as a way to "clear the air" within the community of developers reverse engineering the SMB protocol, who had adopted a policy of secrecy to prevent their improvement from being co-opted by their competitors. At the time, using the GPL helped Samba become the SMB implementation of note -- to the extent that he had to stop accepting corporate copyrights on contributions, because contributors were attempting to use their patches to entrap their non-GPL-compliant competitors. (The GPLv3 has eliminated this technique via compliance grace periods and additional flexibility for source code delivery.) He also shared some amusing anecdotes he'd acquired from his years in Free software development: How the benefits of Free software were made manifest whenever he collaborated with Microsoft ("Oh, you have to write all your own software? That must take forever!"); how the initial jailbreaking of the TiVo was done in Australia by Andrew Tridgell, who wanted to help his friends at the U.S. embassy watch TV shows from back home.
RMS made an appearance right at the end to present the Free Software Awards (to matz and to GNU Health) and to briefly promote the use of LibreJS, which (finally) addresses the issues he brought up several years ago in his essay The JavaScript Trap. The "Stump Stallman" portion of the meeting -- the part where the members line up at the mics to ask RMS pointless questions about the GPL, or try to praise his ideas in ways that confuse and annoy him -- was mercifully absent. Hopefully the organizers finally grokked that this process was doing more harm than good.
By the time the talks were completely wrapped up and all the immediate chatting had subsided, the last free UMass shuttle of the day had already left. The group I was stranded were friendly and interesting (Evan, Mike, Deb), though, and I was glad to walk with them the mile or so to the JFK stop, even though it was a little unclear whether I was going to make my Amtrak connection. Deb and I talked about The Ada Initiative and OpenHatch and how she once booked Peelander-Z at a house party in Somerville. We made it to the T stop, and I said goodbye to everyone and ultimately did end up making my train with a few minutes to spare. I bought a beer on the train and tried to program, but the hops gave me a headache -- or more likely I was just worn out -- and so I just ended up closing my eyes. Jerry and Hanlon texted me in response to a celebratory Tweet I tweeted, and informed me that they happened to be driving back from the Cape at that very moment and would I like to catch a ride back with them. It would have been like the wild west! But I was feeling too depleted to want to be around people and so I said no thanks.
Amtrak to Penn Station; C train to Jay St. When the subway pulled into Fulton St., this hipster dude rushing to make the train took a header down the stairs. He got up, clearly dazed, and scrambled on board. He'd taken damage, though -- blood began to flow insistently down his face from a gash near his hairline. "Oh, dude," said his friend. A well-meaning passenger produced a wet-wipe. I took a step back and watched.