I woke up at 6:00 AM on Saturday the 1st, still sort of drunk and sick from Williamsburg Porkapalooza 2006, and hauled myself to East Broadway, where I was the last person to get on the 7:00 AM Lucky Star bus to Boston. This is distinctly similar to what happened last year. But, yeah, I spent the day at MIT listening to the presentations at the Free Software Foundation's annual members meeting.
I'd assumed that the meeting would be in the same room -- Stata Center 155 -- as it was last year, but when I got inside, there appeared to be some sort of Women-in-Computing seminar going on; the ladies at the desk were nice enough ("Free Software Foundation? Cool!"), but had no idea where I belonged. After some unsuccessful wandering around, I actually called my mom on my cell phone and she looked up the room number for me on their web site. It was Building 34, room 101. Unfortunately, Building 34 turned out to be pretty impossible to find. While I was wandering around hopelessly, though, I ran into none other than Eben Moglen, who was looking for the building himself and had also gotten fairly lost. I tagged around after him as he kind of huffed and puffed up and down a few flights of stairs, but we were pretty ready to admit defeat after about 15 minutes of following promising signs into dead ends. By a stroke of luck, just as we were about to give up we ran into Gerald Sussman, who was going to pick up a projector from his office. He walked us through some dark and austere corridors that we would never have found on our own and eventually we made it to the meeting.
As we were walking, they discussed the difficulty of finding RMS accomodations that would be provably free of smoke (I think). Eben said something like, "At least he's complaining about himself so much right now that he can't complain about how unhappy the state of the world makes him."
I got in in the middle of Geoffrey Knauth's speech -- he's one of the more economics-minded people on the board. He was talking about whether exporting Free Software to the developing world was hurting job prospects in the first world, and I was hoping to pick up some good talking points, not least of all to convince myself, since I'm not clear on a lot of the macro parts of these issues. His argument, though was basically that the first world is still producing the most software expertise and thus exporting the highest quality of Free Software, and that we'll know when the recipients of this expertise stand a chance to move in on our job market when we start seeing high quality Free Software coming out of the developing world. I don't know if I buy this, necessarily. Kind of anthropic.
Afterwards, I bought a neat little lapel pin from someone at the merch table who looked a little bit like RMS's girlfriend, Tania; in retrospect, it wasn't her at all.
Sussman gave what I thought was pretty much the same talk as he gave last year on the importance of interchangeable, standardized components. He did make the interesting point that robustness in biological systems is deeply related to diversity; we need support linguistic diversity in programming languages for the same reasons. He also discussed what he referred to as "paranoid programming," the idea that no input can be trusted, nor can the output of any interchangeable parts that are used by the program; data needs to be annotated with some representation of its "source," so that problems with calculations can be isolated and resolved after the fact. Somehow we got to self-organizing systems -- I guess he was making the point that a satisficing algorithm does not always behave deterministically, or even in a way we might expect. Vein structures in the human hand, for example, differ from person to person because the mechanism for laying out veins is organized around covering an oxygen topology, and the availability of oxygen during vein development is dependent on environment.
After that, we broke for lunch -- they had substantially the same fare as last year, which, you know, was good. I was feeling pretty hung over from Katharine's party, and when I got up from the steps I'd been sitting on while eating, I noticed I'd left a big gross ass-sweat mark. So I went to the men's room and tried to take a crap, but people kept coming in, including a guy who was taking a piss but must have had prostate problems or something, because he could only piss in these weird short little bursts that seemed to require significant abdominal effort on his part -- so much so that he let out this tremendous fart at one point. I made a coughing noise to remind him he wasn't alone, but I don't think he was concerned.
Eben Moglen, who was next, opened his talk with, "Vista will be late, Office will be late, Virtual Server will be late, but the GPLv3 will be on time. Free Software is better." This met with a good deal of appreciative noise from the audience. The brunt of his talk, though was on how GNU/Linux -- and Free Software in general -- are set to make enormous gains in the embedded market because of the economics inherent in that sector. What he said, semi-verbatim, is that if you're Nokia or Siemens or Sony, say, and you've got a set of diverse hardware that you need to sell to consumers, you need to have a software platform that is robust, very well-understood, fully debugged, and absolutely secure. And it needs to be 100% free, financially, because otherwise the guy who makes it is going to eat your lunch. And what meets that need is Free Software -- it's become an essential raw material in consumer electronics manufacturing, and it's not replaceable. However, the move towards "Trusted Computing" has thrown up some stumbling blocks for Free Software, because TC methodologies rely heavily on non-Free cryptographic interfaces to hardware. The GPLv3 will do a lot of work towards making TC and thus DRM irrelevant, but he made the point that the industry's idea of a "trusted" kernel that meets their robustness requirements is basically a pipe dream, given that kernels are, by nature, too big and too volatile to be constantly re-assessed for "trustworthiness." As such, engineers worried about "trust" are moving more towards thin virtualization layers or application-layer DRM, both of which make conflicts with Free Software people less intense.
Winning the war on restrictive hardware, he said, is a conservative activity (I think he really meant "conservationist") -- we need to constantly emphasize the consumer demand for general-purpose computing hardware. But organizing consumers is always difficult.
He also said that the FSF has been watching major technology players get on board with TC and DRM for a long time and warning them that it was dangerous, and then "we made some very reasonable remarks about DRM in the GPLv3 and everyone went nuts. That's really what happened -- they went nuts. And I'm not talking about Linus. Linus did not go nuts by any means."
Ultimately, though, he thinks, The Time Is Right to push on industry.
When he opened the floor to questions, I asked him if he thought the state of mounting software patent aggression had changed since last year, and he gave a very long and interesting answer to the effect that it hadn't changed drastically, but that there'd been some high-profile legal skirmishes that have made a number of big players wary of participating in patent-hoarding. He also mentioned PubPat, which I hadn't known existed, and discussed some cases they'd been involved with.
RMS was up next, and, like last year, he gave a rather poor showing -- a short (16 minutes), rambling talk about the dangers of DRM. I asked him afterwards if he'd changed his position on the necessity for Free licenses for non-software creative works given the argument he'd had with Larry Lessig at last year's meeting, and he vehemently denied getting into an argument with Larry at all -- he claimed I must have read an article that misreported the event, and I was like, well, you know, okay, fine. But he did say that he'd come to believe that Free licenses should be encouraged for certain types of creative work, although he didn't really get too deep into discussing that. I was pleased that fewer people in the audience seemed to be interested in baiting him, though that didn't seem to make him any less ill-tempered.
Henri Poole had somehow wound up with the unpleasant task of soliciting suggestions from the members -- his presentation was called The Member Forum, and was basically all about organizing people into geographic delegations and soliciting suggestions for activism from them. He's sort of the most friendly-looking member of the board, but he also always looks like he's got this secret pain, like he's been gut-shot and is trying to hide it from everyone. I was actually a little bit psyched to meet the NYC / Brooklyn contingent, but it turned out they were all complete douchebags! The two ugliest and dumbest guys there had both been former employees of the FSF and began practically every sentence with, "When I was at the FSF..." And, you know, that wouldn't be a problem if they had anything smart to say, but neither they nor really anyone else there seemed to Get It when it came to what the FSF needs to do to leverage public support. These guys were really hung up on the sort of "reach one person" style of activism, where you give really breathy, earnest, personal speeches about stuff to roomfuls of senior citizens and people with weird and unpleasant disabilities who don't have anywhere else to be in the middle of the day. Look, I don't have any activism experience myself, but it seems to me that what the FSF needs is more public visibility-focused initiatives, like the Firefox full-page ad in the NYTimes. The FSF needs to get on peoples' voting radar, and once they've done that they can focus on handing out free OpenOffice CD-ROMs at the veterans' center. That's a luxury activity. What the FSF does not need is to recruit more pushy, wall-eyed people with acne scars who insist on saying "Treacherous Computing" when they're having a conversation with you; that tack is right for writing a letter to the editor, not for talking to Real Live Humans.
The capstone on the dumbass member forum was the chubby, smug former-FSF beardo saying that having members give speeches at public functions is infeasible because public speaking is so difficult. "We'd basically have to send everyone to Toastmasters," he said. "That's where Richard learned to be such a great public speaker." RMS is probably one of the worst public speakers I've ever seen. I mean, I'm completely devoted to his movement, but he's a surly, slouchy, mumbly piece of crap in front of a microphone. Get off his jock.
So that part of the meeting put me in kind of a bad mood, but then on the way back home on the bus, Maggie talked to me for two hours on my cell phone, which was delightful. Plus I am never going to get sick of riding around in cars and buses in New England looking out the window. I'm sure it's just that I've got so many happy memories of things that happened in Massachussetts (trips with my family and my other family), but I swear that state is the most beautiful in the Union. The flora, the fauna, the sights, the sounds, the smells. I love it.
No comments:
Post a Comment