<D <M <Y
Y> M> D>

Dave Winer's comments in the famous Winer-Lessig-various people discussion are really starting to get to me. I'd like try to explain a few of the reasons.

First, re Lessig's proposals, or ideas:

Lessig suggested that software copyright holders should have to provide their source code to the Library of Congress, in escrow, so that it could be released when its copyright expires. Winer wants to know how this is going to be enforced.

Source code escrow is enforced by making it a condition of copyright, the way there used to be a deposit requirement. The deposit requirement has now been wiped out, but in the past there used to be certain expectations of anybody who wanted the benefit of a copyright. It really ought to be called source code deposit instead of source code escrow.

Source code escrow is not enforced by making it illegal to publish binary-only software. Critics of critics of software copyright keep on saying "Do you want it to be illegal to ...?". That happens to the Free Software Foundation, too: they get accused of wanting to imprison people who publish proprietary software, or something. At most, the FSF would like to cut away some of the legal mechanisms which facilitate keeping software proprietary. (One way to do that is to abolish software copyright, and another is to limit software copyright in any number of ways. Source code deposit could be one of those; compulsory licenses could be one of those; reducing the term of software copyrights could be one of those. I can think of three or four others. All of these are just limitations on what you get with a software copyright or how readily you can receive one.)

Lessig talked about a program Winer developed called MORE (which has disappeared, having been discontinued by the company which bought it, and can't be legally obtained now for love or money). Winer replied with a claim that Lessig's proposals would have precluded MORE's creation:

Had we been forced to release the source, I don't think we could have sold our investors on taking a chance on us, or realized the great return we got from the Symantec deal, and gone on to develop more software. The system you describe just wouldn't work, you wouldn't get any of it.

But this is ridiculous: Lessig suggested that source code needed to be escrowed, not released. There's no reason that the knowledge that the Library of Congress has a copy of your source code in a vault should make it any more difficult to bargain with investors, especially if all of your potential competitors' source code is also in that same vault.

Winer goes on to quote another MORE developer, Brad Pettit, in particular support of his position. Pettit does suggest that the source code wouldn't be very useful to the public today:

Probably 99% MPW Pascal with a little C and even a bit of 68k ASM for good measure. All pre-Universal Headers. Pre-PPC. Dense code. Lots of Pascal language features taken full advantage of: nested procedures, function parameters, etc. But you know that.

Fair enough (though Pascal's an awful lot easier to read than machine code, and it's not as though there aren't Pascal programmers out there; there are even COBOL programmers). Pettit also provides some direct support for Lessig's suggestion that source code ought to be escrowed somewhere:

A few years ago, Symantec asked me where they could find the code (how f-ing irresponsible is that?). I never could quite understand because the source for every release, on multiple floppies, was to have been in some vault somewhere. Go figure. Anyway, they were going to show the source to an outside party, but I don't know why. They tracked it down when I told them about the drive.

That's right: the copyright holder almost lost the source code forever after allowing the project to die. That's a large part of why there used to be a deposit requirement for copyright in general, to prevent creative works from being lost to posterity after they disappear from the mainstream commercial market. According to Pettit, that kind of loss nearly happened with MORE.

Second, in general:

Winer repeatedly accuses Lessig and other proponents of copyright reform of threatening Winer's profession and livelihood. (Sometimes he makes this accusation very vividly, but, hey, Dave Winer's a vivid writer, and I know some others of those.) He responds with a threat of his own: if copyright (after reformers get their way) is not sufficiently protective of his interests, he'll lose his incentive to develop software, and he'll go off and become a potter!

If n years of copyright for published software are not good enough for Dave Winer and he goes off to make pottery, that doesn't mean the end of software development. There was software development long before there was copyright for software at all. There are software developers who don't rely on copyright as any part of their incentive to develop software. I surmise from some of Winer's rhetoric that he doesn't think that those developers are, in general, "innovative" or "original" and that they're all just re-doing things which, you know, the real software developers (who like copyright) actually thought up years and years ago...

This gets to me, too. I've shipped tens of thousands of physical copies of my software product, I've just now spent a couple of hours checking patches into CVS, but I feel like I'm not to be considered a real software developer (by Winer) for some reasons, which could include:

On the other hand, I did develop a software product and shipped tens of thousands of physical copies and I have a couple of dozen glowing reviews from users floating around. (They are really glowing; they say things like "indispensable" and "saved my ass" and so on.) So I say that is real software development experience, even if not in Dave Winer's corner of the programming world.

Part of my confidence that software copyright reform could be non-catastrophic comes from my confidence that the free software community is making steady and substantial progress. I recently installed Debian GNU/Linux on an iBook -- not Debian's native platform -- and I had a couple of observations:

Now, before I installed Debian on this machine, I pronounced myself impressed with MacOS, and what you can accomplish by way of integration and functionality when you have proprietary control over hardware and software. It's really well integrated, and, as people have pointed out, generally easy to use. Apple has been able to define and enforce some very detailed standards for APIs and UIs all throughout the system, and they've been able to debug extensively against a closely controlled platform. And the results of that are impressive. You can expect programs to work together; you can expect aspects of the system to have been thought out with Single Vision (notwithstanding Apple's attempts to contrast themselves with other people who exerted a lot of control over their platform and their users). So, for example, on Linux, your printing subsystem may have been provided by any of several different groups of developers. (I can think of BSD lpr, lprng, and CUPS offhand, and I think there are one or two more. It's complicated enough that, like rival superpowers, the developer groups gathered at a Printing Summit, and it's still pretty complicated, and you're likely to run into integration glitches. And that's not because the software isn't good, necessarily, but it's at least in part because there isn't a single vendor-mandated printing infrastructure that everybody uses.)

But back to my experience with Debian on this Apple hardware.

My experience tends to confirm for me what I already believe: computing works well without proprietary software. Proprietary programs are still ahead in the mindshare and marketshare race, but I think they're still losing ground, and the alternatives actually work.

It's way beyond experiments with fuzz in 1990. I notice that most of the programs which Debian installed weren't even written at all when I started using Linux at home in 1995. (I used to install Slackware, and it used to show me each package it installed before installing it. It could even be instructed to prompt me about whether or not I wanted to install a package.) I noticed this because I literally didn't recognize most of their names. So people keep writing free software all the time.

Very few free software developers are relying on commercial incentives provided by copyright law. But, somehow, they're still writing software. I remember that somebody wrote, in response to a suggestion that economics would preclude free software development, that free software development is a fact, so, if a particular economic theory can't account for it, that theory is deficient; reality isn't.

Software copyright is probably promoting the progress of science and the useful arts by providing people like Dave Winer the opportunity to write software commercially. But people shouldn't be intimidated if he threatens to go make pottery. No single incentive is adequate for everybody, is it?

On the cryptography list, by Bear:

For a while I was thinking cryptographic protocols based on provables, self-interest, and the laws of mathematics (which are relatively constant and unbiased) might be a superior organizing principle for a society to laws. Under careful scrutiny and after acquiring a better understanding of protocol design, however, I concluded that such a society has probably as many problems (opportunities for tyranny and oppression) as the current nation-state concept. Its only definite advantage might be lower legal fees.

I went to David's movie night and saw a very strange animated series called Lain.

I didn't actually attend Worldcon, but I did go to San Jose to meet with a delegation organized by the UK Publishers Association, which was visiting Worldcon. So I was at least physically present at Worldcon.

The publishers were very nice to us, as were our and their hosts from the U.K. consulate in San Francisco, who helped organize the delegation's itinerary. (I got to see the business cards of some of the other people they'd met with, and I recognized quite a lot of the names.) It seemed to me that U.K. publishers are much less radical and much more reasonable about copyright than U.S. publishers, but maybe that's just because I'm still angry with the AAP about how they handled Dmitry's arrest.

(Amusingly enough, the meeting was held within sight of Adobe headquarters and our parade route from the Free Dmitry march, and in fact I walked past Adobe and walked that very parade route on my way from Caltrain to the meeting. This tended to put me in mind of Dmitry's case, not that it came up during our discussions, oddly.)

I've been feeling that there is a hierarchy of radicalism about copyright within the copyright industries. Movie publishers' trade association (MPAA) is most radical; music recordings' publishers (RIAA) less so; print publishers (AAP) again less so. There are, of course, dozens of other trade associations within the U.S. copyright industries, but none of them seem quite so outspoken as those three publishers' associations.

Print publishers do face a different situation from the entertainment publishers, in various ways. Most of their publishing is not now digital; most of their publishing is not a "recording"; most of their publishing is in a format whose physical characteristics are significant; most of their publishing is not usually considered "entertainment". The ways and reasons people buy books are different from the ways and reasons they buy other copyrighted works.

But I was saying that the U.K. publishers were nice to us and interesting to talk to. I'm glad we got the opportunity. One of the things they're working on at the moment is notice-and-takedown in the U.K.; they'd like to have it, but, it seems, they'd like it to be a little better than U.S. notice and takedown (17 USC 512). We suggested that they take a look at Chilling Effects to see a bit of the variety of the ways 512 is being used today.

What is it with these powers of two? (The CBDTPA is S. 2048.)

The publishers' delegation seemed very supportive of the right to parody and the right to excerpt for criticism, and they were disturbed that notice and takedown was being used against people who did those things. As far as they were concerned, notice and takedown was supposed to be used against people who were copying a work in its entirety.

As you might imagine, we spent a long time talking to the publishers about the evils of anticircumvention, and why any anticircumvention provision should require an underlying act of copyright infringement, or the intent to commit an infringement. We should see if the Publishers Association comments on the implementation of the EUCD.

I had the honor at our meeting of meeting Patrick Nielsen Hayden, and a little later on of meeting Craig Newmark.

(The description below is mostly cribbed from an e-mail message I wrote.)

After the meeting, and a good deal of socializing, I accidentally lost track of the person who'd offered to let me share his hotel room. (I'd missed the last Caltrain during that socializing, and Caltrain service is temporarily suspended on weekends this summer anyway, in favor of some strange bus service.) That meant that I had no place to stay in San Jose and no way home, short of a taxi.

So I walked back to the hotel where we'd held the meeting and asked them if they knew a way back to San Francisco after midnight. They didn't.

I went outside to a VTA bus stop and found that there was a bus from the Caltrain station (which I considered walking distance away, since I'd walked from Caltrain to the hotel) at 6:30a which would take me to the Fremont BART, from which is merely a long BART ride home. So I thought I could possibly stay awake, maybe using my laptop at the hotel to do some writing, or some reading, or some playing video games, as long as the hotel didn't decide to kick me out -- and, if they did, I could try to sit in a public park until about 6:00a, and then walk to Caltrain. However, the "staying awake until 6:30a" part seemed kind of challenging.

So a light rail train pulled up and I asked them how to get to San Francisco. They were a "test train" (no passengers), but the train operator kindly calling in to VTA headquarters to ask them to advise me. They said "take light rail to San Fernando, take the 22 to Menlo Park, and catch a SamTrans bus there to San Francisco".

San Fernando seemed far away -- like Southern California, right? -- but I asked some other people who were waiting for light rail and they told me it was a street in San Jose, two blocks away. While we were talking about this, a 22 bus passed by us. Oops! So I waited at a bus stop where the 22 had gone by for a while until a woman waiting there for a different bus warned me that it wasn't a 22 bus stop, merely a bus stop which happened to be along the 22's route, and that I really should go to San Fernando the way everybody had advised me to.

So I walked over to San Fernando and finally found a bus stop which looked promising. After waiting there for quite a while, I caught a northbound 22 VTA bus. I fell asleep on the bus somewhere around Sunnyvale and missed the Menlo Park stop, and I only woke up at the last stop on the 22 line, which is the Palo Alto Caltrain station. Oops again.

It turns out that there is a SamTrans bus from the Palo Alto Caltrain station to San Francisco, though -- the 397. (I think that's a different bus from the one I was supposed to catch at Menlo Park, which I think was the KX.) One of those eventually showed up, and I fell asleep again, and woke up somewhere along Mission Street in San Francisco, shortly before the 397's last stop at the Transbay Terminal (1st and Mission).

There, as I'd hoped, I was able to catch a 14 Mission MUNI bus and take it (falling asleep again, as you might expect) 23 blocks to 24th and Mission, and walk home, getting in just before 6:00a.

So, my non-Caltrain trip home took five hours in all, including waiting, walking, and riding.

I'm reminded of one time I got stuck without a place to stay in Berkeley and missed not only the last BART train but the last AC Transit F bus. It turns out that there is an all-night bus from somewhere near downtown Berkeley to downtown Oakland (12th Street), and then you can wait a while for the all-night A, which has service to the Transbay Terminal. (And from there, of course, you could take the 14 Mission, although I lived within walking distance from the Transbay Terminal at the time.)

So it turns out that there is actually 24-hour service between Berkeley and San Francisco, and between San Jose and San Francisco, but you're likely to spend at least an hour waiting at bus stops, and maybe several hours. I had some good books, too, but I was too tired to focus on reading.

Matt Obert posted to crackmonkey with the subject "El-Haraty vs. Moglen":

I don't get Emad, I get Eben.

[Main]
Support Bloggers' Rights!
Support Bloggers' Rights!


Contact: Seth David Schoen