Why Larry Lessig gets an 'F' in software (News.com) [LWN.net]

Here's a News.com article about free software advocate Lawrence Lessig. "But Lessig is also going further. In his latest book, "The Future of Ideas: The Fate of the Commons in a Connected World," he draws a distinction between the intellectual property developed by, say, an Ernest Hemingway, and the intellectual property created by a code jockey."

to post comments

Why Larry Lessig gets an 'F' in software (News.com)

Posted Aug 24, 2002 10:30 UTC (Sat) by ninjaz (guest, #2083) [Link] (1 responses)

The crux of the matter is that Lessig discusses copyright as defined in the US Constitution "to promote the Progress of Science and useful Arts", while Cooper, the news.com opinionist only considers it a moneymaking device.

Since with binary-only software, the progress isn't being revealed, there is no justification for removing the right of people to share binaries according to how the Constitution has it.

Why Larry Lessig gets an 'F' in software (News.com)

Posted Aug 29, 2002 11:14 UTC (Thu) by tres (guest, #352) [Link]

You are right on the money when you state that the Founding Fathers wanted to promote progress. Providing an incentive, through copyright, to authors gives them an incentive to write the code but the real purpose is in promoting the public good. The ultimate goal is to benefit ALL! That means providing an incentive for the authors to write the code but at some point unexpiring copyrights benefit the authors to the detriment of the rest of us. That shouldn't be the case. We need to have a better line so that after the authors have gained a reasonable return on their investment that everyone else can start using their work. After all NO ONE writes code from scratch. When was the last time you started writing a program in hexidecimal opcodes? Every programmer in every case is building on the work of his/her predecessors.

Why Larry Lessig gets an 'F' in software (News.com)

Posted Aug 24, 2002 17:47 UTC (Sat) by lance (guest, #2263) [Link] (1 responses)

This article is a joke.

10 years is too short, the article says. And yet let's look at the state of software ten years ago. Windows 3.1, 3.11 is out, that might be useful source code . . . not.

Having a short copyright on software enhances innovation by allowing more people access to the code, they have an ability to learn from what their peers were doing.

This as opposed to everyone re-inventing the wheel just because Unisys holds patent on an encryption algorithm. The saying is, "Standing on the shoulders of giants.", and it means that people are able to build on what the masters of history have created.

lance

Why Larry Lessig gets an 'F' in software (News.com)

Posted Aug 26, 2002 1:29 UTC (Mon) by Peter (guest, #1127) [Link]

10 years is too short, the article says. And yet let's look at the state of software ten years ago. Windows 3.1, 3.11 is out, that might be useful source code . . . not.

Yeah, that's exactly what I thought. I almost fell out of my chair when I read that because the software industry is different from other industries, copyright terms should actually be longer.

I was discussing this with my dad a couple weeks ago, and he immediately picked up on the the point that, since software lives in Internet Time (ooooh I sound so 90s), the existing copyright terms are just stupid. And he is most definitely not a "computer person" and has never heard of the EFF, Professor Lessig, or even Internet Time.

Why Charles Cooper doesn't know what he's talking about

Posted Aug 24, 2002 18:47 UTC (Sat) by origz (guest, #1985) [Link] (5 responses)

The author (C. Cooper) seems to be a moron. He ignores several important facts, and makes many bad assumptions:

- 10 years is ages when you are talking about software. There is virtually no software that was made 10 years ago and that is still useful. What can you build on top of Windows 3.1 or DOS?

- Providing the source code does not impact the software company's ability to make money by selling software, even if the copyright length is 10 years. There are many for-sale perl scripts, and the authors seem to have no problem selling them. The same copyright protection that applies to binaries can also apply to source code.

If the source code for Windows 3.1 goes into public domain and M$ still uses it in Win2K (for example), nobody stops them from including that code in their product.

If the source code is open, companies will not be able to hide protocol, API, and implementation details from developers of competitive products. I can't see how this favors big companies, like Cooper seems to suggest.

Why Charles Cooper doesn't know what he's talking about

Posted Aug 26, 2002 13:07 UTC (Mon) by raindog (guest, #1235) [Link] (3 responses)

Actually, while I agree that copyright should be shortened at least to its original length, consider that at 10 years old, the first embryonic versions of the Linux kernel would become public domain, meaning the GPL would no longer apply to them and anyone could use the code in closed source software.

So what, you say, it's obsolete code. But over time it would become that much easier for an abuser to claim he was using an old public domain version of the kernel and was not subject to the GPL. It'd be like if the earliest Mickey Mouse films had been allowed to pass into the public domain as they were supposed to, and people started doing derivative works whose main character looked an awful lot like the modern Mickey Mouse. They'd probably still get in trouble, but it would be that much harder to make the case stick which is why Disney bought the Sonny Bono law. And at any rate, I don't think Disney could force them not to use the Mickey Mouse name to promote their derived-from-public-domain work (but I don't know; do they sue people over Winnie the Pooh?) So Linus' trademark could become diluted in some fashion too.

The same is true of GCC, emacs, and countless other mature free software packages where the core hasn't changed that much in the last decade. Maybe they should be public domain like Lessig suggests DOS and Windows 3.1 should be, but the fact that public domain cuts both ways is something advocates of his argument should keep in mind.

Clearly something needs to be done to re-establish the commons, but the two methods described here - the "legislate a shorter term" method and the "viral license" method - have some fundamental incompatibilities.

Why Charles Cooper doesn't know what he's talking about

Posted Aug 26, 2002 15:36 UTC (Mon) by jeremiah (subscriber, #1221) [Link]

That's kinda spooky really....We depend so much on the fact that our code can't be used without GPL'd repercussions, but a shot copyright time would kill that. On the other hand, the code from the Kernel of 10 years ago, is about as different as win 3.1 and XP or macOS1 and OSX. So it's probably not that big of a loss.

-jj-

Why Charles Cooper doesn't know what he's talking about

Posted Aug 28, 2002 0:02 UTC (Wed) by JoeBuck (subscriber, #2330) [Link] (1 responses)

It wouldn't make a great deal of difference if 10-year-old Linux entered the public domain, as there isn't much of use in that code. What you're forgetting is that in 1992, a lot of folks had source code tapes for Unix, the real thing, which in 1992 was way ahead of what Linus had. If copyright on software expired in 10 years, it would be much more valuable than 1992 Linux. The current Unix owners could probably try to use trade secret law to block its distribution though, I don't know.

There's tons of scientific and CAD software that's 10 years old that would make a huge impact if the copyright were suddenly made to expire; in many cases that stuff was coded better than what people are putting out today, because the machines were 100x worse, so people almost killed themselves to make it as efficient as possible.

My view: 10 years is on the short side, but what the founding fathers chose would be good enough for me: a 14-year term, renewable once for another 14 years. Furthermore you have to give the Library of Congress a copy.

Why Charles Cooper doesn't know what he's talking about

Posted Aug 29, 2002 7:23 UTC (Thu) by Peter (guest, #1127) [Link]

It wouldn't make a great deal of difference if 10-year-old Linux entered the public domain, as there isn't much of use in that code.

Bingo! Anything that old may still be useful, but will never give you a competitive advantage over the (former) owner of said code. In my opinion, trying to hold on to a code copyright for 10 years is about equivalent to trying to hold on to a movie copyright for 50 years: just plain greedy. That's why I've been trying to figure out the best way to express a "voluntary time-delayed public domain notice" for my own works - I'm still not sure of a suitable wording for what I want, which is that any code copyright I hold will expire after 10 years and (in the absence of others' copyrights on the same code) the work will fall into the public domain.

Anyone have facts or opinions of whether it's possible to donate a work to the public domain in a time-delay fashion - i.e. whether someone wanting to use said code several years later in some arbitrary jurisdiction could legally feel comfortable doing so without worrying about any copyright issues, and without having to contact me for clarification / reassurance?

Regardless of what I do with my own copyright notices, I really like the idea of mandatory short(er) copyright expiration for software, partly because it would mean that there is an upper bound to how many old copyright notices / license agreements you have to reproduce in your source code comments. Dozens of programs today have, embedded in the license agreement somewhere, a notice about Henry Spencer's regex library which he wrote in 1988 or so. Usually only one line saying it's completely unrestricted, but what if Spencer had attached some mutant version of the GPL or something - then everyone would have to reproduce a 17-kilobyte file just so they could use a 5-kilobyte regex engine.

Why Charles Cooper doesn't know what he's talking about

Posted Aug 29, 2002 10:58 UTC (Thu) by tres (guest, #352) [Link]

raindog stated:

This is exactly how M$ maintains its monopoly in the office suites. People can't write a document in one format and give a copy to anyone that doesn't have the software to view it. This even works to force people into upgrading their software because if their boss - who always has the latest and greatest software - writes something that his underlings cannot read the solution is to upgrade all of the underlings software - thus perpetuating the monopoly. This has NO bearing on the new features in the new software - just the new format of the files. This most definitely favors big companies!!