Whoop! 

Random acts, cont'd 

Finding agreement 

All the colors in the monochrome 

Feel good, automatically 

Going offsore 

Blogseeding 

Can they die even sooner, pleeeease? 

discuss

I tried to counter and think I got the major jist right on my weblog..

Basically both sides are arguiing for return to common good, they are just arguing from different experiences. Which does nto by itslef produce opposing views..they are just talking in different experience languages..that is all..they both want the same exact thing!

Josh tries to make apolitical arguement out of it.. whereas I don't see how one group chosing lib, demo, or repub and the other group chosing something else makes opposing views since both choices are being made on the basis who has the most honest chance in getting the wrong bills voted dwon rather than made into law..

discuss

discuss

discuss

discuss

It would also be easy to see Josh's post as a classic divide & conquer tactic. A few years ago Microsoft was on the sidelines about issues of copyright and software patents: avoiding the issues, basically. But in the last few years Microsoft has gravitated toward the copyright absolutisms of the RIAA, Eisner and the MPAA.

But I don't believe Josh had a corporate agenda in his post. He apparently just doesn't like Larry Lessig and open source, and sees them in absolute terms. That's why he thinks all open source advocates line up behind efforts to push a Digital Software Security Act, and why he also thinks Larry Lessig is "a collectivist opposed to individual property rights."

discuss

Lessig makes me uncomfortable. He doesn't say enough to make me believe we see things the same way. I've never heard Lessig say that he supports copyright at all. Have you??

BTW, fourteen years isn't enough. We haven't yet recouped our investment on Frontier, and if fourteen years was the standard for copyright, it would have expired in April of this year.

Has Lessig studied commercial software? Have you ever heard him talk about commercial software without binding to Microsoft? He's supposed to be a learned man, a scholar, a thinker and researcher -- but I find his arguments are missing a lot of substance. So much so that I suspect we are not allies or friends on most issues.

From Missouri,

Dave

discuss

linkage: http://www.wamu.org/ram/2001/r2011220.ram

commercial software without binding to microsoft.

linkage: http://www.wired.com/wired/archive/9.12/lessig_pr.html

Dave, I would be interested in hearing where you guys specifically disagree. Where, in your opinion, do his arguments lack substance. Larry has a lot of his articles, presentations and interviews available on his website www.lessig.com.

regards, kevin mcmahon

discuss

discuss

Further, why would if matter if Lessig studied commercial software (he did) or not? If he didn't, does this make him "unqualified" in some way, then? If so, similarly, one could ask if Dave studied the law. Since Dave did not, does this make Dave "unqualified" in legal matters? The answer is "no" in either case.

FWIW, I don't think Lessig and Dave see things the same way and there's nothing wrong with that. Lessig is coming into the arena from a legal perspective. He's saying, "Look, all you developers, open source or commercial, you better start organizing POLITICALLY, or your technical genius is going to be neutered by politicians opposed to you."

discuss

discuss

If society is to give software producers more protection than they would otherwise take, then we should get something in return. And one thing we could get would be access to the source code after the copyright expires.

I believe the remark about copy protection, plus the whole obsession with source code, serves Dave's case.

Perhaps copyright law is concerned about the protective qualities of closed source. But the real reason to keep it closed is simple: it's easier to sell, and to ask a price that's north of zero. The market actually wants it that way, on both the buy and the sell side. That's the lesson VA Linux (now VA Software) learned. That's the lesson CollabNet learned too. That's the what's behind the new dual- licensing scheme that CollabNet is helping put together for RealNetworks. As Rahul points out, there are other variables. It's more complicated than it appears.

I still don't think these guys are that far apart. We need to learn from each other. Larry has learned a lot from the open source guys. Now the commercial guys who aren't Microsoft need to teach him the rest.

discuss

P.S. Lessig is 40. He has seen civil disobedience up close. Indeed, he helped write some constitutions in Eastern Europe after the fall of the Soviet Union. Again, all of this is in his writings.

discuss

discuss

discuss

We can sit in our offices and write software forever and call ourselves freedom fighters but if the government prevents us from sharing then it doesn't matter. We must reform the government directly. I believe that is Lessig's point.

I also agree with this point: "Once a company that produces a certain product goes out of business, it has no simple way to uncover how its product encoded data. The code is thus lost, and the software is inaccessible. Knowledge has been destroyed."

I write software for a living but I don't write products. I do enterprise softwaree - I'm an itinerant software architect basically. I buy components from venders and stitch them together and automate business processes. Its software engineering in the collossal.

Selecting components is incredibly risky. Apart from looking at the software, I have to look at the financial health of the company that sells it. If the company vanishes tomorrow, what happens to the value of my purchase? We are all familiar with bit-rot. Every new OS release breaks some software. You can't say, well just stick with the old version because network effects eventually pull you forward. This is the strategy MS has been using for years to force either OS upgrads or App upgrades. I don't think I have to explain it - we've all been victims of it.

It is for this reason that I prefer open source components over proprietary every single time. Because there is no safety in commercial components - they can vanish at any time. OTOH, if the source code is available - then I feel secure that I can keep the system runnning by taking on the maintenance of the component. If the component is really good, then a community will arise to take this over and we can share the cost of keeping our systems running.

In fact this is exactly why there is an opensource movement at all. Because almost every single developer that contributes to opensource has been screwed by the loss of some important copyrighted software. Some important chunk of work we've done that uses or builds on the hidden software was made worthless instantly.

So I'll agree with Lessig that software source code should be required to be opened. In practice, it turns out that this isn't the big vulnerability that the software vendors claim. Any sizeable chunk of work requires a great investment of resources to understand and very few people are motivated to undertake that. Most of us have bigger fish to fry.

Incidentally, Apple's OS code is (mostly) open for inspection via the darwin project. Solaris has been opened as well. We aren't seeing a lot of "ripping off" of that stuff because it would cost a lot to do it, the license forbids it anyhow, and there's plenty of other more interesting work to do. Its also interesting to note that what's not open is being cloned via the GnuStep project.

So you can open your source code and still maintain control of your property. Besides, it would be easier to tell if you were being ripped off (plagarised) because you'd be able to spot inclusions of your code into other systems because they'd be open too. Plagarism is still prosecutable, right?

discuss

As for maintaining control of source code while leaving it open, yes, that's quite possible (Apple, Sun, Real and many others are doing that with core products). But generally companies are doing this with foundational code, not with the commercial code they build on top of it.

The serious questions Dave keeps raising are commercial and creative: How much more can you charge for closed bits than for open ones?; and How much creative software would never get written if there were no money to be made with it?

The answer to the both is apparently a lot. This is why surviving "open source companies" like Caldera, VA Software and CollabNet have been selling closed as well as open source software, and helping their customers do the same.

Out in the real world marketplace, it's not a black-or-white issue. This is why making laws to force one or the other are so misguided. Dave once said open and closed source are best seen like hot and cold running water. You need to be able to mix the two. I haven't run across a better metaphor than that.

I think it's still early in this thing. I've said before that the software industry when it matures will be much more like the construction industry than the manufacturing-like industry it resembles today. But let's cut it some slack: we've been building with bits for only a few decades, and with rocks and wood for thousands of years.

What matters most right now (to me, at least) is that we keep the Net free. On that issue Dave and Larry are on the same side.

discuss

Perish the thought that some in Washington might actually listen to him and legislate based on his beliefs. I think the first person to straighten out is the one who shows up at the conferences and gives speeches based on an incorrect understanding of what we do in the software business. I don't think I'm on the same side with Lessig on very much, despite what you say Doc.

I take copyright for granted, that when I create something I put a copyright on it. Lessig thinks my work doesn't justify that. Basic disconnect. Disrespect like that makes me angry.

discuss

If true then closed source is hot and if you use too much of it you will get burned. Open is safer.

There's another force at work here. The only companies that can afford closed source are the really large ones. Think about that. If you are not a really big company, I probably won't buy your product unless I can get source code. Oracle can stay closed source, IBM, Microsoft, but anyone smaller and I'm not going to use your product unless I get source. Its just too risky for my (typically huge) client.

So having closed source and being small is costing the small developer the ability to sell to large clients.

Sounds unfair perhaps, but should Pac Bell build a customer service application in something like Dan's killer scripting language? No way. If Dan hits the lotto or gets hit by a truck then millions of dollars in development effort evaporate instantly.

Its a hard sell and I wouldn't risk my professional reputation by even presenting it. Its a lot easier to push something like php for the same job because we know that if the php developers suddenly move on, we're not up the creek.

I'm not trying to argue a point, just share a view. Closing your source closes doors.

discuss

What would the web be without open source? The web is open sourced. That is one of it's main advantages. It is why everyone can easily make a webpage. People are making money from the web. The source of a webpage is still subject to copyright. And it doesn't come with copy-protection schemes.

Dave, I thought the net developer community exists because of open source. That's why it's bad when Microsoft tries to make Microsoft proprietary codes/tags for IE, it breaks the web. Haven't you raised the specter in your own writing of Microsoft "closing" the web?

Lessig's argument is simple. Open source is useful. It's more useful to the public than closed source. It is the people, through goverment that grants copyright, and copyright should serve the interests of the people, not the copyright holder. Granting copyright to an author is a means to end, not an end in itself.

"Clause 8: To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries"

"Writings" here is the code. Copyright is granted to the what is authored or discovered...unless compilation is authoriship the copyright is granted to the source. Promoting the progress of Science and the Useful arts is better achieved through open source than closed source. I know this through personal experience if only with networking related communication...think token ring vs ethernet, tcp/ip vs appletalk/ipx/etc..., http/html vs opendoc/cyberdog or msn/blackbird

Doc, you said you want to preserve a free net. Well, it's free as in open, not free as in beer.

-calvin

discuss

This is not lost on commercial open source developers like Collabnet, Caldera, Trolltech, VA Software and Borland. They're answering those questions with mixed commercial and open source licensing schemes. It's not a comfortable subject in the open source world (I'm about the only writer talking out loud about it) but that's what's happening. And it's because these companies need to do business in the real world marketplace.

Again, I don't think this needs to be as complicated as it seems, but it is, because it's political.

Craig Burton says most of our problems are technical, political, or both; and the political ones are the toughest to solve.

Once again, the problem with the Open vs. Closed battle is the OR logic that comes out of the politics. In the real world, the answer most often is AND.

discuss

You mean "published" closed source or private? I think COBOL is still the biggest language in the world in terms of lines of code in existence that are currently in use. How much of that is "published"? Practically none of it. The so-called "commercial software" is probably less than 10 percent of all the software in use today. Maybe more like 1 percent. I'll wager that PacBell has more lines of code in service to run its business than Microsoft has written in its entire history. Thats one company in one industry. I'm guessing we leave this out though since its mostly unpublished.

Of course not all source code needs to be published. Only those programs that are offered for sale under copyright.

As is mentioned before, Hemingway, Stephen King, even Danielle Steele don't have the benefit of closed source. But they publish all in plain text and still manage to make money. Why, pray tell, is software supposed to be different? Developers get the same protections under law that every other creative author gets and still they're not happy with it. Why are they special?

I'm going to step onto a soapbox for a second here. The state of the practice in software development technology has been in a state of deevolution since 1980 or so. Not only has progress stopped, it has been rolled backwards and its largely due to closed source. I'm now watching with an extremly jaded eye a replay of the 90's. CORBA has been relabeled web services and its proponents are about to rediscover from scratch the value of naming services, transactions, and all the rest of the distributed infrastructure - only now in bloated XML formats rather than the compact binary message formats used before.

Progress? Where? Its all a big circle jerk and I'm on my second loop (having joined around 1989 or so) because nothing endures and we have a whole new generation of prognosticators who are completely ignorant of what was learned on computers science in the last decade driving the so-called "new" standards. Jane stop this crazy thing. Where is the body of knowledge in our field upon which we are supposed to build our future?

Gone. Up in smoke. Poof. The entire body of lessons learned in the last decade are locked away in the backup tape vaults of companies that have either been crushed by our favorite monopolist or simply changed tactics and discoutinued the "out of fashion" line.

Bottom line. So-called intellectual property rights are destroying America's competitive edge and ability to innovate (note lack of trademark and capitalization). If we can't stand on the shoulders of the giants of the past, then we'll never rise above their heads. We're being robbed.

I'll step off my soapbox now. Thanks for listening.

discuss

Dave: "Lessig's analogy doesn't work. You can not see how Hemingway wrote, you can only see the words he published, the publication does not reveal the process. Similarly with software, students are free to study the published work, and use that as prior art (patents notwithstanding). Lessig's analogy is wrong."

Me: I agree that Lessig's analogy isn't perfect, but its not nearly as far off as you claim. Without the source code to an OS, virtually nothing can be learned about the algorithms and techniques involved. I think its a mistake to say that programs reveal as much about their creation as novels do. It is true, however, that a fair amount can be learned from GUI apps without access to source code.

Dave: "Further, I can sing a song I heard on the radio, but with published source, anyone would be able to sing the song as well as the person who created it."

Me: Here is a failed analogy on your part. Publishing the source code does not give anyone without a great deal of time and intelligence the ability to do anything with it as well as the author. Sure, I can USE the software as well as the author, but I could already do that without the source. What I can't do as well as the author is make changes to the code, at least not without expending a great deal of effort to learn the software.

Dave: "There's a reason why open source software hasn't produced very many memorable melodies. Programmers have to make a buck to keep programming. It's pretty simple."

Me: I think your really misunderstanding Lessig here. Source can be "open" and still not be freely distributable. Copyright and "open source" are not mutually exclusive. Lessig's view is quite different from Stallman's. This is something thats sort of a pet peeve of mine, and its something I wish people talked about a lot more. You can give a paying customer the source code to your software without giving them the right to redistribute it or incorporate it into their own commercial products. Of course, the latter would be quite difficult to enforce without laws requiring source code to be open.

Dave: "Or take Lessig's course from 1989 and get just as well educated. Software is unique among creative and thoughtful work, in this way. The source gives you an exact replica of the original. And software needs to be upgraded. It costs money to do that. That's why we don't publish the source code, so we can keep working on it."

Me: As I've stated above, its takes a lot of time and effort to grok source code well enough to make changes/upgrades to the code nearly as efficiently as the original author. Giving customers your source is not going to stop them from asking for improvements. I doubt developers like you would take any noticeable hit at all. Oftentimes, open sourcing a product ADDS value.

The bottom line is what I call my asshole principle, "If you can improve the world without any significant cost to yourself, then why not do it?" Only an asshole wouldn't. Come on Dave! Don't be an asshole. Lessig is not trying to destroy copyright or rob you of your capacity to earn a living.

Jonathan Simpson

discuss

I also believe Larry is doing extremely valuable -- even heroic -- work.

But that's not what we've been talking about (or trying to talk about). We've been trying to talk about the software business.

Dave believes (and I'm sure he'll correct me if I get this wrong) that Larry doesn't understand the software business itself, in all its real-world complexities. Larry seems most familiar with the venalities of Microsoft and the virtues of the open source community. But there's a heckuva lot more going on out there -- even within Microsoft and the open source communities.

Some big and obvious companies, like Microsoft and Adobe, give away as little as they can, try to create industry-sized dependencies on their stuff, and generally don't give a whole lot to the free and open common infrastructure we call the Net. Worse, when some of them die or abandon old code, it isn't given up to the public domain so the rest of us can make use of it. That's a problem. But it's not what characterizes the whole industry.

Some sompanies, like Apple, IBM, HPm Sun and Real, now have open source strategies of one kind or another (in the latter case for the first time), now that they see what good can be had from participating in the free & open parts of the industry's ecology.

Meanwhile, companies like Dave's have been deeply involved for a long time in the market ecology where commercial software sits on the Net. These companies make constant choices between sharing stuff on one hand (XML-RPC, SOAP, RSS, MacBird) and making money on the other (Manila, RadioUserland). They also know, intimately, that their best work wouldn't get done if they didn't get paid for it, and that hey wouldn't get paid for it if it weren't sold as compiled binaries.

My point: in the software businessthere are an endless range of choices, of distinctions, and of many variables, just one of which is closed/open source. Frankly, I've been around that business one way or another for 25 years and I still don't feel comfortable claiming I understand it.

But I am clear about what I'm *fighting* here, and that's unnecessary lawmaking and regulating. On that count I find Dave and Larry on the same side.

discuss

Here's the correction you were looking for.

There's no way to know how he got to that place. I think perhaps Ernie the Attorney had the best explanation. There is an attorney point of view of software. Ernie explains it well. It would be arrogant for me to say that anyone doesn't understand the software business, I only understand one slice of it.

Anyway, after endless wheedling, let's try another angle, and forget about Lessig for a bit -- he isn't here, so who can tell what he would say about this or that.

Question: Where's the harm in keeping the system as-is, minus the patents.

Give me an indefinite copyright on my own work. But I can't stop anyone from competing with me. So if I don't open up some aspect of my code, and that's a mistake, the market will correct me, and one of my competitors will eat my lunch.

Where is the problem with that?

discuss

A legal argument against indefinite copyright is the violation of the Constitution - the "limited Times" clause. This is why Jack Valenti would like copyright to be "forever minus a day".

The economic argument against indefinite copyright is that monopolies decrease net social benefit. And let's not gloss over the fact that copyright is goverment-sponsored monopoly. It is communism at its finest and it's written into the Constitution. (I personally find it ironic that those who cheer for capitalism the loudest are also the one's who want goverment-backed monopolies to help them out.) Economically, society wants to grant copyright (and patents) for a length of term that will maximize net social benefit. Indefinite copyright does not maximize net social benefit. The most compelling brief filed in the Eldred case is not Lessig's brief, but the brief of a wide range of economists who show that extending the copyright term does not confer social benefit to society. Indeed, I believe most economists would claim that 75 years of protection would not maximize net social benefit and is thus is too long.

Finally, what would happen if, heaven forbid, a developer passed away suddenly? All of his work is protected by indefinite copyright. Legally, the developer's work would be essentially locked in a vault perpetually. Society would never be able to benefit from the developer's copyrighted work. This is destruction of knowledge and a social loss. And copyright's goal is to serve the public, not the copyright holder.

discuss

discuss

"Question: Where's the harm in keeping the system as-is, minus the patents.

Give me an indefinite copyright on my own work...Where is the problem with that?"

Regarding legal protection of IP (copyright and patents) the Instapundit legal paper you linked to yesterday (which I thought was quite good) says:

They must be "for a limited time" because part of the public's payback is that the invention or creation will eventually enter the public domain. They must also be reasonably calculated to "promote the progress of science and the useful arts," meaning that they must bear some relationship to creativity, not simply political clout.

To sum up. Copyright and patent law is a compromise between the creators of IP and the general public at large. The government wants to foster progress and innovation but progress (in anything) implies the ability to build on the past. Imagine the state of physics if Newton had copyrighted and kept secret differential calculus and only published the results of his calculations without ever revealing how they were done. Thus all creative work *must* eventually pass into the public domain.

So the government makes a compromise. Works are protected for a limited time to allow the creator to cash in for a bit - but then the work has to go into the public domain.

I think the problem Lessig is on about is that closed most source software never makes it to the public domain in any meaningful way. So the software industry isn't holding up its part of the bargain. Fine, you cashed in - now where's the advancement of the art we can all build upon? We (the commons) never get it.

Thats the problem with leaving the system as it is.

Possibly, the answer lies in depositing all commercial source code into trust to be opened when copyright expires. Thats just an idea off the top of my head. I haven't thought it through but I think it would satisfy the aims.

In some way - I think the open source community is a compensating mechanism. Those old unix utilities should have been released into the public domain eventually but they weren't - so the developers are busy essentially reinventing them from first principles to rectify the situation and ensure progress of the practice as a whole.

discuss

Our philosophy is to fight on two fronts. The first is to go talk to Congress and see if they can repeal this bad piece of legislation. Then on the practical matter, while this legislation is still out there, can we create a lightweight subscription radio kind of service offering that will make it easy for any of these stations to kind of either stay on or come back on the air with sustainable economics for themselves. The first is an industry-wide effort, and the second might make a difference in terms of what the economics are for Internet radio. It's just unfortunate than Congress did something that will temporarily stop the growth of the industry. I don't think it will eliminate or permanently stop that growth.

discuss

Our philosophy is to fight on two fronts. The first is to go talk to Congress and see if they can repeal this bad piece of legislation. Then on the practical matter, while this legislation is still out there, can we create a lightweight subscription radio kind of service offering that will make it easy for any of these stations to kind of either stay on or come back on the air with sustainable economics for themselves. The first is an industry-wide effort, and the second might make a difference in terms of what the economics are for Internet radio. It's just unfortunate than Congress did something that will temporarily stop the growth of the industry. I don't think it will eliminate or permanently stop that growth.

discuss

>recouped our investment on Frontier, and if fourteen

>years was the standard for copyright, it would have

>expired in April of this year.

Two points:

First, tough. Many texts and programmes have been written that did not make back their investment. The RIAA and MPAA claim that some 80% of new CDs and movies do not recover their investment. If one of Shakespeare's plays didn't make money, should it be removed from the public domain? How about all of van Gogh's paintings? Many items in other fields of endeavour are not guaranteed profit. You can always change your business plan.

Second, if you publish a renewed book (e.g. an encyclopaedia or dictionary, I can't think of a good expression for this) then the copyright date is based upon the year of that edition. Thus, the Catholic Encyclopaedia of 1908 is the most recent edition to enter the public domain. Similarly, the most recently freed edition of Webster's was published in 1913 and 1911 is the corresponding date for the 11th edition of Britannica.

While I believe that this is counterproductive, the law is what it is (c.f. Dickens).

However, you could apply similar logic to software. When was version 1.0 of Frontier published? April 1988? Okay, in April 2002, version 1.0 enters the public domain. When was version 2 released? Add fourteen years to get the year of its entrance into the public domain.

In addition to the copyright on the code, you also have a trademarque on the name, Frontier. Anybody can use the version 1.0 code but they cannot market it using your marque. I don't have a finished plan for this but there has to be an alternative to 75 years after you die (excepting of course that a corporation owns the code and corporations are like flesh-eating zombies. The earth keeps regurgitating them. When a given company fails, it's assets are acquired and the cycle begins anew.

Incidentally, I'd like to know just what the most recently published book is that has naturally entered the public domain. I suspect that no book published subsequent to Edwardian times is available. If a different mindset is not applied to preparing these laws, the same will be said in a hundred years.

discuss

If a hide-the-source mentality had prevailed among all writers of scientific software from the beginning of scientific computing, learning and research would have been much harder and slower.

discuss