A New Model for Software Innovation

An anonymous reader writes "In this whitepaper published at LinuxDevices.com, Matt Asay (former Linux naysayer-turned-disciple) analyzes the GPL, picking apart what it means (and does not mean) for users, and whether it is enforceable. Assay also details how its terms inhibit and foster innovation, and why we should care. In this next generation of software, those who understand 'copyleft' licenses like the GPL will have the upper-hand, and will be best positioned to take on closed-source shops like Microsoft. Assay wrote this paper while attending Stanford Law School, where he studied the the GNU General Public License under Professor Larry Lessig." A thoughtful piece that answers - as well as they can be answered - a lot of the questions about the GPL that we get for Ask Slashdot, as well as examining the economics of it. Good reading for anyone developing or selling software.

Innovation

From the article link (not the whitepaper):
"In turn, this most 'anti-IP' of licenses is arguably doing more to foster innovation than patents or copyrights ever have."

Patents and copyrights are about making money, NOT fostering innovation.

Re:Innovation

After you loose your patent, everyone will be able to use your method and build upon it.

That's the way it's supposed to work, but the reality is that after you lose your patent, it's useless to everyone else - at least where most software patents are concerned. The state of the art will have progressed too far, and there will thus be no real quid-pro-quo for the public.

Bruce

Re:Innovation

From the US constitution:

The Congress shall have power ... 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;

The original intent of patents and copyright was about fostering innovation.

Not that the corruption of the idea we have today is about that tho.

Re:Innovation

Is Mozilla more innovative than Internet Explorer or Opera?

Is TCP/IP and the Internet more innovative than xns? or netbios (which came much later)?

Is Sendmail / SMTP more innovative than DEC's messaging system from the early '80s, or IBM's? (damned If I can remember their names).

Is NFS more innovative than RFS?? (actually, I'd say that they were about even -- but NFS won out .. possibly because it was more open than RFS).

______

When you think about it: VHS beat out Beta -- not because it was better but because it was more 'open' (just about anybody could produce a VHS machine, but only Sony could produce Beta).

Similarly, many people ascribed DOS's ascendency over the obviously superior MAC to openness -- the fact that anybody could (and did) build a DOS box, and put whatever they wanted into it.

Openness does drive innovation.
With two products of equivalent inovativeness, the one that is more open tends to drive more collateral innovation -- and is thus most likely to thrive.

Part of the reason for patents and Copyrights was the fear that, without them, big-money own-it-alls would usurp the works of truly innovative (usually small) authors and scientists -- make all of the money off of their work and, this remove any incentive for them to make their work known.

Copyright and Patents have now swung to the other end of the pendulum. They've become and end in, and of, themselves. They've been 'strengthened' to the point where we'll soon be 'celebrating' a Century of Intellectual wilderness -- A century during which substantially nothing has been put into copyright that has subsequently flowed into the public domain -- as envisioned by the writers of the first amendment.

Much of what was created in the early 1900s will be lost into this wilderness. Only the most famous (and, in some cases, the most mundane) works will survive. Much of the rest is caught in a legal limbo. Nobody knows who 'owns' it -- thus, nobody can obtain the right to copy it. By the time it's legal to redistribute it, the originals will be incapable of being duplicated.

Consider that recordings of the technical conversations with the Apollo astronauts were only barely recoverable -- and this after less than 30 years in storage. Imagine what another 100+ years of languishing would do to less famous works -- Those who would like to preserve them risk criminalizing themselves in the process.

Had the original works been protected under current Copyright rules, Disny's Snow White would never have been made. Similarly, most of Shakespeares works are known to have had their stories lifted from other authors.

We build on the works of others. Copyright and Patent laws are useful to the extent to which they lubricate the path from creation to public availability. The moment that these laws block that path, they become repugnant to their moral and constitutional underpinning.

Re:Innovation

Do you think that is all right? Do you think programming a computer is easier or harder than walking down a runway?

"Which is harder" has nothing to do with anything. The only question is which is more valuable in monetary terms. The model isn't getting paid to walk, the model is getting paid to use her physical assets to promote the fashion designer's product. If the fashion design didn't think it was worth it, she wouldn't get paid. Now how many people have the physical assets to do that, versus how many people can crank out code? That's why she makes $50K and you don't. Supply and demand rules all.

And by the way, if you're going to complain that she's "just" using her beauty that she was lucky enough to be born with, and turn up your nose about how superior you are, be sure and realize that you are just using a particular mental talent that you were lucky enough to be born with.

Software Law

What I am really curious about is if the field of Software Law is growing, or if it just falls under the more general patent law? I am interested in both the areas of software development and engineering (being one myself) and law.

Worldwide monopolies?

From the whitepaper this time:

"It just replaces Microsoft, a monopoly
by one, with a global development effort, a monopoly by all, as it were (which, of course, is really no
monopoly at all)."

Right. It's a polypoly.

Re:Worldwide monopolies?

polyopoly " - or perhaps "poliopoly", but that sounds a bit too much like someone who is infected by multiple polio viruses.

But -- as was pointed out: Jonas Salk placed his vaxine into the public domain -- and thus saved millions of lives and made polio almost extinct (with the exception of 'defensive' bioweapons stores).

From one story about Salk [pbs.org]:

The success of the vaccination effort won Jonas Salk unsought fame. The March of Dimes, hoping to boost publicity and donations to fund vaccination programs, lionized Salk to the point of offending his colleagues. He had applied the findings of others in a successful bid to prevent disease. Other researchers and doctors grumbled that he hadn't found anything new; he had just applied what was there. But the timing of his successful vaccine at the peak of polio's devastation made the public blind to that.

Had todays IP laws been in place back then, much of the work that Salk depended on would have probably been patented. He might not have been able to create the Smallpox vaccine, and many of us here today would have thus been dead.

GPL Revision

The paper refers to a proposed revision of the GPL. A link to it is on the page.

This GPL revision seems like a completely new GPL. It's language is much simpler, and devoid of legalese. Does anyone else worry about such a complete re-write being enforcable?

Re:GPL Revision

Do you worry about the old GPL being enforcable? It's never been challenged in a court of law. It's never been tested to see if it stands up. RMS acknowledges that.

Dear AC,

No, I don't worry at all. If the GPL tends not to be valid, the people who have challenged it are then left with the default in copyright law, which is all rights reserved. Even if they win, they lose. Would a court say "yes, you can use and redistribut GPL code without any obligation to anyone else"? If they did, that interpretation would apply to code under other licenses as well - including that stuff in the boxes from Microsoft. Forget about that happening.

Bruce

A different perspective, perhaps?

While I welcome any insight or debate into the GPL and CopyLefting, frankly, I'm not getting anything new here. And the fact that he studied under Lawrence Lessig is going to imply some bias on his part, considering Lessig's positions. Frankly, other than speculation and name calling, there's not much real analysis out there on the issue of licensing in open source. What I'd like to see is some expert opposing viewpoints, starting a real debate on the legality of the issue. I've yet to see an article from the other side of the opinion fence regarding the GPL's legality, and this is strange, considering Microsoft's attack against the license, and the fear among many in the Open Source movement that it's unenforcable. How about it, Slashdot? How about some legal opinions from people other than Lessig or his students, pro AND con? That would be an interesting debate.

psych

Whatever the GPL says or omits, being 'best-positioned to take on closed source shops like msft' will not be possible without understanding why people will pay $550 for a word processor and spreadsheet. And dismissing that behavior as just 'lusers being stupid' isn't an answer.
When people will pay $550 for an easy, intuitive word processor that comes with restrictions (Office), rather than a more primitive, ugly, yet free and somewhat as functional word processor (Emacs or other unix WP), there's something begging to be explained. And the licening details are of secondary importance.

lol

former Linux naysayer-turned-disciple?

That's funny... How come I don't read about former Microsoft naysayer-turned-disciple??? Anyone???

Sounds great on paper

I've read lots of papers like this, going back to Eric Raymond's rants. And you know, it sounds great on paper. It really does. And I love the idea of the GPL. But in practice, we're just not seeing all the innovation that you would expect. The classic examples are always cited--Perl, gcc, Emacs, Apache, the Linux kernel--but for a model that's supposed to be so superior, the success stories are fewer than you'd expect.

One striking example is that the GPLed games for Linux always tend to variations of Tetris, Boulder Dash, Missile Command, etc. You would expect the innovative games to be coming out for Linux: no pressure from marketing, free development tools, big community. But the games with spark, like The Sims and Grand Theft Auto 3 are coming from elsewhere.

Re:Sounds great on paper

The gamers are playing games instead of coding :-)

The lure of Free Software development mostly has to do with useful software that can be developed incrementaly. It doesn't necessarily make much sense for games. The fact that both KDE and GNOME produced great desktops in 4 years is pretty impressive, given what folks have spent on CDE and so-on.

Bruce

You are looking in the wrong places

I've read lots of papers like this, going back to Eric Raymond's rants. And you know, it sounds great on paper. It really does. And I love the idea of the GPL...ut for a model that's supposed to be so superior, the success stories are fewer than you'd expect.

That is because you are looking in the wrong places for your "success stories." You are wondering why some GPLed mapping program hasn't unseated Microsoft's Streetsmart, or why program X under the GPL so resembles commercial program Y (though often it is program Y which mimicks the GPLed program X ... much of OS X's eye candy can be traced back to the first program to ever use images in window decorations and borders: Rastorman's enlightenment, but I digress).

What you are ignoring are the tens of thousands of small firms, like my own, like the dozen's a friend of mine consults for, and thousands of others around the world, who have used free software (GNU/Linux in particular) to build their infrastructure, their inhouse software, and their business out from under the everpresent heal of Microsoft, Sun, or Apple (though the latter has become a kinder master of late than the other two, it nevertheless remains a master). The success stories you are ignoring range from the traders I work for, who earn a sizable chunk of money thanks in no small part to GNU/Linux's superiority over their competitor's infrastructrue, to Pengiun Airlines, to Wallmart, to BMW, to the German government, the Chinese Government, the Brazilian government, and others too numerous to list, all of whome are able to run their IT infrastructure on their own terms, within their own budgets, and free from the coerced upgrades and chronic, forced obsolescence of their software and their hardware at the hands of their operating system vendor, as is chronically the case for anyone subject to Microsoft, Sun Microsystems, and yes, even kinder, gentler Apple.

That is where you will find the innovation and the flourishing of the technology more than anywhere else, amongst the users which have been liberated by the Freedoms expounded upon by the Free Software Foundation, people like myself and many, many others who have been able to flourish even during these difficult times, thanks primarilly to the GPL and the user's freedoms which it helps to guarentee.

Freedoms that allow us to run our businesses and earn a living, without being subject to the whims and deprivations of a copyright cartel or a convicted monopolist.

Re:Sounds great on paper

But in practice, we're just not seeing all the innovation that you would expect.

What innovation are you expecting? Are you looking to be swept off your feet by the next handsome OS that you see in the elevator?

Linux and GNOME are a unique twist on the old idea of a UNIX desktop. Linux has matured a great deal over the years to rival old giants like Microsoft, and GNOME has become a very good desktop environment for many people. While they aren't earth-shatteringly innovative, they provide some relief from the stagnant mainstream desktops out there.

Then there are the countless tools that made my life livable over the years: GCC, gnuplot, ghostscript, Emacs, LaTeX, ispell, XFree86, Mozilla, and OpenOffice.org for my desktop, and sendmail, apache, and BIND on the servers. There are many many many more tools that I won't take the time to list out, but that doesn't mean they aren't there doing their part.

Many Open Source software projects are not truly innovative but provide best-of-breed implementations of standard protocols. Nearly all of the standard Internet protocols have been implemented for Linux and the BSDs. Ones like Apache steal the show, and ones like the BSD TCP/IP stack are "borrowed" by smart and successful people like Microsoft.

Perhaps the biggest contribution of Open Source is properly moving things into the "public domain" after no longer being commercially worthwhile or after becoming commodities. The protocol implementations I mentioned certainly fall into this category. For example, I seriously think the idea of a commercial web browser is obselete, and ones like Internet Explorer really serve to hold back progress rather than help it.

Another way that Open Source contributes in ways that companies can't is to provide uncorrupted software. By "corrupted" I mean: integrated DRM, proprietary file formats, proprietary communications protocols, unfair EULAs, and any other blatant company or government-driven attempt to limit what computers can do. The frequent advocacy by Open Source enthusiasts for open file formats is one way to drive innovation that really can be earth-shattering (imagine a world without .doc!).

So, I'm not sure what you're looking for, but I'm pretty convinced that there is more innovation occuring than we can shake a stick at. It isn't always suprising, revolutionary, or even obvious, but to say it isn't happening is to be in denial.

Innovation and interest

There's a quote of Lincoln on one side of the patent office building in Washington DC. It says something about the patent system coupling innovation with interest, which I guess means providing a reward for innovation. That's where the money comes in.

But of course there are any number of situations where that reward is not the optimal means of promoting innovation, or isn't even appropriate - one example is the case of public-funded research resulting in private patents, another would be when the monopoly grants not merely methods but entire categories of business, as with today's broadly-interpreted business method patents.

Bruce

IP

regarding the ip/anti-ip content, this is a very old argument. As old as the american copyright system.

Ben Franklin: I'll make my inventions free so humanity can reap the rewards.
Tommy Jefferson: We'll make a patent system so that profiting from IP will promote invention.

Because there's not overwhelming evidence that one or the other positions is superior, it usually becomes faith-based reasoning--the worst kind.

A new model for software innovation

I see this as the perfect chance to plug openchallenge [openchallenge.org] and maybe get good comments on how to improve the concept.

Mace Moneta [67.80.55.197] summarized the idea well:
" get those that have the skills needed to tackle a problem together with those that have the problem, and solve it as Open Source. The OpenChallenge web site [openchallenge.org] lets you submit problems; those looking for a project to work on can browse the list of requests for something that they find interesting. "

Stallman must be obsessive-compulsive

Either that or a robot. On page 17 he goes off on the "GNU/Linux" thing without provocation. Unless you consider the word "Linux" provocation. Anyhow, I just had to laugh. I guess I should have seen it coming but the FSF guy replied to to question in such a normal way and RMS had been replying so normally that it caught me off-guard.

esay

Assay also details how its terms ...

It's a great esay written by Assay.

proposed revision of the GPL

An interesting streamlining of the GPL. However he takes out all the "social statements" which may or may not be a good thing.

A few comments:

You can use GNU GPL Software for any purpose. You can modify, copy and distribute GNU GPL Software, including derivatives, for commercial or non-commercial purposes.

In return, you must agree:

No no, you can use the Software for any purpose. Period. If you re-distribute the software, THEN you must agree to the terms.

The license should emphasize that it only covers "copying, distribution and modification". No one should feeled compelled to accept the GPL, just because they are using the software. This also helps dispel the stupid notion that you don't have a right to use software you buy or download unless someone grants it to you.

This, in my opinion, is one of the greatest thing about the GPL. It lies dormant and only applies to you if you distribute copies. You don't need to read it, you don't need to "click-through" it. You only need to care when distributing copies, which you aren't allowed to do to begin with. In fact, if I were writing the GPL, I'd put that up at the very beginning (after a warranty disclaimer if that's necessary): "you don't need to read this unless distributing copies" or something similar.

if you distribute GNU GPL Software, you include two things: a copy of this license and a ready means of obtaining the source code for the Software.

Define "ready means"..

That if you distribute a derivative of GNU GPL Software, you include a notice explaining the changes you made.

(Nitpick) why isn't this included with the previous requirement? What constitutes a "notice explaining the change"? A diff? A paragraph? Is this important?

The GPL is wordy, partly because it spells out some of this stuff in more detail (for instance, see item #3 in it). The "revised GPL" should also be specific whenever possible.

Anyway, the GPL does need to be streamlined and clarified to a certain extent, and perhaps the FSF can get some ideas from this revision. For instance, cutting out the Preamble of the GPL entirely might be a good idea, and replace it with a link to the FSF philosophies.

Re:proposed revision of the GPL

You left out the part about linking. If we are to clear up that part, I'd much prefer to make it clear that some forms of dynamic linking and server-izing (as with CORBA) do create derivative works. I think the key concept of a derivative work in software has nothing to do with linking, and everything to do with the span of the thread of control in a computer program. This can indeed happen across physical system boundaries. The boundary for the copyright should probably be at an API that is explicitly designated as an interface to separate works, such as the kernel system call interface. Not just any API.

Thanks

Bruce

His next paper...

...will be on the Slashdot effect.

IANAL++

This guy deserves a new acronym

IANALBISL

I Am Not A Lawyer, But I Study Law.

-Jon

Taking best advantage of IP and software

I have just skimmed his paper (interesting stuff to me started on page 31), so this is tentative: I have been thinking of GPLing my software technology for semantic extraction and natural language processing.

My problem is that I have invested a little under 2 years of work on my NLP tools, and although I only make about $5K a year selling them, that money does help pay the bills and I believe the long term market for my software is excellent.

I have received many great bug fixes and code improvements for my Open Source projects and even more great help with material for my free web books, but these projects have not cost me nearly two years of work.

I guess that the real question is if companies (and perhaps individuals) who use GPLed software to make money in their own closed commercial products are honest about paying a fair license fee to get a non-GPL license to use my work. I do receive many generous donations for my free web books, so I do trust in many people's generousity and fairness.

An alternative to using the GPL is to license software for free non-commercial use and then hope that commercial use users act fairly.

I think that many developers would very much like to have their work freely used by people who make no money off of their work and at the same time collect reasonable payments from anyone making money on one's work. Tough problem - please share any good ideas :-)

In any case, it is tough for me to make a decision here.

-Mark

Is linking closed code to GPL library "fair use"?

Interesting paper. Since the GPL only grants permission to do things that would otherwise be forbidden under copyright law, I wonder if linking proprietary code to a GPL library could be defended as "fair use"? Ignoring the GPL license. Much the same as "reading it" could be considered fair use of a copyrighted book.

If so, would that mean that linking to any software library, whether proprietary or not and whether validly licensed or not could also be defended as fair use?

I am no lawyer, natch.

The Microsoft Factor

One item only partly addressed in the document has been the increasingly burdensome requirements of Microsoft licenses (product keys in manuals, product keys stuck to the back of PCs - good idea MS!, online product activation and shotgun EULAs giving MS the "right" to alter your system). In combination with the increased cost of licensing, this is prodding all Windows consumers to re-evaluate their options. Not to mention MS's famed security...

Pronouncing 'Asay'

This guy is actually someone I know from my San Jose days. Really good person, very smart, and I'm enjoying reading his paper. But anyway, his name is pronounced like "Ace-ee". If it were spelled "Acey" you might not need help. But there it is.

Making money on hardware versus software

What is interesting about this paper is that most of its examples, at the end, of companies adopting and using Linux are people who are *PRIMARILY* hardware companies and view the hardware as the product and software as a cost.

Why not use Linux rather than pay VxWorks or some other company, when you are primarily a box company. Lindows is in the same boat. They sell a PC and then make money selling software to the end users. Some of the software may even be free, but end users don't necessarily know about site like SourceForge. IBM likes Linux because it is a box and services company, and you'll certainly need a lot of services to start using Linux in a corporate environment. Their UNIX OS isn't a great success and is tied to their proprietary hardware. Their mainframe OS isn't growing. Their database software and Notes have no real Linux challengers (see below).

What the author does not address is the issue of making money from SOFTWARE in the Linux world. Oracle will make money because they have a proprietary lock on the data and programs that are written to run with their database. Over time, MySQL or HSQL might replace them, but remember that the cost of switching is huge and the risk high for a large corporation. Same is true of DB2 and Notes.

And that is the issue. For new development or new systems, Linux and its various middleware is free and greatly reduces the cost of deployment and potentially the cost of development in terms of tools and desktop costs, not project costs. What GPL doesn't do is provide a software product industry. In the end we will be left with the only way to make money as a developer being to work for a corporation writing dull accounting programs, because we'll have given away our efforts to destroy the revenue streams of the companies that provided us with a living.

Was it Lenin or Marx who said, "When the time comes to hang the capitalists, they will fight for the right to sell us the rope"?

The Author Does Not Understand the GPL!

He's confused about licenses versus copyright, e.g:

The GPL's language, then, tries to swallow every piece of software that incorporates a GPL program or portions of it. However, it probably cannot do so under the copyright law it purports to follow. The definition of a derivative work proves highly controversial in the software world, and may not stand up in court. [Page 15.]

This has nothing to do with the GPL!

If I create a derivative work from a copyrighted piece of GPL software, I can release it under my own terms without accepting the GPL. I then probably get sued by the COPYRIGHT holder, and the courts then decide if this was a derivative work, a new creation, a copyright infridgement, or whatever.

Alternatively, I can accept the license (the GPL,) in which case copyright law no longer applies: I have freely entered into a contract that gives me certain benefits and takes away certain others (e.g. my right to claim copyright trumps the license I just agreed to.) If I now start doing evil like violating the GPL, I get sued for violating a LICENSE, not copyright law.

The GPL does not "build on" copyright law - it's a license that I can agreed to or not. If I don't agree, copyright is the operative thing, if I do agree, the license if the operative thing!

Please don't read it

Normally we have trouble with folks posting here without even bothering to read the article in question. But in this case I'd argue that would be a good thing.

I read through about the first half of it, but his explanations of the GPL are so confused and horribly unsound that attempting to follow it was starting to muddle my brain.

In the first paragraph it struck me as odd that he claims that he was recently convinced of the GPL's innovation harming qualities, but now is convinced that the complete opposite is true. Executive summary: I know I sounded convinced yesterday. But I was actually full of shit then, and just didn't realize it. But today I really do know what I'm talking about. Honestly

However, after reading his embarassing attempts at logic, its perfectly understandable how something like that could happen to him. With logic that muddled, tomorrow he might be convinced the GPL has manged to retroactively kill the dinosarus.

My suggestion is that anyone who doesn't have a real good handle on the GPL already avoid this whitepaper like the plague. Folks who fully understand the GPL should take the Monty-Python "World's Funniest Joke" approach, and split up into teams with each person analyzing a small bit of it seperately. That way perhaps no one person will read enough of it to become fatally confused.

I wouldn't exactly trust this guy's conclusions...

Let's see ... on the page of the paper, the author asserts that the public and corporate perception of Linux has undergone a 180 degree reversal in the last year. (Simply not true ... a year, even two or three years ago, Linux was very much on most of the mentioned companies' radar. He's trying to pass off his change of mind by claiming that it was the industry's) He also states that "Linux and its global network of developers have long sought nothing less than 'total world domination.'" (Again, untrue; 'Linux,' as an operating system, doesn't have any desires. Linus Torvalds, the head of the "global network of developers," has repeatedly stated that Linux is a hobby, not a grand political scheme. The only one interested in world domination is RMS, however, in that case the article shouldn't be about Linux at all; it should be about GNU and the FSF.)

I'd take any paper that starts that way with a grain of salt.

how 'bout a little puppet show

<RMS MODE ON>

Any part of Mr. Asay's article where he implies the GPL is bad or could use some work is merely due to his failing to comprehend or understand the meaning and scope of what the GPL in its entirety.

<RMS MODE OFF>

Isn't that pretty much what we're going to hear in a few days? I'll thank the moderators for at least attempting to have something resembling a sense of humor.

P.S. No offense meant to RMS or his devout apostles.

An FSF Response to Asay's article

We at FSF have spent a few minutes reviewing Mr. Asay's paper. We are glad to see that someone who once took a clear anti-GPL stance now claims an "admiration" for it. We noticed the paper contained a number of email quotes from our founder, Richard Stallman, and our General Counsel, Eben Moglen. The quotes appear to be taken from various email conversations that Asay conducted with Moglen and Stallman separately. Quoting email out of context is always tricky, and the conflict and confusion with our official statements regarding the GNU GPL are apparent rather than real.

We do, from time to time, have occasion to clarify our views on the GNU GPL, and we work hard to be responsive to questions from the community on them. This does serve to generate a lot of email traffic concerning the Foundation's opinions on GPL, and out of context pieces of the whole of such correspondence can be confusing. When we get pieces together that are particular good and clear, we post them on our philosophy page [gnu.org], as we did with Eben Moglen's essay, Enforcing the GNU GPL [gnu.org], and in our GNU GPL FAQ [gnu.org].

As for the GNU GPL, version 3, we had indeed long ago seen Mr. Harris' proposal, but we did not feel it was representative of our plans for next version of the license. The copyright of the so-called draft of GPL Version 3 distributed alongside the article, thus, is incorrectly attributed to the Free Software Foundation. We do not hold copyright on it, nor did we have a hand in drafting it.

Efforts for development of real GNU GPL, Version 3 continue within the Foundation. We of course expect to have extensive public discussion before any draft is finalized as the official new version of the license. In March 2002, We approved the Affero GPL [gnu.org], which includes a draft of one major provision we are considering for the next version of the GNU GPL. Of course, when we do announce that a draft copy of actual "GNU GPL, Version 3" is available, it will certainly come directly from the Foundation.

Bradley M. Kuhn
Executive Director, Free Software Foundation

Still Not Clued

Remarkably, after all this time, he's still only partially clued. As might be expected when somebody was totally wrong and announces the fact, when he explains his new understanding there's little more reason to take him seriously than before.

From such little details as getting the names of the licenses wrong -- leaving the word GNU out of the title of his paper, and elaborating the LGPL as the "Library" GPL rather than its correct name, the Lesser GPL -- to totally missing the true competitive value of the GPL to businesses, as well as the very simple economic motivation for developers to participate, he still has a lot to learn.

If there's anything new or interesting here, I didn't find it. He's articulate, though, and evidently open to reason after he has been pummeled sufficiently, so he might even get it right on the third try. In any case he's no worse than ESR.

Leave GNU/Linux & GPL alone!

Why don't you people leave GNU/Linux & GPL software alone??

There is so much hypocrisy in your message.

If you do not like GPL software, please do not use it. Use something else, use MS Windows, BSD, whatever you like.

You would like to use GPL software but you do not like the GPL because it will not let you profit from the work of others available to you for free. How is that?

tooth fairy

That reputation argument is a crock. It sounds like some kind of monetary surrogate in the minds of people who can't organize reality in any other way.

I participate in open source because I still believe in the tooth fairly. I put a tooth with a cavity under my pillow at night, I wake up the next morning and the cavity is gone.

Once upon a time it was possible to have incredible wealth, yet still have bad teeth. You don't brush your teeth to improve your reputation. Perhaps those who bleach their teeth are in it for personal profit, but the rest of us just want good teeth _without_ spending a fortune on dental work. Is it really that hard to understand?

As a recent Linux adopter...

I completly lost the author when I read this quote:

Very few applications exist in the open-source world, at least those that would be useful to the average user.

Completely false! Especially for "average user". Tons of applications to suit all my need were installed automatically when I installed Mandrake. And I could easily complete by looking on the web (gimp, OpenOffice).

The average user needs a browser. Konqueror, or Mozilla, or derivatives will do.
The average user need a mail reader. Mozilla, Kmail, will do.
The average user needs a word processor / spreadsheet. KOffice, or OpenOffice, will do.
The average user needs a MP3 player, and instant messaging client, .... There is some on linux already.

What else do you REALLY need? The author certainly didn't look much before making this affirmation. There are plenty of applications for Linux. And more are coming our way every day.

The author also says that the GPL is an incetive killer for developper. For COMMERCIAL applications only. The way I see it, there is a lot more programming done with no commercial intentions. I like to code; and I always though of releasing some of my stuff as shareware or freeware. GPL is even better! Someone might improve it! What I think the author doesn't get is that the GPL intends to change the buisness model of software makers. And that is a good idea.

Holes in his proposed GPL...

• AS IS isn't capitalized. The UCC requires for it to be.
• It doesn't define "source code".
• It doesn't require a distributor of a derivative work to distribute that derivative work under the terms of the GPL unless that distributor is the same person who created the derivative work.
• You only have to include a notice describing the changes you made, not the changes made by others. That effectively makes the requirement meaningless.
• A distributor cannot impose license restrictions, but the creator of the derivative work can.
• Clause 5 is equivalent to a restriction against removing the "do not remove" sticker on a mattress. I should be allowed to do whatever I want with my copy of the software.
• I'm sure there are others.

GNU advocates stealing Apache credit

Yet another GPL/GNU advocate steals credit from The Apache Software Foundation while claiming that Apache's success supports the GPL. Ironic how the same movement that is trying to advocate the name GNU Linux is shamelessly citing Apache's success as its own. Of course the Apache Software License is incompatible with the GPL so that code from the Apache and GNU projects cannot be mixed whatsoever, but that doesn't stop the GNU advocates from citing Apache as support for the GPL.

How many more of these articles are we going to see that make bizarre claims like Apache is GPL software [newsforge.com]? (Note the passage containing "today thousands of GNU GPL software packages are already available including operating systems (like the famous Linux OS), office packages, Internet servers (like Apache running on 30% of Internet Servers".

Coming Up Next

How whip cream fosters weight loss, how unfettered immigration promotes security, and how adultry prevents AIDS.

Sure there are people who eat all they want, live next door to strangers from Iran, screw all the time and don't end up as fat disease bags screaming in a pool of their own blood at a bombed-out supermarket. However, the smart money says "why chance it?".

I hate having to download PDF Files!

Go to http://www.adobe.com/products/acrobat/access_simpl e_form.html

and enter the URL of the paper

http://www.linuxdevices.com/files/misc/asay-pape r. pdf .

*ONE* good point

His whitepaper brings up one good point, in my opinion. That in the free software world, consumers can develop software for other consumers rather than businesses developing software for profit. His example of Mozilla giving the user control over privacy with cookies, etc. is a good one. A better one would be the popup-disabling support that AOL/Netscape (tried to) remove from the commercial version of the browser.

The rest of his paper, IMNSHO, is rubbish.

Almost but not completely a load of dingo's kidney

One footnote stood out:
...there is no such thing as intellectual property in the Open Source world, since all code must be open to the view of the user, who is free to modify the code...

Mr Asay has one or two good points. These are methodically buried in a series of categorical statements like the one above. He proffers statements that are clearly inaccurate or directly false. He has made his research, but evidently not understood what it was about.

Consider this sentence, "The GPL turns all of this on its head. Not only does the GPL remove the property right (or, at least, the ability to profit from it), the GPL skews the definition of "common good" for which copyright and patents provide. That is, the GPL focuses on developers, rather than the end user." Aha. And everybody who ever installed RedHat is a developer? I think not. End users benefit hugely from using Free Software, in that they never have to be nervous about EvilEmpirism. For the man on the street, Open Source is like playing with an open deck of cards. You might not understand the game but at least the on-lookers can see if your opponent is cheating.

I am sorry to say, that Matt Asay has shredded his credibiliy with me. He seems intelligent enough but a paper can only have so many blatant blunders before the reader loses respect completely.

Self-Determined Time Limits

Hi, all.

It occurred to me, why is the government deciding the time limits for copyrights and patents? Since each application requires a document unto itself, wouldn't it make sense for there to be a space to assign your own time limit, and might this provide a way to encourage short time limits on tech patents?

For example, in patenting software, let's say software patents require a time limit (to be imposed by the author), after which, the software source will be forced into the public domain under an OS license.

As example, say you have two creators, that create competing products at the same time.

Creator A chooses 5 years because he knows he can make a lot of money with the software immediately (there's a strong demand for the software RIGHT NOW), but at the same time wants the software to gain market share and beat out other competing products in the long run. He advertised that it will be OSS in five years, and his patent is registered (along with his code) for five years, after which the patent office will release his code on the internet.

Creator B chooses lifetime because he thinks, no matter what, it's my idea, and if someone uses it, I should be reimbursed.

Now, since A can advertise "Here's this great product, and, btw, after five years it's OSS and you can do with it what you want" whereas B can only say "Here's this great product".

Would companies be more inclined to buy things that will eventually be free to them? Will other companies license this technology because they know they only have to pay for it for five years? Will companies wait for five years, possibly losing a competitive advantage, in order to avoid paying?

How might these ideas play out?

Would people embrace technologies with shorter time limits, because it would imply eventual cheaper adoption later on? Or, would everyone just make unreasonably long lifespans? Would creators use this idea to work against publishers who would just milk a product dry even after the creator no longer works for the company?

Just curious if any work has been done on that idea?

M.

Re:barf!

If anyone posted this garbage here, they'd be moderated to -1 in a hurry.

Are you under the impression that slashdot moderation has anything to do with the quality of a post?

Re:barf!

``Free source software has a long history of playing catch-up to proprietary software.''

Just curious: why do you disagree with this statement? As others have noted, the majority of open source software is described as "an open source version of [insert proprietary software name here]". Even GNU was meant to be a free replacement for Unix.

Re:It's in PDF format!

How would you see the cool graphs? ASCII art?

Re:barf!

You apparently didn't read the papers. Both of them.

The author makes it very clear in his second paper that he has completely changed his opinion on the GPL as a viable licensing model.

He spends a lot of time sorting out both the myths and truths surrounding the GPL's effect on closed-source companies, OSS developers, and consumers/end users. Especially interesting was his analysis of how dynamic linking to GPL code effects proprietary software development/licensing. In short, there was a lot more to the documents than the topic sentences you grabbed from the first couple of pages.

As an aside and a personal opinion, the first of your quotations is actually a truth, not garbage. OSS has been playing catch up for quite some time, as it had a much later start. Granted, it's come a lot farther in a lot less time, but that doesn't change history.

Anyways, if this sounds like a rant, it probably is. After reading all 47 pages, the last thing I wanted to see was a blatent troll cluttering the forums.

Re:WTF is a whitepaper?

Look here [dictionary.com].

Re:WTF is a whitepaper?

We've also used the term greenpaper meaning a whitepaper that isn't ready yet (ie, typically just incoherent ramblings like /.).

Re:You don't HAVE to give away your GPL'd apps

Actually, everything he said was 99% true. Not eloquently stated but defiantly true.

If you create a piece of software that is based on another GPL'd software then your software must be covered by the original GPL license. You are not under any obligation to release the binaries or source. However, if you do release the binaries you must make the source available upon request.

This effectively prevents someone from making changes to a GPL'd product and then sell the changes for a high price. They could do it but the buyer would have the right to distribute the code for free thus killing the seller's market.

The only part that I disagree with is that because the programmer is being paid to product the product his employer would have the exclusive right to initially release the code. So unless the employer released to code to the programmer, the programmer couldn't release it to the general public.

If I am incorrect, perhaps you could point me to the paragraph in the GPL license that supports your position.

Re:yay editors!

That type of error is quite hard to spot. When it comes on a line break. More so. (-dot)