ESR Writes About O'Reilly and FSF Differences

dopplex writes: "Over here at Linux Today, Eric S. Raymond has written an amusing piece in which A.) He analyzes the way in which we use the word freedom, B.) Examines the point of view of both O'Reilly and the FSF on 'freedom' and C.) Coins the term 'flerbage,' which I hereby suggest be put into immediate use, just because it's a really cool word." It's cheesy but it is a good way for people to understand the difference between Open Source and Free Software. (Oh, and I figured I'd just mention that I'll never use that F word since I think its stupid)

Comments

Linux Today...

[NewbieSpaz]

Shouldn't that be "GNU/LINUX Today"...?

;)

Re:Linux Today...

[Waffle Iron]

For those who want to call it Linux, I'd just suggest this: try running your favorite distro after subtracting all of the GNU system. Have fun.

Try running your favorite distro after subtracting all of the NON-GNU software (Apache, Perl, Python, vim, etc..). Not much fun either; yet none of those developers are clamoring to get their names prefixed to the system.

Excellent

[Reality Master 101]

He has neatly summarized my problem with Stallman, the FSF and the GPL. The big problem with Stallman is that he believes that users should have power over programmers, which I find absurd. The programmers are the creator of the work, and thus should have the "freedom" (there's that word again) to choose how their work is used.

It seems like the height of tyranny for an ungrateful rabble of users to in essence say, "Thanks for creating this product that we find useful. However, that's not damn good enough. It's not enough for us to have the freedom whether to use your product or not, you should be required to develop your software according to OUR requirements."

I hate to borrow from Libertarian philosophy, but a right is not a right if you require coercion of another person.

I also look forward to hearing Stallman's response to whether they would be in favor of laws enforcing software "freedom" GPL-style.

Re:Excellent

[Uruk]

It seems like the height of tyranny for an ungrateful rabble of users to in essence say, "Thanks for creating this product that we find useful. However, that's not damn good enough

The fundamental divide I think is that the FSF doesn't think software should have owners. It sounds like you're getting pissed at users acting that way because the developer should own and control the software - or as you put it "the freedom to choose how their work is used".

GNU believes that programs are generally useful technical information. Most people think that patenting math formulas is bullshit. I don't really see much of a difference, since in both cases they're generally useful technical information.

I hate to borrow from Libertarian philosophy, but a right is not a right if you require coercion of another person

I agree - but put a different spin on it. Who are you to say that I cannot cooperate with my neighbor by sharing generally useful technical information with them? Who are you to throw me in jail because I copied something I bought to a CD? Who are you to call me a 'pirate' (and equate me with someone who robs, murders and rapes ships) when I'm helping my friends and coworkers out? From my perspective, distributing non-free software is coercion - it's coercing users *not* to help their friends. It's coercing them to avoid using and spreading generally useful technical information in many circumstances.

Life is all about gathering, spreading, and exploiting technically useful information. If you disagree with me, disagree with me that computer programs are technically useful information. But without the ability to use, spread, and gather good technically useful information, we're never going to evolve and get off of this planet.

Re:Excellent

[rgmoore]

I agree - but put a different spin on it. Who are you to say that I cannot cooperate with my neighbor by sharing generally useful technical information with them? Who are you to throw me in jail because I copied something I bought to a CD?

Hear, Hear! This is exactly the point that RMS gets and ESR misses- which is surprising given that ESR is a libertarian. All software licenses are inherently coercive; they use the power of the State through the means of copyright to restrict the rights of the user. The difference between a Free Software license and a proprietary license is that a Free Software license uses that power for the benefit of all (by restricting obnoxious behavior) while a proprietary license uses it for the sole benefit of the writer (by restricting socially beneficial uses like sharing). And sadly, the mere existence of Free Software does not defang the power of proprietary licenses. Big software houses like Microsoft can still engage in serious legal harrassment of just about any computer using business even if they don't actually use any Microsoft software.

Stallman

[sourcehunter]

Yet Kuhn and Stallman say they don't like this world. It appears that they would prefer a world in which people who write software cannot choose the proprietary licenses that Kuhn and Stallman dislike.

"Stallman" is starting to sound a little too much like "Stalin" for my tastes.

It is my property, I can and will choose a license that fits MY needs as a developer.

And like it or not, proprietary licenses DO foster innovation on a corporate scale. I'm not saying free licenses don't foster innovation - but they do so at a different level - the community level... and companies (and individuals) have to get paid. I gotta keep food on my table SOMEHOW! is RMS/FSF going to send me a check every month? Or is that the government's job? This is sounding more and more like communism/socialism.

Economics

[underwhelm]

The author of the article happly ignores one monkeywrench: economics.

The rules of economics aren't "laws," just predictors of how people will behave to further their interests. Unfortunately, the predictions of economics imply a necessary reduction of our "flerbage" in the case of a monopoly.

Specifically, ESR carries the libertarian mantle blind to the "Network Effect" of proprietary software that makes the Windows Monopoly so powerful. By flouting standards Microsoft makes their proprietary software more valuable because there is no adequate substitute in the marketplace. They have an interest in breaking campatibility with all but their own products or those they sanction in order to maintain and strengthen their monopoly (ignoring government regulation or anti-trust lawsuits that modify their behavior).

The Network Effect means that my flerbage is certainly decreased in the case of proprietary software. Microsoft operates in their best interest to break compatibility, and in order to do business I am forced to purchase their product whenever they do. My choice appears "flerb," but is really "double-plus-unflerb" because of economics. Sure, I'm free to use linux, it's just that I have to expend by precious time and resources circumventing Microsoft's artifical barriers to compatibility. That's not flerb at all.

I'm not sure if this is endemic to the libertarian view, but from my standpoint being forced to do something by the market is just as bad as being forced to do it by the government. Sure, the market can't put me in "prison" with "guns" but it sure feels like it when I don't have any *actual* choice.

Liberty and proprietary software

[Arker]

Sorry, ESR, but the flerbage nonsense is stupid. There's already a perfectly good word for that, it's called Liberty.

Proprietary software may be able to coexist with Liberty, but it will certainly take some work. The whole trend since the beginning of closed software has been for it to threaten Liberty, to eliminate it bit by bit, as quickly as it's producers believe they can get away with. The pace is ever increasing, and every time a draconian measure is abandoned because of public backlash... it returns with less fanfare a little later on.

And no, I'm not in favour of criminalising the offering of software under proprietary licenses. I don't believe the FSF is either - in that respect this article is simply an audacious straw man.

When the government uses my tax money to buy proprietary software, when it publishes public information which I am legally entitled to access, but in a form that is useless to me unless I pay microsoft their monopoly rent, my liberty is under assault. When they grant artificial monopoly privileges (software patents are one excellent example of this) and enforce them via law, this is no better than enforcing one particular license on developers would be.

Under no circumstances should tax money ever be spent on proprietary software. Under no circumstances should public information be published locked in a proprietary format. Instead of trying to break down monopolies with anti-trust law, the government should quit creating them and nurturing them in the first place.

Why ESR doesn't understand the FSF point of view

[phaze3000]

Essentially, what ESR is saying is that in his opinion, everyone should have the right to publish software under whatever license they want, proprietary or otherwise.

What the FSF would argue is, to quote Star Trek, 'The needs of the many outweigh the needs of the few'. Thus, whilst stopping people from publishing software under a proprietary license does, to some extent reduce the freedom of that induvidual (or corporate entity, etc.), it does so only in order to stop the rest of the world from having their freedom violated, namely the freedom to alter said software as one wishes.

Personally, I'm inclined to agree with the FSF; I believe it is more important for everyone to have freedom, even if it does reduce individual freedom.

Herein lies the fundamental difference between the 'Free Software' movement and the 'Open Source' movement, and the reason RMS and co get so uppity when the two terms are used interchangably.

Case studies

[BlackStar]

Very interesting points from ESR, although I think RMS isn't *that* fanatical. Well, ok maybe. ANYWAYS.

What about looking at some cases in real life? Proprietary licenses. Say.... Windows. Very successful. Likley due to cost, marketing and standardization of a chaotic platform long in the past. Solved a lot of problems with proprietary platforms only running 1-2 applications you needed, so you almost bought one machine per application in some situations. Seems OK for the time. Now, however, with viable alternatives, there are some things like open sourcing (NOT GPL) that may be useful if the modifications could be redistributed, but MSFT still owned the rights to the parts they feel they need to. (Asbestos enabled)

Why not GPL? Enter point number 2. BSD/Mozilla/Extend and contribute like licenses. The SCSL from SUN. Specifically, Java. If this thing was GPL'd off the bat, it would be another fragmented, proprietary implementation of a screwed up standard left in the past not unlike CDE, or even C++ in it's early life. With SUN owning a brand, and enforcing a standard that they don't actually unilaterally define, it's a workable, reliable, and standardized open platform.

Those two cases in point, one must ask WHY the GPL seems to have such problems creating the defining third case study where GPL is the only thing that worked. Well, maybe it has. Let's take Linux. If it was proprietary, it would likely be as big as CP/M about now. If it was SCSL, the buy-in by the GPL crowd would be nill, (err... null, err.. nevermind) and the corporate adoptions to the benefit of the community wouldn't have occurred.

So, we've got three broad and incomplete categories of license, and three broad and incompletely analyzed case studies showing success in each case, and why in those particular cases that license modality was the correct choice for the goals.

So bascially, I would side with Tim on the side of choice, and promote the said flerbage as the yardstick. Evolution finds optimal solutions through excessive choice. Seems to have worked out fairly well. Odd that it still resulted in the occasional individual that opposes the primary mechanism that gave rise to them. I savour the irony.

I'd love to hear why the power of choice would be a bad thing, even if you choose a proprietary license. Try and write a cheat-resistant multiplayer game with open source on both client and server, and see just how far you get before the cheats make the game unplayable except among friends. Lots of papers and discussions on that as well.

And don't raise the "web of trust" and such there RMS and cadre. Defintion of trust on that level would have removed the success of the GPL in the case of Linux, as there could be code in there that trusted people back-doored, but no one has bothered to review. Trust is perception, and perception is in it's very nature incomplete.

Sometimes you just gotta say no when someone wants your recipe. :-)

Respond with thought or not at all if you please.

Libertarians should hate ESR for this

[abe ferlman]

I hate to borrow from Libertarian philosophy, but a right is not a right if you require coercion of another person.

Copyright is nothing but a government granted monopoly. Wake up! If the government does not coerce people *not* to copy your stuff, then copyright doesn't exist.

Libertarianism requires government non-interference into the marketplace. Copyright is a direct intervention into the marketplace. Despite the fact that its intentions are good, it does not work, and it causes a whole heap of coercion along the way.

Property rights make some sense when they are attached to items that are scarce. Information, however, is not naturally scarce (although the ability to create it may be). Would libertarian ethics allow other sorts of interventions into the marketplace to guarantee innovation? For instance, what if it was found that better music would be created if only people with masters degrees in composition (or licensed students) were allowed to create music. Think of how much crappy music wouldn't get made if you needed 6 years of school and a license before you could strum an A chord! Is this a legitimate type of coercion? Think!

Bryguy

Re:Libertarians should hate ESR for this

[Jerf]

'Information, however, is not naturally scarce (although the ability to create it may be).'

I finally figured out why that statement has bothered me for so long. It's this: "Information" in the abstract is not scarce. But guess what? "Physical goods" in the abstract are also not scarce!

Useful information, information you want, in other words, "information" in the concrete, is scarce and will remain so for the forseeable future. In some isolated catagories, there may be a glut of "information" (like mail clients), but in the concrete, there does not exist so much "music I like" in the world that it can be said to be a commodity. Unless all you are interested in is stuff in the past, your current desires will always not have enough information to be met.

Water isn't scarce in many places. There's as much of it as you could possibly want. But unless all you want is water, you should anticipate needing to pay somebody for your food, clothing, and shelter. Isolated examples of non-scarcity do not prove the general case of non-scarcity.

You might say I'm missing the point, because once information is created, it can be infinitely copied. And I say in return that you (the reader) would be missing the point. In the old economy, the cost of distribution may have been the primary constraining factor, but the cost of production is still non-zero. The digital economy may make the cost of production the defining factor, but it does little to affect that cost. (Useful information, virtually by definition, requires significant effort to create. If it did not, you would not come to me for the information, you'd simply (re-)create it yourself.)

Even if the physical goods could be distributed for free, you'd still need to pay for production and creation. Even if the costs of production could be reduced to zero, you'd still need to pay for creation... unless everything you wanted was already designed, a situation not likely to happen for a very long time.

I think when you see arguments like "information isn't a scarce resource", you're seeing a confounding of cost of distribution vs. cost of production. The reality is, information is still costly to produce and you can't just wave your hands around and wish it away. While there is historical proof that some software can be developed in the Open Source manner, some catagories of software don't fare so well. Nor is all "information" like software (a massive oversimplification), and for those categories, and for those categories, there is little to no evidence that "novels" or "blueprints" are in a "scarcity-free" world.

In the abstract, copyright recognizes this scarcity by granting the author certain limited rights. In the concrete, most of the problem lies in the absurd nature of those rights. I stand against the DMCA, I stand against the absurdly long copyright terms holders have nowadays, but in the abstract, copyright still works. In fact, the guiding principles of copyright are standing amazingly well against the "onslaught" of technology, even if many of the details aren't faring so well.

Until such time as you can effectively wave a magic wand (i.e., a super-human intelligence) and recieve the answer to any question you can ask ("How can I cure my hippocampus cancer without removing large chunks of my brain and without killing me with an allergic reaction, since I'm allergic to the dyes used in scanning technologies?" That's information that's decidedly scarce and isn't going to not be anytime soon!), information will continue to not be free, and economic models and ethical arguments predicated on those models will continue to not be grounded in reality, with all that that implies.

That said, one might make the case that software is a special case. However, I submit that if that were true, it's one of those things that would be obvious to all concerned. Personally, when I write software which didn't exist before, I find the effort to be non-zero, and one way or another, society needs to support my ability to do that, or I won't do it any more. As it turns out, I'm not getting paid directly. What that effectively boils down to is that I'm 'paying myself' to do it (If I didn't get money from some source, I would not write this software), but that still doesn't mean the effort was zero, and arguments that assume zero monetary cost -> zero effort -> zero scarcity to create are doomed to inaccuracy and failure. Overall, I'd say software is no more immune to the costs of production then any other kind of information.

ESR misses the point

[Mr. Slippery]

While I respect ESR, in this case he could not more perfectly fail to grasp the point.

Proprietary licenses, whereby a state-designated owner can use state power to declare some string of bits "property" and do nasty things to you if you copy them, are an infringement of "flerbage". (Or "freedom", if you prefer.)

Maximum "flerbage" would be the absense of copyright - not passing new restrictions on proprietary licences, but rather removing the exisitng restrictions that make proprietary licences possible.

(I'm not - for the present - arguing for or against such a change. Just arguing that outlawing certain uses of photocopiers, tape recorders, computers, etcetera, is not moving in the direction of maximum "flerbage".)

I'm disappointed that a self-described anarchist doesn't understand the difference.

Raymond is a libertarian

[prizog]

This is more standard libertarian rhetoric. Consider: If I "choose" to work a low-income job, and therefore "choose" to live in a high-crime area, men with guns will occasionally forcibly divest me of my property. Or, as happened to a friend of mine, they won't have guns - they'll just have lead pipes, and instead of just taking my property, they'll beat the shit out of me, putting me in the hospital for weeks and *then* take my wallet.

That's what I call freedom.

Sure, I could choose to live on Monaco, were there's virtually no crime as much as I could choose to build a rocket ship and live on Mars - that is, sure in an ideal fantasy world, but not in reality.

Raymond is comparing his ideal world, in which I would have the flerbage not to use Microsoft products, to the real world in which I have the choice of significantly fewer jobs if I make that choice. If I were a secretary, I might have the "choice" of using Microsoft products or finding a new profession, probably at lower wages.

Anyway, even if he weren't an adherent to a utopian philosophy, it's not the case that being disallowed from releasing proprietary software reduces my flerbage.

Here's what Raymond says:
"If I walk up to someone and offer them the same
proprietary license that I did before the law was passed, police may come to my house to drag me off to jail, or kill me if I resist arrest. My flerbage has seriously decreased."

And here's the definition of flerbage:
"I have the condition of flerbage when I can behave in the confidence that nobody will take my life, my physical property, or my time without my consent."

Well, obviously that doesn't include acts which are illegal - you can't expect to kill someone and get away with it. So, if you say "My flerbage is decreased because I can't break the law", well, tough. Or you might say "the law is bad because it decreases my flerbage," well, that's what all laws do - but we pass them to increase the sum flerbage of each person more than it decreases your flerbage. So, you're back in the same situation where flerbage means freedom. Oops.

Here's an interesting commentary on libertarianism:
http://world.std.com/~mhuben/faq.html

Flerbage, Schmurbage

[bwt]

I have the condition of flerbage when I can behave in the confidence that nobody will take my life, my physical property, or my time without my consent.

Larry Lessig once admonished ESR for what Lessig described as ESR's advocacy of "warmed over Ayn Rand". I would describe flerbage as "a moldy and rotting 'Ayn Rand'-like substance".

The problem with flerbage, is the "without my consent" part. Essentially it claims an individual right to veto laws that protect more than life and personal property. Ironically, given ESR's position, Copyright violations don't qualify as life, property, time deprivations so I don't know why or if ESR would support the government depriving software pirates of flerbage when they have not deprived him of it.

Like most radical libertarianism, flerbage simply misunderstands the role of public policy in recognizing property and trade. Neither property nor contracts exist outside of a context of government because both involve a concept of government force being used to redress transgressions. You do not have property or a contract unless the public, through it's agent the governement agrees to enforce it. This involves the public use of force that requires resources and the public has the right and the power to expend its resources in a way that is consistent with the heirarchy of legal principles established by a Constitution and due process of law thereunder, which includes some majoritarian lawmaking processes that predictably may occassionally result in rules that fail your personal "without my consent" test.

Enough people consent to the process to call it consensus, and you are never offered freedom from attack by the delegated power of the people if you don't consent to a particular rule and ignore it. Nobody really cares if you "never signed no steenkin social contract". If you don't consent to the process then you've declared anarchy and rebellion, so don't come whining when bad things happen to you such as your flerbage being violated. The Declaration of Independence states the principle that if the government becomes destructive to the will of the people that the people may overthrow it. Civil disobediance, peaceful or violent, is sometimes the morally correct thing to do, but nobody ever said it doesn't come with great peril precisely because it is outside of the rule of law.

Re:Why ESR doesn't understand the FSF point of vie

[Russ Nelson]

I believe it is more important for everyone to have freedom, even if it does reduce individual freedom.

Spoken like a true fascist. No, really, that's what they used to say. Go read about it if you don't believe me.
-russ

Freedom is not having to bark on command.

[Futurepower(tm)]

To me, the article seems to confuse the main issue. An example of the main issue, for me, is that if you use, or program for, Microsoft Windows, you are effectively a dog on Bill Gates' leash. Bill can do whatever he wants with you. He can refuse to support new hardware. He can decide that your copy of Windows is obsolete. He can decide that your old hardware is obsolete. He can, under the DMCA, remotely disable your entire OS. He can support U.S. spy agencies in a hidden way. He can avoid fixing bugs because he wants to save some so that you will be interested in buying a new release.

The GPL is a sophisticated way of avoiding being under the control of a dictator. That's where the word "freedom" applies.