Does free software taste great, or is open source less filling?

Short URL:


Which do you like best: the satisfying, rich taste of principle in free software? Or do you prefer the less morally filling and pragmatic goodness of open source? Do you wish people would stop endlessly rehashing the whole question of "free" versus "open source?" Or do you enjoy the chance to talk about goals and philosophy? As you might suspect, since I'm bringing it up...

...I like to talk about it. Is it too soon, though, following the lengthy debate stirred up by the recent "What should we call it?" FSM poll? Too dividing? Pointless? Do I ask a lot of questions?

Let's turn to our lodestar, Richard Stallman, for some answers. What's that, you say? "That RMS guy is the cause of all the trouble! He's a divider, not a uniter!" I disagree. Emphatically. In reading Richard's essays and listening to his speeches, I see and hear a very consistent and reasonable message: That freedom should come first.

Here's the canonical essay on the subject from the GNU philosophy rack: Why 'Free Software' is better than 'Open Source'. To me, it lays things out very clearly. Among other things, RMS writes:

We disagree on the basic principles, but agree more or less on the practical recommendations. So we can and do work together on many specific projects. We don't think of the Open Source movement as an enemy. The enemy is proprietary software.

That doesn't sound like someone who wants to drive a big wedge between the camps. Although first you might ask if we should say that there are separate camps. Isn't that part of the problem with that kind of language? Aren't we all talking about the same thing, and by constantly harping on the subject, isn't he creating a division where none should exist? I don't think so. Some people may use the terms interchangeably and they may understand the essential freedom that is at stake, but not all.

And I think there is a definite philosophical difference that is demonstrated by current events. There is danger that we might lose ground if people don't value freedom. And! It doesn't seem like harping to me when it is in support of principles. What are we trying to achieve? For those interested in freedom first, it makes sense to highlight the philosophy as much as possible.

I think it's absolutely necessary to be clear on the philosophy, as shown in the GPLv3 discussion. First, let's take a look at a question from a interview with RMS published earlier this year: Members of the open source camp tend to describe themselves using words like "pragmatic" and "realistic," while describing the free software camp as being driven by "ideology." Do you think that's a fair description?. Stallman: We in the free software movement are generally as pragmatic and realistic as anyone, in pursuit of our goals. Our methods have been rather effective, as [evidenced by] what we have achieved. Where we differ from the proponents of open source is in what those goals are. The open source viewpoint cites only practical-convenience goals, such as making software powerful and reliable. Our primary goals are freedom and community. We appreciate convenience too, of course, but we do not put that above freedom.. To win freedom for the long term, it is crucial for people to value freedom. If we could hand everyone free software for all jobs today, they would have freedom tomorrow (at least in the use of their computers). But would they still have freedom in five years? If they do not learn to value freedom, they will probably let it slip away. Thus, while the open source camp's strategy of getting corporations interested in releasing free software has contributed important free programs to our community, that approach alone cannot assure our freedom in the long term. Only talking about freedom can do that..

It's similar to what is covered in the GNU essay, but it highlights even more clearly the issue for me, and reinforces the desire to keep talking about this. If people don't value freedom, they won't stand up for it. And look what we have going on today with the GPLv3 debate: some people are upset about the anti-DRM clause and what effect it might have. (Disclaimer: I'm not following the debate closely. Instead I trust the Free Software Foundation to promote our interests in freedom.)

With the anti-DRM clause, we're seeing some heated discussions about how corporations will be affected and how they might respond. There are some who talk about the millions and billions of dollars that companies have riding on "open source" software and that there is too much at stake to be letting the "radical" FSF make all these "anti-commercial" decisions.

Some of these people are philosophically against all free software, but seeing that they're losing ground in the battle, have chosen to stake themselves to the old license and try to cause dissension by crying foul at the new provisions.

Other people may believe in free software, more or less, but they've taken the more practical view of the open source philosophy. They may appreciate the contributions of the GNU project, but are afraid its leader, RMS, will be too unpalatable to their new corporate masters. They may even agree that DRM is a bad thing, but they are more worried about losing business credibility and the support of corporations. Their desire for open source to prosper is stronger than their belief in freedom, and they may be more likely to want to try an appeasement strategy.

No way. Let's not do this. Let's not get off the freedom train so soon. The Free Software Foundation has been nurturing this vision along for so many years, and a little commercial success doesn't change the ultimate goal. I'm unable to articulate very well what I see as the problem here: that this vision of free software and a free society has grown steadily for so many years, but because we've gone through a spurt of rapid growth thanks to corporations realizing one aspect of the value of free software (its low cost and power), suddenly people want to say the FSF's approach is no longer relevant. That they're clinging to some outmoded and unrealistic ideal. Bah. Let's keep talking about it, and try to help people understand what the true goal is. One more time, where Mr. Stallman says it better than me:

Spreading the idea of freedom is a big job--it needs your help. That's why we stick to the term "free software'' in the GNU Project, so we can help do that job. If you feel that freedom and community are important for their own sake--not just for the convenience they bring--please join us in using the term "free software''.

And that's why I use and promote the term, Free Software.

Afterword I: On Corporations

Please don't misunderstand me. I'm not saying corporations are evil. Most of us earn our paychecks doing honest work for them. They are us. Doing business through corporations is largely how we've chosen to organize our economic activity. But by necessity, they often look at the present and the future very narrowly; unable to see the benefit of disruption. Free software is very disruptive and therefore threatening to established businesses. That's why I think the Free Software Foundation is right to discount business considerations in furthering the cause of freedom. And of course, despite the shrill cry of critics, the FSF can't force anyone to do anything, except to comply with the terms of the licenses they've already agreed to.

Afterword II: The Troublesome Mr. Hancock

Some final words about Terry and the troubles he has caused me with this post. First, I swear that I already was planning on using the "tastes great / less filling" title before I saw this comment. Now he'll probably sue me for some kind of idea infringement. Second, his thoughtful comments in the "naming names" poll caused me to miss last week's entry as I had to back up and think some more about it. (In the end, I just went with my gut anyway. But I valued his opinions.) Third, he just wrote this nice article about world peace and now here I am bringing up this potentially divisive subject again. Thanks for nothing, Terry! :-)

Back at the ranch

Several new articles can be had over at the Moving to Freedom web site ( There is a review of Linus's book, Just for Fun, that goes in to some of the same territory as the article you just read, and a post about intellectual property where I (briefly) connect the emerging networked age with man (and woman!) coming out of the caves.


Reusable with this attribution (including hyperlinks), and please note if modifications are made: Copyright © Scott Carpenter, 2006. Originally published in Free Software Magazine. Creative Commons Attribution-ShareAlike License (CC-BY-SA-2.5).



Terry Hancock's picture

So glad I could help you out. ;-)

Of course, the irony is that the people arguing to allow DRM (i.e. who object to the "corresponding source key" requirement in the GPLv3 and the "anti-TPM clause" in the CCPL3) are doing so for idealogical reasons. They argue that these provisions excessively restrict "user freedoms".

The pragmatic argument can be made too, but it's always an adjunct.

Fundamentally these are questions of "allowable copyleft". The FSF wants a stronger copyleft, because some groups have found a loophole that allows them to effectively defeat copyleft.

However, this conflicts with the one genuine difference between the "free software definition" and the "open source definition", which is that the open source definition permits a slightly stronger copyleft: a requirement that forces changes to be published. This, of course, was an earlier suggestion to defeat an earlier loophole used to defeat copyleft (the one used by Google: they don't have to publish their changes, because they don't "distribute" the software, they just let you use it on their site—forced publication would require those changes to be published anyway).

Trying to rectify that as a "principalled" argument requires some pretty serious logical gymnastics. The truth is that you just have different people adopting different nuanced positions in the big fuzzy gray space, ultimately in service to the same goal. There's no bright lines to cross here, just some hand-placed chalk marks upon a wide-open field that happen to disagree a little bit about exactly where the foul line is.

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

"Fundamentally these are questions of 'allowable copyleft'." - Fundamentally, it's about the freedom to run modified versions, which the intention of the GPL. If I can get the source code, but don't have the freedom to modify it and use it on my hardware, what good is it?

When I use Google, I am definitely not using their software-- I am using the output of their software. That's a very important distinction.

Being able to modify free software for my own personal use without being forced to redistribute my changes is a nice, business-friendly feature, not a bug. Besides, any provision to force redistribution in that case would be unenforceable, and thus pointless.

Dont' forget, the term "copyleft" is just a play on words. Every so-called copyleft license is based on, and worthless without, copyright law.

Terry Hancock's picture

"Fundamentally, it's about the freedom to run modified versions, which the intention of the GPL. If I can get the source code, but don't have the freedom to modify it and use it on my hardware, what good is it?"

Yep. That's the case for both FSD and OSD licenses.

"When I use Google, I am definitely not using their software-- I am using the output of their software. That's a very important distinction."

Sort of. It's a matter of point of view (by this definition, you aren't using your browser software either, you're just interacting with it. Why the distinction? Is it because you own the CPU? Are you using software you run on someone else's computer? If I install software on a remote sever and use it through SSH, am I "not using" it?)

But of course, it can be claimed to be a stronger ideological value to share that work back to the community that created it (as easily as not). This is just the principle of "copyleft" at work: sharing improvements back to the community that created the work (you can think of this as a kind of "payment" for the work).

My point though isn't that Google should be forced to share, but that some people think they should. And categorizing those people as "more pragmatic / less idealistic" than the people who think they shouldn't is a massive misrepresentation. Obviously both sides are making arguments based on both principled and pragmatic reasons.

"Being able to modify free software for my own personal use without being forced to redistribute my changes is a nice, business-friendly feature, not a bug. Besides, any provision to force redistribution in that case would be unenforceable, and thus pointless."

This argues for the difference being trivial, not an all-important ideological distinction. Which again, is my point. There is no all-important ideological distinction. It's a picky little detail. More importantly, the difference does not hinge on this mythical "pragmatic / ideological" dividing line: pragmatic and ideological arguments can be (and have been) made for both sides of the issue.

Bear in mind that most open source licenses do not have this requirement, and it is regarded as "bad practice" by both "free software" and "open source software" advocates. It's just a matter of a line being drawn to circumscribe allowed behavior. And the difference between "forced publication if used" and "forced distribution to anyone to whom the binary is distributed" is subtle. In the vast majority of cases, they amount to the same requirement (e.g. if I put the binary on my website, I need to put the source there too).

"Dont' forget, the term "copyleft" is just a play on words. Every so-called copyleft license is based on, and worthless without, copyright law."

By and large correct.

However, just FYI, this statement depends on jurisdiction and license. In some jurisdictions and for some licenses, the license is a contract ("license" not being a distinct legal concept in those jurisdictions, or the license being actually a EULA).

A contract or "license agreement" can remove "fair use"/"fair dealing" rights. I don't like the idea any more than you do, but it's the law. That's why the Creative Commons licenses explicitly grant "fair use/fair dealing" permission (because despite what the legally obsolete preamble to the GPL says, these rights can be taken away by the "license". IIRC, GPLv3 will be fixing this problem as one of its many uncontroversial changes).

Scott Carpenter's picture

Thanks for your comments, Terry. I really did appreciate your views in the other discussion, and I agree that some approaches can be less helpful. In this entry here I decided to take more of a straight advocacy approach because I really do believe in the idea of spreading the message and the debate provides that opportunity, even at the risk of alienating some people.

The question of how to treat code like Google's is an interesting one. Crosbie Fitch over at Digital Productions has some interesting things to say about this. I share his interests in privacy, but there are also all these gray area challenges. I don't want to force people to publish their private modifications, but as technology evolves and the methods of distributing software change, I'd hate for there to be ways around copyleft.

(EDIT: Durnit -- I hate it when I don't attach the reply to the comment.)

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

I been using GNU/Linux since 1994 when I found out about it in my university in alicante, Spain.
when I start it using it, it was not cause it worked better but for the ideas behind it.. now I see a lot of people don't care about the ideas but just that "it works", everyone to his own.. but well that said, remember that even if you dont agree with the FSF, some of us do and mostly we are the ones who started this and been using it for some time, if you dont agree with the ideals fine.. but don't step over them, just respect them.
my english is rusty but I hope to make a point here.
I vote for Free Software over anything.

Scott Carpenter's picture

Hi! Your English makes the point just fine. Thanks for commenting -- it's good to hear from people who share this view. For my own part, I believe strongly in the ideology behind free software, but now I need to get to work actually *using* it.

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

I have to say Thank you for your blog, our members former GNU/Linux Boston user group now named binaryfreedom( ) much apreciate this magazine and the articles and content you bring us, most of the times we print them and give them to our members in meetings and such.
Please keep the Free Software spirit alive.

Chris Fernandez
PS: BTW how can I become a member or help we will be glad to do so.

Scott Carpenter's picture

Thank you, Chris. I have a long way to go technically with using GNU/Linux and contributing to development of free projects as I'd like to do, but I do what I can with my writing to promote it. It's good to see web sites like yours -- do you have an RSS feed or plans to add one?

As far as helping out, I'm sure FSM would appreciate whatever you are able to offer, whether that is donations, articles, or comments on entries. I think comments are great and help to continue building the community here. (And promoting articles on social sites like digg that you think merit wider reading is great for sending more people our way :-)

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

I've actually always thought of Free Software as a subset of Open Source, since you can open the source code without hailing the four great freedoms of whatever. But it we must lump them together, one point of contention may be the fact that the Free Software nerds are a bit aggressive on how the word freedom is used. Stallman only pushes "freedom" if it fits into his 4 part vision. For people like me, Open Source contains other freedoms that "Free Software" does not. Like the freedom to actually do whatever you want with the code (like sell it and not give away your modifications). Or the freedom to include binary NVIDIA drivers that dont suck. Or the freedom to at least attempt to prevent your customers from screwing around with your hardware. Or the freedom use trademarked artwork. Or the freedom to use terms like "linux operating system" or "intellectual property" without getting an irrelevant and unwanted lecture by a language hippie on why we must talk different.

Some people like ideologies. Some people need a techy religion/school of thought. But some people just want cool stuff for free, preferably cool stuff that uses open protocols and patent-free formats and maybe even lets them pretend they are somehow fighting the man--a.k.a. Microsoft. Free, for these people, has nothing to do with the legendary fixing of printer drivers.

fhulleman's picture
Submitted by fhulleman on

By not giving away your modifications to the source code, you are denying your customers their freedom to modify the software.
Similarly, customers screwing around with your hardware may just be exercising their freedom to adapt it to their own needs.

With DRM one can also take away that freedom. With DRM, after the slightest modification, software may simply not work anymore.

The other point you seem to miss is that free software is not about money.

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

Or the freedom to at least attempt to prevent your customers from screwing around with your hardware.

Excuse me, but if I am your customer and purchase your hardware, then the bit of hardware I purchased belongs to me. It's now mine. I'll do what I damn well please with it - and if the seller suggests that I cannot screw around with my property - that is what is really f***ing annoying!

Terry Hancock's picture

The Free Software Foundation is the one responsible for this definition of "Open Source".

Your perception that Open Source doesn't include the four freedoms is in direct conflict with the "Open Source Definition" to precisely the same degree that the idea that "Free Software" can not have source available is (i.e. what the FSF calls "non-free" software).

Likewise, the illusion that Open Source includes non-copyleft license, but Free Software does not.

These are the result of misinformation deliberately spread by the Free Software Foundation. None of them has ever been advocated or claimed by the people using the term "Open Source Software".

Always pay attention to the sources of information.

On the whole, I respect the Free Software Foundation, but this deliberate attack campaign is a truly unworthy activity.

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

Agreed, his perception of open source is skewed. But he clearly is not the sort of person who's been paying any attention to anything the Free Software Foundation says, and his conception of them seems at least equally skewed. I think it's pretty repugnant for you to be smearing the FSF as responsible for his poor information.
If nothing else, he's basically claiming that the FSF would regard the BSD license as not Free software, which is clearly a smear on the FSF (and which I don't see you doing anything to dispel). They certainly say they prefer the GPL and copylefted licenses in general, but they've never said the BSD is nonfree.
Now, the original poster seems basically just not to know anything more about Free Software or Open Source than misleading slogans, presumably gathered from the mainstream media, so it's hard to blame him for his position. But you are apparently a knowledgeable person--you should be ashamed.

Terry Hancock's picture

One need look no further than the direct quotes from Richard Stallman appearing in this article to see examples which refute your position. But of course, there are many more on mailing lists, articles, and FSF blogs.

The FSF deliberately spreads several myths, the main one being that "open source software" is conceptually or ideologically different from "free software". This alone does enormous damage because it causes casual observers to flail about looking for what this distinction is: unsurprisingly they latch onto genuine community divisions such as "copyleft status", "GPL compatibility", "availability of source code", or "license to modify source code", but none of these distinctions actually applies.

More subtlely, they encourage particular misinterpretations. Broadly (and metaphorically wrongly) slamming "open source software" as "free beer" (I can attest to personal experience with this behavior by FSF partisans) which encourages conflation of the "availability of source" or "license to modify source" issues.

They talk about "ideological" or "ethical" bases in contrast to "pragmatic" issues, in a specific context of discussing the copyleft principle and/or the importance of what FSF calls the "four freedoms".

Both practices lend credence to the idea that the Open Source Definition somehow does not encourage copyleft or that it permits licenses which violate the "four freedoms".

These kinds of communications are what cause the erroneous press reporting that you refer to, and are thus the ultimate source of the misinformation that this poster is repeating.

Scott Carpenter's picture

I think one of the things that highlights the real difference between the terms is that people will promote things that are not free as "open source" and hope to gain the cachet of FOSS, even if they're only OSS. For example, there is an open source Jitterbit integration product that looks very interesting, and it is based on the MPL license, but as far as I can tell, it is not free.

I commented about it on an On Lamp article about the product and it drew a response from someone over at Jitterbit, but he didn't follow up on my reply. I'm still left with the impression that it is not free in a key way. But the source is open, so "open source" is certainly an accurate way to promote the product.

I think this is one of the things that will happen if people "freely" use an alternative term that doesn't suggest freedom and all that it implies.


Terry Hancock's picture

By FAR the more common error is to promote non-open, zero-cost software as "free software".

Both are misconstructions of the meaning given by the promoting organizations. "Open source" is abused far less than "free software", so which is the better term?

Of course, neither is a complete definition, so both are subject to some misinterpretation, which is why we get beasts like "free-licensed open-source software" or "FLOSS". This at least has the advantage of being sufficiently complete when spelled out as to leave no doubt and being sufficiently unique when abbreviated that you can search for a definition. OTOH, it does sound like something you should clean your teeth with.

WRT to Jitterbit, BTW, the JPL is a "free" license.

I re-read the FAQ and the response to your post. What the authors are trying to express (albeit clumsily) is that you cannot take the code proprietary if you make changes to the JPL-covered code. They are actually opening up the copyleft to allow for separate, proprietary-licensed plugins, called "Jitterpaks" which may be run-time linked with the Jitterbit code.

This makes the license similar to the LGPL in intent (it may be linked with code that is not JPL-licensed, but the code itself must conform to the license).

The authors' error was in equating "sell" to "make proprietary" (which is an altogether different conflation that neither "open source" nor "free software" makes clear!)

Anonymous visitor's picture
Submitted by Anonymous visitor (not verified) on

I like tools like synaptic package manager.

Synaptic comes pre-linked to 19,000+ (free) software packages.
You basically just check the packages you want and click apply.

Synaptic also does security updates.

blytheworks's picture

I have only been around GNU/Linux for a few years and I really never thought of Free and Open Source Software as being two different things. To me, it seems like Open Source Software is like a car:

Someone creates it (new)
Someone acquires (in the case of cars - buys) & uses it
There are possible modifications: new paint, better stereo system, new tires, engine overhaul (you get the idea) but it is still a car
After modifications, that person sells it (or gives it away, it is up to them anyway because they own that version now)
The new user repeats the modification steps to meet their own desires
The car is then traded, sold, given away or discarded after many repeats of the above cycle.

The creator isn't concerned with the cycle as long as they get credit for the original work. A Ford is still a Ford, a Porsche is still a Porsche, etc. No amount of modification short of utter destruction, can change that.

Free (as in "free beer") allows everyone to enter the above cycle without shelling out a dime to do so (no purchase necessary) AND it supports and encourages the users' right to participate fully in the above cycle ("freedom").

To me, Free and Open Source co-exist peacefully in the advancement of Software design and implementation. Freedom 1 - The freedom to study how the program works, and adapt it to your needs (in which case, the source code is required for this) ensures that the program is Open Source. I think all too often we have confused "free" with "free beer" instead of using it in the context of "free speech".

If a program is modified and those modifications are sold along with the original source code, then that does not stop the next user (there really are no 'end' users where open source softwre is concerned) from using or further modifying the code to meet their needs. All it did was cause that user to spend money (a little or a lot) but in no way does that violate any of the four freedoms as long as ALL the source code is available WITH the purchase.

By definition and design "Free" software is "Open Source" software.

Author information

Scott Carpenter's picture


Scott Carpenter has been lurking around the fringe of the free software movement since 1998 and in 2006 started a more concentrated effort to "move to freedom." (Chronicled at the Moving to Freedom blog:

He has worked as a professional software developer/analyst since 1997, currently in enterprise application integration.

(Views expressed here and at are strictly his own and do not represent those of his employer. Nor of miscellaneous associates including friends and family. Nor of his dog. It's possible they're representative of his cats' opinions, but unlikely. Void where prohibited. Local sales tax applies.)