Comment by mpk | Hacker News Reader

Comment by mpk | Hacker News Reader

Parent Full thread

> This makes them sound like mortal enemies.

That makes it sounds more dramatic than it is. Stallman is an idealist and attaches moral value to the licensing of code. Torvalds is an engineer and just wants to ship code. His license of choice matches his model for doing this.

Basically Stallman is ideologically opposed to the existence of proprietary software and Torvalds is fine with a mixture of free/open software mixed with proprietary software.

The two of them are often mentioned together because the largest results of their labours (the GNU userland and the linux kernel) are combined to form 'linux based distributions', generally just referred to as 'Linux' (or 'GNU/Linux' as Stallman insists, because he justifiably wants some more public recognition of the impact of his project even if that just isn't the way things work).

Considering their differences and the way in which their works are combined it shouldn't be that surprising that there's some friction and snarky comments that spill out from time to time.

I can't find consistency in Stallman's idealism. It seems that he's against proprietary software, yet is completely fine with legal protection for intellectual property rights. The GPL is a copyright. The GPL, just like any copyright, limits certain freedoms of the users of the software. Does Stallman have some idealogical problem with the public domain?

I'm not speaking for Stallman of GNU here, this is pure conjecture based on too many years of following these discussions. The "problem" with public domain software, (and BSD/MIT) style licenses, is people can take that code, make changes, and seal it back up into proprietary binaries where users lose access to the source code. The ideal world that the Richard Stallman character wants is one where if you're using something that was built with code, you have access to that code, and can't be stopped from using that code. Everything that people in the GNU camp do is towards that goal. Their choice to assert a GPL copyright (or copyleft, if you want to get really hippie) is a pragmatic one, because they recognize that the power dynamics of business lead companies to stop sharing most of what they're working on.

It always amazes me how frequently this question comes up. I cannot decide whether it is genuinely something lots of people fail to understand, or whether a certain type of person just enjoys pointing out that the GPL isn't 'free' according to their own personal definition.

But, to answer your question, Stallman is interested in the freedom of the end user to modify the code and redistribute the modifications to that code freely. He has been almost freakishly consistent on this topic for the last 200 years, restating that stance in a multitude of articles, interviews and talks.

Anything that ultimately results in code being 'locked down' and unavailable to the end user to modify and then redistribute, he has a problem with. This includes the public domain.

I probably shouldn't have used the word "consistency" in my accusation; I didn't mean to imply that Stallman has wavered or flip flopped over the years. I suppose that his position is quite logical and internally consistent if you specify his main interest/goal as "the freedom of the end user to modify the code and redistribute the modifications to that code freely."

To me, however, it seems like the goal of GNU/Free Software is usually conveyed as maximizing freedom in regards to software. They themselves stress that word "free," and point out that it's "free as in speech, not as in beer" without emphasizing that they really just mean "free for the end user to see and modify the source."

In my opinion (and yes, under my personal definition of "free") the GPL does not maximize freedom and instead is exactly like other copyrights except that the restrictions it places on software are perhaps different than other copyrights. I personally do not think that software should be covered by "property" protection laws, because I think the analogy of software as property is fatally flawed and leads to serious contradictions and ambiguities. Also, on the more practical level, I do not think that intellectual property protection for software is necessary for a thriving and innovative software industry.

There are many different kinds of freedom, it is true, I dont recall RMS ever talking about maximizing freedom in any kind of general sense, and I suspect it is because to do so would be nonsense. The phrase 'maximize freedom' is meaningless without context, too many freedoms are contradictory.

It is freedom as in speech, because it is the freedom for users of code to take that code and use it to express their own ideas, and to redistribute that expression to their friends and anyone else who may be interested. In that context, code is literally a form of speech, and the software an expression of an idea. Under the freedoms provided by something like the GPL I can take someone else's idea, change it to suit myself and redistribute that change.

The GPL clearly does not attempt to maximize the freedom of developers or distributors who wish to deny end users the freedoms that the GPL provides.

Again, this is ground that seems to be covered over and over. It is hard for me to believe that someone like you, who is clearly intelligent, has clearly thought about the subject and discussed it previously, has truly never had it explained to them, or has entirely failed to understand the distinction. I hope this has helped a little.

My above comment (the grandparent of this comment) was a clarification of my earlier comment. The disagreement between RMS and myself indeed comes down to my personal definition of "freedom" and my personal views on intellectual property protection. I have indeed heard your argument before, and I suppose I can concede that RMS's philosophy is internally logical given his definition/usage of "freedom."

Still, I find the GPL to be more restrictive and less free than the public domain, and therefore less desirable. I take issue with the way RMS and GNU uses the words "free" and "freedom" because I see their philosophy not as a principled desire for freedom, but rather merely as a desire for certain restrictions on freedom (granted, a different set of restrictions than proprietary software licenses).

I think that "The Free Software Movement" is an unfortunate title. Not only do they have to constantly remind average readers of their definition of "free as in speech, not as in beer," they also should be clarifying their definition of "free" for people like myself who believe that freedom is an utmost concern. Perhaps a better though less catchy title for their movements would be "The Publicly Available Source Code Movement."

The GPL clearly does provide restrictions on the use of software released under it.

'people like myself who believe that freedom is an utmost concern'

which freedoms matter to you, and for whom? without context your statement is pretty much content free.

IME almost everyone believes that freedom is an utmost concern, even Brutalog, the Despotic Dictator Of Zoogtzargle believed that he was fighting the good fight for freedom.

The interesting differences seem to lie in the context under which they define the term.

Really, it's about either maximizing freedom for users or maximizing freedom for developers. The GPL is for the former, while the WTFPL is for the latter.

Regarding practicality, RMS claims that the GPL forces companies to open-source software that would otherwise by proprietary, thus opening up the information contained within to everyone, for their use and improvement. In reality, I see more cases where the company instead recreates the software using a proprietary (or, if we're lucky, permissive open-source) license (PySide and Clang come to mind).

The public domain is actually a remarkably poorly defined IP space. Most countries don't actually have a legal mechanism to recognize PD. Places that have a recognized PD (the U.S. for example) are the exception.

Here's an example of this

http://arstechnica.com/old/content/2007/10/european-copyrigh...

The GPL, or CC (or some other license) sidestep this issue by working within the nearly universally accepted legal framework of copyright vs. the Public Domain. In Stallman's view, because PD cannot be guaranteed to use the force of law to maintain his view of freedom, it's unworkable for his purposes.

Reply on news.ycombinator.com