Is Stallman Wrong About Proprietary Software?
April 25, 2018•2,891 words
"With software there are only two possibilities: either the users control the program or the program controls the users. If the program controls the users, and the developer controls the program, then the program is an instrument of unjust power." - Richard Stallman
This quote is likely the most famous from Richard Stallman and encompasses well the primary argument made for 'free software,' which is more of a political philosophy regarding the treatment of software users rather than a philosophy regarding its development. Stallman draws a line between software that is 'open source' and 'free as in freedom' or 'libre,' meaning software that people are free to use, study, change and redistribute as they see fit. This distinction makes sense, because open source software could have its code be publicly auditible, but not allow the modification, private use or redistribution of it. Stallman fears that open source could be used purely as a marketing bullet point with little consideration for the ethical principles behind free software. While getting behind 'open source' software is a good start, it does not automatically entail the further ethical principles behind 'free/libre' software.
Stallman has helped build a community based around these ideals regarding free software, and I find myself in agreement with his ethical goals in principle. Although, the argument Stallman and many free software advocates make does not sit well with me. This is mostly due to the the incredibly hard and fast claims that are made regarding the nature of freedom and proprietary software.
The argument Stallman makes rests on the idea that proprietary software necessarily leads to the abuse of its users' freedoms. Though what is meant by this? Is it true? Of course, if software is proprietary it necessarily means that users cannot audit the code and there are clear limtis to its distribution. But, couldn't we have proprietary software that does respect the freedom of the user? Additionally, what sort of assumptions are we making about the user's desires when using such software?
The definition of 'freedom' here is very particular, and seems to assume that freedom necessarily requires complete control. I believe this is where the argument breaks down, as such control is likely too extreme a standard to reasonably operate with and claim as a necessity. There is a lot to be said here philosophically, in terms of freedom of the will, but lets keep this reasonably focused.
A Strict Definition of Freedom
What do we mean when we say 'freedom'?
By normative standards, it means we are able to and are free to act in accordance with our desires, and that we are free from both force or coercion when we act. The question is then, are we forced or coerced by proprietary software when we use it? Does it force us to use it in particular ways?
We tread into some heavily opinionated territory asking these sorts of questions, because the answers will likely vary hugely depending on who you ask. So, I want to illustrate the perspectives of two likely responses to the questions above; (1) from the perspective of an ordinary person, and (2), from the perspective of a technically savvy person or maybe more specifically a computer programmer.
First the ordinary person, who we can assume has a low technical understanding of computer hardware and software. This is your everyday person who uses technology because it makes their lives easier, they think little about the software and hardware they use, so long as it does its job. This person does not have the ability to modify software and knows little of how it actually works, just that it helps them do the things they would like to achieve.
Let's say this person uses a proprietary word processing program made and distributed by a large private corporation. This program does everything that they need it to do to get their desired work done; it does not force the user to use features they do not desire to have and is not missing any features that this person needs. Not much needs to be said, there is clearly no threat to freedom here from their perspective and the software itself does not force the user to use it (though maybe you could argue that their occupation or circumstances requires them, but this raises further questions and issues outside the scope of this discussion). This person would respond with a shrug, they feel the program adequately respects their freedoms, they can use the program to the extent they require and that it is not manipulating them in the way in which they may use it (e.g. refusing to save certain documents, restricting access to basic program features after already paying for a license, etc.).
Next is the tech-savvy user/programmer, who has a high degree of technical knowledge of computer hardware and software. This person also uses technology because it is a useful tool to achieve desired goals, although they have a deeper relationship with technology and thus require more fine-grain control over it. They have the ability to modify and/or redistribute software and know how to most effectively use it to their advantage.
This problem exists primarily for the technically savvy and not the ordinary user, but could we argue that the ordinary user only has the illusion of freedom? Not really, since their needs are in fact being met by the program entirely, it seems difficult to argue that the program itself is exerting any sort of coercion or force on the user regarding its use. Saying otherwise is like arguing that a hammer forces/coerces users using it for hammering because it can't be made into a screwdriver. You could argue that some proprietary software coerces users into using it (e.g. 'hacking' their brains, invading privacy), but once again this doesn't seem like a necessary feature of proprietary software.
Freedom in this case is too strictly defined, and involves being able to do more than what any user could potentially desire of a program and shedding any potential restrictions that could even theoretically come about.
Is it more likely that harm could be done to users through proprietary software? Objectively I believe the answer is yes, though, is it necessarily true that this harm is done? Absolutely not, and to make this claim is to make a circular argument about proprietary software. (i.e. proprietary software is bad because it is proprietary)
A More Charitable Example - Proprietary Software Liscences (DRM)
It is possible that I may be focusing too much on simple user-program interactions and reducing the argument to absurd levels, so lets refocus on the strongest aspect of the position; software licenses.
One of the more effective arguments for free software comes from a short article from Stallman called, The Right to Read and gives a non-technical example of why free software is so important.
The central dilemma of this example includes a person named Dan, who wants to lend his computer to his friend, Lissa. The example is set in what seems to be an authoritarian near-future type of scenario, where a 'software protection authority' is able to discern whether someone other than the owner of a book reads it. Dan wants to lend Lissa his computer to help her complete her midterm, but is torn because it is likely that she may read the books off of his computer, which she does not own the license for and may get him in trouble. There are many ways around this dilemma, but they are unsavoury options which include; Dan getting expelled for sharing his computer password, getting his friend to 'crack' the DRM to allow her to freely read it (and risk jail time) or letting her use the laptop and go to jail for infringing on the license terms.
This short story is more of an argument against Digital Restrictions Management or 'DRM' for short. Although, DRM is commonly included in proprietary software and ensures that only the person who paid to use the software or media is the one who uses it. Assuming that one day DRM is taken this seriously, and users could risk jail time for copyright infringement, this is a very serious concern. Also, we have seen many times how companies themselves exert control over the type of media people can consume on their software platforms, in some cases directly removing disagreeable files from users' devices.
Why Not Both?
I often find the answer to such issues are best found in compromises, so it may be best to try and find some middle-ground here.
It would appear we would have no problem at all if, for example, companies like Amazon never exercised (or never had) their power to remove media off of user devices. Additionally, deeper beneath Stallman's story there is the idea that people should be able to freely distribute and modify the copyrighted intellectual property of others, but this is a separate issue about the nature of private property and capitalism, which needs a separate discussion. (It further seems like Stallman may have further issue with people being able to own things, and in order to support free software in totality, it may involve a commitment to abolishing private ownership in some sense)
The question remains, is there an issue with proprietary software itself or is there an issue with proprietary software licensing and the companies behind them?
Companies like Amazon, Facebook, Microsoft, Google, etc. all remain justified in their actions because of the sole ownership of their software, they get to decide the boundaries of what people can do with it. Further issues regarding their ability to inject code to restrict user freedoms is not something unique to proprietary software and Stallman himself seems to be aware of this. Open source software is not necessarily free software, but does this necessarily mean proprietary software cannot be? Free software is quite clearly an ethical commitment, but it seems to me that free software itself does not exclude proprietary software. As long as the user is not unreasonably restricted in the way they use the software, what issues regarding freedom truly remain? There may be issues with trust, but ideally this could be dealt with in adequate user protections in case of violations.
Stallman attempts to argue that proprietary software is necessarily harmful to users, but this seems like a poorly founded argument. In order to argue that proprietary software cannot be free software, you must also argue that proprietary software is necessarily harmful, which is not the case. It is a stretch to claim that a program exerts control over a user, especially when you consider the nature of computer programs as useful tools. I believe this is why the average person does not care for free software, because most people would agree that they are perfectly free in the current use of their proprietary software. Sure, they may not be strictly speaking 'free' in the most libertarian and perfect sense, but it is likely more than adequate.
Instead, it may be more important to focus on more generally protecting user freedoms with ethical software licenses. This means that Dan would not go to jail for sharing his book with Lissa (or even giving it away), and rather he would only see trouble if he actually 'gave away' multiple free copies of his books to many people without a license. Users could use proprietary software/hardware such as on Amazon's Kindle e-readers and be protected from Amazon potentially removing the files they put on the devices. It is not totally necessary for a user to be able to modify the program as they see fit, as seen in the case of the everyday person, this is not even a consideration. In order to be free they must be free from force or coercion, so the focus should be rather on mass redistribution and not exerting control over content once users have purchased it. Avoiding problems such as the Amazon one are primarily licensing issues, people did not own the books they bought, and this is something that could reasonably change.
I would even go as far to say that proprietary software need not even restrict modification, and that users could modify their programs so long as they respect the terms of the license agreement (not reverse engineering code and reselling/distributing the program).This way, we could have peoples' intellectual property remain protected and paid for while still defending the users right to its use it, barring only the most egregious offenses (e.g. mass redistribution/reselling).
None of this is to say we don't need open source or free software, it would absolutely remain the most appealing form of software due to its superior security, learning, trust and modification qualities. Not to mention the 'freedom' it can offer, in the strongest sense of the word.
The purpose of this discussion was primarily to promote caution in the hard and fast claims of such arguments and point out some potential flaws in their reasoning. It's these sorts of flaws that keep such arguments ineffective in practice because they too quickly dismiss the opposition, rather than attempting to find a more practical solution. I love open source, free software and what it stands for, but arguments like Stallman's rub me the wrong way.
I make these articles to get people thinking about these issues and to get people thinking twice about arguments that make these strong and extraordinary claims, which influence people to take some pretty strong positions online and in life. I feel that particularly in the space of open source and free software, the discussion is too often crowded with IT professionals, who are naturally biased in their judgements here. It is pretty difficult to find contrary arguments on the internet on such topics, but here I am attempting to provide one despite my own biases.