Free software, will it work?

Aaron Swartz in his response to Nick Bradbury's post on software piracy is trying to imply that all software should be free.

On Aaron's home page he refers to GNU's Free Software Definition. This definition lays down four freedoms:

-The freedom to run the program, for any purpose (freedom 0).
-The freedom to study how the program works, and adapt it to your needs (freedom 1). Access to the source code is a precondition for this.
-The freedom to redistribute copies so you can help your neighbor (freedom 2).
-The freedom to improve the program, and release your improvements to the public, so that the whole community benefits (freedom 3). Access to the source code is a precondition for this.

Where as I agree with the first two (0, 1). I find it hard to believe that the programming community can sustain itself, if freedoms 2 and 3 are aloud. The problem is that these two fail to protect the value of the software.

You might argue that this is no different from your right to sell a car second hand. The difference is that when selling a second hand car there is a two-way exchange of value. Not only does the buyer give up value by handing over cash, the seller gives up value by handing over the car. As a result, there is an incentive for the seller to charge a reasonable price. The incentive that sellers have to keep prices up maintains the value of cars as a whole and means that car manufacturers can keep selling cars.

With software, the equation is a little different. As it costs nothing to duplicate software, when you give someone a copy, you suffer no loss. If there is no loss, there is no incentive for the seller to make a charge. Put another way there is nothing to stop two sellers undercutting each other until they are giving the software away free.

An example:

A developer works on a project for three months. It costs him 10,000hmu (hypothetical monetary units) to sustain a reasonable standard of living for this period. 10,000hmu is rather a lot to pay for a piece of software so he decides to sell it for 100hmu in the hope that he can sell it to 100+ people.

The person A decides to buy the software, and decides to sell copies to regain his 100hmu loss. Because in a free market with two people selling the same product the cheapest man wins, he decides to sell it for 1hmu. The 99hmu difference between the 100hmu, person A bought the software for, and the 10hmu it is being sold for does not hurt as the software is not lost in the transaction.

Because of person A's decision, the developer can now only sell his software for 10hmu.

Applying the same logic person B having bought the product from person A might sell the product for 0.01hmu. This not only hurts the developer but it also hurts person A as neither of them can get what they wanted for the software and so the price cutting goes on!

There is a flip side to copyright; it gives authors a potential monopoly. If society grants authors total copy protection, it is easy for a seller to charge way in excess of what a product cost to develop. In most cases, the buyer has a choice of other products that he can use for the same purpose thus generating competition. In other instances, Operating Systems, for example, he may not and this is when the monopolies start.

In conclusion, while there will always be sources of free software, this model of software funding will not scale to the development industry as whole, but until another form of value protection is developed, there will continue to be a risk of monopolisation.

P.S. Is it a coincidence that most producers of free software are academics (funded by government) or students (funded by their parents)?


Re: Free software, will it work?

First, I think you have it wrong about the people who produce Free Software - they're not mostly academics and students. I'm not sure where you got that from; although it's true that academia does have a propensity to release software as Free Software (that's a tradition thing, though.)

I understand your argument (I've heard it many times :) but I'm not sure you understand the case being put forward. Let me state it in a slightly different manner, and see if your response is the same.

If you've seen StarTrek, you'll know the idea behind replicators. You walk up to this hole in the wall, and you say "I'll have a cup of coffee, please". The replicator immediately responds by generating a cup of coffee, from some store of "matter". The cup is strong, and the coffee is hot and good. The StarTrek community don't actually have to worry about food any more, because they have replicators which work on raw matter. Replicators are fairly simple machines (in their time) and can do a lot of work, so producing food stuffs (and other useful tools) is easy and cheap. In fact, it's basically zero-cost, because matter is plentiful throughout the universe.

Now, in your scenario about, the "value" of food is being diminished by replicators. All the effort that goes into preparing a dish is basically wasted, because a replicator can do it faster and cheaper. Think of all the restaurants that will be put out of business! Not to mention the farmers. In fact, rather than having replicators and solving world hunger, what we should actually do is *license* them. We should sell someone a replicator, but only on the condition that they don't share it with their friends, and that they don't compete with commercial suppliers using it. This food "development package" must be used solely for personal reasons, and you have to buy matter from your local grocery, and it costs £5 per kilo (which is more than the equivilent weight in cabbages, but so what). We'll also ensure that once something is replicated once, there is an ID on the item so it cannot be replicated again. Simple copy protection.

Is that really what you want?

(this example may be hypothetical. But, when industries expire, when they become obsolete, we push them aside. We don't let them impede progress. We lost our hosiery industry in this country. We lost our shipbuilding, and our miners. We're in the process of losing our agriculture. Proprietary software is the same; it's an outmoded concept that has long since been useful).

Oh, and your comment box is way, way too small :)

Comment from Alex Hudson at Saturday, 10 January 2004 07:35PM (GMT+00)

Re: Free software, will it work?

I admit that my comment about academics was perhaps a little bigoted, however it leaves me with one question Â?Why does anyone do it?Â?

I am assuming here that a replicator does not have the ability to come up with a coffee all by itself. Someone has to tell it at some point what a coffee is. People might develop coffee just for the hell of it, as it does not really take that much effort, you just pick the beans and boil them.

If you take a car, you have to do lots of development, testing and tuning before you have a good car. A team of fifty might take two years to do this (in reality its probably more). Are people going to do this just for the hell of it? I donÂ?t believe they will. Why would they? Even if they did do it, would they share their car design with anyone else? Having a car would give them a big advantage over everyone that is still on a horse.

Another way to look at it is this: With a food replicator, the cup coffee comes in would always be the same shape and colour. No one would ever program your replicator with a different shape or colour cup, as there would be no reward for doing so. If all cups were exactly the same, wouldnÂ?t life be boring. In addition, what would happen if one day you felt like having more coffee and wanted a larger cup?

You could include the cost of developing all the required design patterns in the cost of the replicator. The design patterns needed however, would be infinite, and the cost to the replicatorÂ?s manufacturer would also be infinite.

I suppose the Government could pay for it all, but would you really want that?

Comment from Martin G. Brown at Monday, 12 January 2004 11:31PM (GMT+00)

Re: Free software, will it work?

Why do people do it? Because it's fun? It's educational? Because they're paid to? There are all sorts of reasons. Some of it is commercial. Some of it is academic research. I don't think the licensing terms speak much about the actual development means or method - I would guess it does have an effect, but I don't think it's even 30% of the whole story.

All cups being the same... I think you're missing the point. You're arguing against people "devaluing" software because they shouldn't be allowed to give it away, or something? Really, who are you to tell people what they should and shouldn't do with something they have written?

I think you have been sadly brain-washed by our current state of affairs. Take Microsoft as an example. They have $40 billion of *cash in the bank*. They make between 60%-80% clear profit on Windows and Office systems. That means that they could easily take a 50% price cut, and they would still be making money.

The truth of the matter is that software is cheap. Dirt cheap. Anyone can write it (it costs maybe £10 for a GNU/Linux set, which comes complete with source code, tutorials, development environments...) and if you're willing to use the banks of free software code out there (such as Perl's CPAN), you can create quite complex apps simply by stitching other pieces together. This is not rocket-science, nor is it particularly valuable. When you have a good user base, you can also get people to test out new versions, send in patches and improvements, help with documentation, etc., and the cost of development is suddenly spread across tens if not hundreds of people.

I think you vastly over-estimate the cost of producing software, under-estimate the commercial value of free software, and this is why I find your original piece so misguided: we don't need to "protect" the value of software, we let the market sort that out. That's what a free market is for. Free software is also not a source of monopolisation; I'm afraid you need to look to the proprietary market for that.

Your title is "Free software, will it work?". Well, the GNU project - one of the most well know projects - has been successfully (and commercially) producing free software since 1984. Linux has been around as long as Windows NT, and other free OSes (like FreeBSD) have been around even longer. The question should not be, "will it work?", rather, "why are you not using it?" :o) It does work, and it's here to stay and grow.

Comment from Alex Hudson at Friday, 23 January 2004 11:40PM (GMT+00)

Re: Free software, will it work?

I know web logs aren't strictly accurate, but they give a reasonable idea. I notice from looking at the logs of a number of large web sites I have been involved with that nearly 95% of people use Internet Explorer. What happened to Netscape? I believe it died a death when MS gave IE away for free!

So Linux is a success, but ask yourself this. Would it be a more or less of a success if Microsoft sold Windows for just £10? Linus Torvalds (Linux's Author) was one of those Academics I was talking about, he was at the University of Helsinki in Finland when he wrote it. Anyway, how much did he or the university make for his trouble? Not as much as I think they deserve. ItÂ?s the likes of Red Hat that made all of the money.

I do believe that Microsoft is having a joke with the price of their software, but I donÂ?t believe having them charge nothing for it is a good idea either. I'm not trying to tell authors of software what to do, just pointing out that there is a flip side to giving software away free.

I agree entirely with you when you say Â?Free software is also not a source of monopolisation; I'm afraid you need to look to the proprietary market for thatÂ?. I feel that I might have made a wording mistake at the end of my post. What I meant to say was: Â?Â?but until a form of value protection *other than copyright* is developed, there will continue to be a risk of monopolisation.Â?

Comment from Martin G. Brown at Friday, 30 January 2004 03:51PM (GMT+00)

Re: Free software, will it work?

Microsofts IE marketshare correlates strongly to it's desktop share; which tells you something: it's actually not the price that got them marketshare, it's because they were illegally able to tie their browser to the desktop. The price of Netscape never really meant much; after all there was free (gratis) competition already.

Also, Linus' project was a personal one, not a Government-funded academic project. Hi s University had very little to do with it.

I'm afraid I didn't understand the rest of the argument; you castigate Red Hat and others for making money from GNU/Linux, and then say giving it away free destroys the value of software. So, which is it to be? (The answer is actually that it's not given away for free; for example, Red Hat now charge for every OS they sell and have always charged for most of them). Free (as in freedom) software is not really much about the price; the price is irrelevant to the argument whether or not software should be Free.

I'm also still intrigued by your monopolisation argument. Have looked at how many vendors sell GNU/Linux OSes commercially, for example?

Comment from Alex Hudson at Tuesday, 03 February 2004 04:28PM (GMT+00)

Sorry, this post is no longer accepting comments.