Really Making Money with FOSS
So many people have attempted to describe, explore and probe the economic workings of the Free and Open Source Software business. Recently Nick Fox has given us his thoughts on what this means.
And this is my critical article explaining why he is wrong, sorry Nick got to be critical. The first half of the article is fairly correct as far as I know. so I’m just going to skip to the bad parts:
Commercial software being generally closed source is a necessary evil.
This assumes proprietary software is the only model for profitable commercial software. It is not. It also assumes that FOSS can’t possibly be commercial, a big mistake and a common myth. You can take a copy of a GPL licensed program and sell (that’s right, for money) the software to someone. So long as you don’t remove the recipients freedoms and they get to redistribute, that may sound like it crushes your commercial opportunities and it does sort of, but I’ll get to that.
However creating software as a business requires a level of production protection that is not usually accepted in the Linux circles. Compiled and protected sources are a bad thing for software freedom and progress, however they are good for free market business.
This part floored me. The idea that proprietary software could be in any way free market is so absurd that I can’t understand how this idea has come about. The nature of the free market is that goods or services will be priced very close to the costs of replication and distribution when supply is above demand, software had infinite supply and will always be above demand.
The costs of software replication and distribution are very close to zero. So in a free market, all software is free of cost by order of the invisible hand. What is NOT free of cost is the creative production. But because the creative industries can not yet find a secure way of funding their production; they have gotten government regulation (copyright) to warp the nature of the free market to create temporary monopolies on distribution instead. Shifting costs around. And if you’ve been following your economics, a monopoly will tend to price things at the very maximum a customer is willing to pay, not the minimum it’s economically sustainable to charge (as in the free market).
The separation of first creation and replication of copies, I think is important to understanding the nature of these economic processes. Writing software is creative production, copying software is replication production, they are not the same thing.
So, what we have here is an industry that is not only removing user and developer freedoms, but it’s doing it at the expense of the free market too. On the other hand, FOSS is free market, it doesn’t claim to have created an economic rewards system to drive the cost of one economic activity into another one. It’s goal is to create a social and legal framework for collaboration. The software production is priced accordingly through commission or (more normally) through the needs of the user’s time and the distribution is priced very close to zero.
The incentive to make money inside the Linux community will help to break the cycle. When businesses find there is money to be made by producing Linux based applications for busness users, it will help bring Linux to more desktops. While I very highly advocate the Free and Open Source movement, I am suggesting that closed source software for sale does have it’s place, and in fact may help bring Linux to more desktops.
I recommend watching this video on motivation first:
As we’ve already discussed, there is money to be made from FOSS, it’s just you have to follow the economic landscape. You can’t go begging on government to bail you out of your unprofitable software distribution business with anti-free market copyright laws. You have to make your money from the development of software, not the distribution of it. This shift in thought needs to accompany the shift to FOSS, because without it FOSS will be uneconomic in the old mindsets.
Closed, proprietary software has no place in a rational, enlightened, scientific and honest economic world. It is NOT a necessary evil, it is a plain misunderstanding of economic mechanics. An attempt to create rents on distribution instead of maintaining the economic costs of production. It’s is not good, useful or progressive and the longer we hold onto some of these mythologies the longer it’s going to take to drag our industry out of the dark ages.
My own thoughts are that in order to fund software production properly, we need to have ways of getting money from users who want software to be made or changed, to programmers who want to earn money writing software. It’s not an easy task.
Tags: business, Economics, floss, foss, free software, linux, money, open source
“You have to make your money from the development of software, not the distribution of it.”
That is what closed-source/proprietary software and it’s developers are doing when they make money from the development of software.
Yes, the cost of Windows CDs for example and all of the packaging and marketing does take some cash; but let’s put that all aside and assume that someone can legally download the ISO images and burn their own CDs/DVDs, except they must pay for the software; then in that case the money is all going towards the development of the software, not the marketing and distribution of it. So instead of Windows costing $300, it may cost $200 let’s say (probably not in reality because companies will sell something for as much as they can get for it, but you get the idea anyway.)
You can’t make money off of F/OSS in the developing it because according to the GPL you need to openly and freely distribute the source code; so once you do that, then why would anyone pay you to develop something they can now do for themselves? If you custom develop some software for a company, chances are they aren’t going to let that out into the public, so it’s not GPL’d and F/OSS, it’s exclusive, private, closed-source, proprietary to and for that company etc.
I think you are off the mark on this one Martin.
The key is not in the development of F/OSS, but in the SUPPORT of it.
Let’s say you code BlahServer and you give away the source, GPL it and all of that, yup, people CAN write their own modifications, change things around, etc, BUT if they want something they don’t know how to do, want YOU to do it, to train them etc, then they can go to you and pay for such a service. After all, isn’t that the model Canonical is based upon? Ubuntu is free, but if you want them to train you or support it, it costs.
Now granted Mark made his money elsewhere and he bankrolls Canonical because quite honestly, it’s just not a great business model with a lot of potential for ROI. It CAN work, but it’s not McDonalds, ya know?
This is a similar situation as to what plagues the recording industry – and outdated business model that no longer works. I think you are probably closer on the mark with the training of people, you can make money there.
You may agree – or not — and that’s fine because we both have our opinions. I would wager though that you do not make your living from F/OSS development, and you’d be hard pressed to find someone who does. Possible? Yes. Probable? No.
> The creative industries can not yet find a secure way of funding their production.
Indeed. If FOSS/GPL was such a way, everyone would readily switch. However, – not for business users, not for custom software, but for mass market like personal computers – there is no money in support (forums, LUG, IRC), in customizing (users generally do not know what they want, individual suggestions is hard to deal with). I’d be happy to be proven otherwise, but right now there is no stable solution to the free rider problem.
“…goods or services will be priced very close to the costs of replication and distribution.”
That isn’t true for (say) airplanes. The cost of engineering/design must be spread over the # of units. If #units is very large then you’d be correct, but that isn’t true for all markets. Example: medical software.
Absolutely. the cost of developing medications is another one such example, of which there are many.
[...] a comment » Martin just posted a lengthy critique of an article by Nick Fox on Linux business models. I’m not going to address Nick’s [...]
You know, I’d be interested in seeing a study done relating to the sales of Windows or the sales of support for it and which one earns more money?
I’m going to wager that more money is generated by companies that support Windows because there is only 1 company that makes Windows, but thousands of companies that support it. Microsoft has a stranglehold on the source code and programmers had to beg and plead to get the API access they did. there are some parts of Windows that will NEVER EVER see the light of day outside of Redmond.
In the F/OSS World, we have thousands of people writing for Linux AND also supporting it; so if all things were equal in that regard between Windows and Linux (assuming both were F/OSS), I’d imagine that both would have an equal time trying to make money off of what is freely given away.
As I said before, I think the key is education, and I see Martin (and others) really leading the way on that. Something like 1% of all PCs run Linux, and of that 1%, 30% run Ubuntu as their choice of client/desktop OS.
There’s ways to make money in F/OSS, but software development isn’t it. Even support of it is *kind of* a way to make money, but it’s still tough…possible, yes, but tough. I think it has to be more along the lines of training, education, maybe even slowly migrating people and companies over to Linux, etc.
Martin, I really enjoy the debate of ideas. Thanks for your post, I have written a reply article you might want to read: http://www.rubmyubuntu.com/LinuxSoftwareBusiness/Part2
I look forward to your response.
Nick, In your response to Martin, you brought up many of the same points I brought up, which I believe you came up with on your own, not after reading what I’ve posted here.
For the record though, Canonical is funded by Mark who made his money elsewhere. Sun got bought out by Oracle, so that FOSS company wasn’t making any money or else they wouldn’t have been purchased. On a side-note, I’d like to see someone write about how Sun who bought MySQL is now owned by Oracle. Oracle owns MySQL, Java, Seibel, and quite a few other things. Oracle (in my mind) is getting to be a sort of smaller version of Microsoft with it’s pending monopolistic status.
Anyway…back on point….. like I said, we agree. I don’t see money being made in FOSS development. If there was the cash in it, more people would be living off of it, getting paid (even for part-time work), etc. I hope Martin allows this reply of mine as well as another one that is pending, because I think such discussions (important because of the 3 blog posts surrounding this topic) have opened a can ‘o worms.
Maybe this has been discussed before, probably has, but what has been really DONE about it? I would like to suggest that someone form a group to teach others. Come up with a plan, come up with a consistent method and format for training people, come up with the basic, intermediate and advanced levels, come up with real-world examples, solutions, scenarios, and run with it.
Martin, I recall you putting together a bunch of PCs, why not make a non-profit, get a PC lab going, setup a training lab or school if you will, and charge people a NOMINAL fee to attend there. Right now it is so cost prohibitive to attend school, jobs are scarce in the USA, etc.
Instead of FOLLOWING the old trend of arguing over this stuff and never getting anywhere, let’s START a trend and LEAD by EXAMPLE.
You have my email address. I’m in. Who else is with me?
It’s funny you should mention that, I have been teaching at the SETC community center for over a year and a half. Each week a session is taught on Ubuntu Desktop and it involves numerous LoCo volunteers.
Not content with that, I’ve written a very alpha systems administration course and I’m a board member for the Ubuntu Learning project. a project seeking to create the background materials required for teaching around the world.
So I’m already leading by example, already doing more than just procrastination. I hope you can appreciate that I’m not just a disembodied voice in the community.
> You can’t make money off of F/OSS in the developing it because according to the GPL you need to openly and freely distribute the source code; so once you do that, then why would anyone pay you to develop something they can now do for themselves? If you custom develop some software for a company, chances are they aren’t going to let that out into the public, so it’s not GPL’d and F/OSS, it’s exclusive, private, closed-source, proprietary to and for that company etc.
This is wrong. Look at the economies created around gnome, there are several companies who hire gnome contributors to improve and develop gnome technologies. The companies earn money by improving and developing software to meet other peoples requirements. (Nokia for the nx00 series, intel for the moblin laptops etc)
Once some GPL software is released no-one is going to pay you to develop the same thing again, someone might pay you to make improvements. This is likely to be cost effective to them because you have skills and experience in the project already.
Some cases where it is in a companies interest for your work to be open source:
1 It was to improve an existing piece of open source software and they’re going to be distributing it
2. It was to improve an existing piece of open source software and they’re not redistributing it, but they don’t want to have to keep rebasing your patch on new releases
3.It will speed up development of the software
4. Your software is in an area that is not the companies focus and will bring no to little competitive advantage. e.g. a company fixing bugs in a 3rd party software library
“The idea that proprietary software could be in any way free market is so absurd that I can’t understand how this idea has come about.”
Yeah it’s just propaganda, not really an idea. Software isn’t scarce so applying it directly to economic theory is mistaken but if one really wants to, the best label for Free software is free market capitalism. Proprietary software would be mercantilism. And the closest thing to communism might be semi-free software.