In association with heise online

26 April 2010, 11:11

Why Making Money from Free Software Matters

by Glyn Moody

Free software began as a political movement: its central aim was – and remains – the propagation of freedom. Later, it became a development methodology too, largely at the hands of Linus, whose geographical isolation in Finland forced him to develop ways of using the Internet to coordinate a new kind of massive, but decentralised, global collaboration. Later still, free software also became a way of making serious money – something that Stallman has repeatedly said he is quite happy with, contrary to much FUD claiming otherwise.

FO$$ Few would deny the importance of the first two aspects of free software – as an ideology and as a methodology – but the future impact of its business model is probably under-appreciated. And yet in its own way, that aspect is just as revolutionary as the other two.

Businesses based around free software have to address a central conundrum: how is it possible to make money from something that by definition is freely available? Richard Stallman himself provided an example of how that might be done, when he started selling GNU Emacs on tapes – at $150 a time. This meant he was financially independent during the early days of the GNU project – a crucial development for him, as he told me in 1999: Emacs logo

I had quit my job at MIT to start the GNU project. And the reason is I wanted to make sure that MIT would not be in a position to interfere with making GNU software free in the way I wanted to do it. I didn't want to have to beg the MIT administration for permission to do this and then maybe they would insist on spoiling it in some way.

Despite the high price tag – and the fact that anyone could make a copy – Stallman's tapes soon sold quite well:

I was getting some eight to ten orders a month and that was enough for me to live on.

Although one reason that Stallman was able to sell copies of his free GNU Emacs might have been the relatively slow speeds of the Internet back in the mid 1980s, another is the fact that people were getting the code from the source: in effect, they were paying for the “official” version. This remains an important facet of free goods: given the choice between getting something for free, but unvouched-for, and paying for something that comes from an official source, many will still choose the latter.

That fact also helped drive the rise of the first paid-for GNU/Linux distros in the 1990s. The very earliest Linux-based systems were produced by Linus in the form of two floppy discs, called “boot” and “root”. Linus placed copies of those discs on a server at Helsinki University. Because they were freely distributable, several mirrors sprang up.

The operators of one of those mirrors, at the Manchester Computing Centre, decided to modify the code to create their own distribution. The reason they did so is significant. As the Readme explained:

The MCC Interim versions of Linux are designed to allow people who are not Unix experts to install a version of the Linux operating system on a PC.

This soon became an important selling point for distros that required payment, which were often designed to make installation of GNU/Linux easier than existing versions. Of course, those same distros could be copied freely, but then it came down to people preferring, in general, to buy from the source of a product, rather than simply an unvouched-for copy.

The move from floppy discs to CD-ROMs, begun by a company called Yggdrasil Computing, brought another element into play. Because the then-new CD-ROMs could store many more programs than before, it became necessary to make a choice about which ones were included. This was an editorial decision, and people who bought Yggdrasil Computing's discs were really buying that company's expertise in selecting applications.

The arrival of Red Hat , which is perhaps the best-known GNU/Linux distro, brought with it yet another feature that people were evidently willing to pay for: ease of configuration and package management. As the company founder, Marc Ewing, told me a decade ago:

It was clear that Linux needed a lot of help in the areas of installation, configuration, and package management. Until Red Hat, there really was no such thing as upgrading to a new Linux distribution; you had to reinstall. That was a huge drawback. It was also difficult to incrementally upgrade your machine. You either went to the sources and configured and rebuilt them yourself, or you took a chance with pre-built binaries.

Red Hat went on to extend the help it gave to users to include support, perhaps the most common way today of making money from free software. A whole generation of companies based around open source have now sprung up that use support as one of the main ways of generating revenue from free code, although this is often allied with dual licensing, and closed-source add-ons.

The rise of this new computing sector is significant for a number of reasons. It has enabled hackers to be paid for their work on free software projects – an important point for attracting new blood into the community. A broad range of support options has allayed fears among general users and companies that free software is risky and difficult to use.

As companies like Red Hat have grown in size and profitability, so has the credibility of free software options among larger enterprises. Profits mean that other, smaller open source companies can be bought, providing a useful payback for entrepreneurs that encourages others to enter the fray by funding new open source start-ups. Money also means more influence at the political table through increased visibility and clout.

But there is another, perhaps less apparent, but ultimately more profound reason why the success of these businesses based around free goods is so crucial: their arrival and success is in many respects a forerunner of coming shifts in many other industries.

For example, like free software, digital music has zero marginal costs. As basic economics teaches us, this means that the price of such goods will tend to zero. That's certainly happening in the world of computing, with Microsoft, for example, offering all kinds of cut-price deals on PCs (notably netbooks) in an attempt to discourage manufacturers from installing GNU/Linux. This knock-on benefit of free software is often overlooked, but is real and increasing as open source applications start to be deployed within companies.

In the world of music, it's more complex. Prices are dropping, not least because most people regard traditional pricing levels as extortionate for digital products that cost far less to produce and distribute than analogue ones. The rise of online file-sharing services has also provided competition, although not of the same kind as free software does to closed source products. As people come to expect digital music to be priced much closer to zero than is currently the case, the recorded music industry is faced with precisely the problem that Stallman and his successors grappled with: how to make money from something that is freely available, or nearly.

Next: The music industry response

Print Version | Permalink:
  • Twitter
  • Facebook
  • submit to slashdot
  • StumbleUpon
  • submit to reddit

  • July's Community Calendar

The H Open

The H Security

The H Developer

The H Internet Toolkit