Commentary-- Most software development is a local activity, done directly for the benefit of the people writing it.
The wonderfully bureaucratic sounding "Directorate General for Enterprise and Industry"
of the European Union recently commissioned an extensive study on what they call FLOSS (Free/Libre/Open Source Software). The result is a mammoth 278-page report you can read for yourself here.
While not exactly a page-turner, it offers some really interesting surprises. One of them was that over two thirds of free/open source software is still produced by individual developers working on their own time and for their own ends. The "hacker in their parents basement" factor is just as strong as it ever was. It really shouldn't surprise me; after all, that's how I started myself (actually in the little bedroom rather than the basement). Somehow, after all these years getting paid in the commercial software industry (and writing free software for most of that time), I've gotten so used to working with others already working in the industry that I forgot my roots. I'm glad to be reminded of what I should never have forgotten: Most software development is a local activity, done directly for the benefit of the people writing it.
Part of the report deals with the little-known (outside of Europe anyway) success of the GNU/Linux deployment in the Extremadura region of Spain, where regional government adopted a Debian Linux distribution for local use. This involved changing the often rather obscure names of free software programs to something more meaningful to non-technical users (why do we use something called "FireFox" to browse the Web?), complete localization, and -- more importantly -- customization of the desktop distribution to make it meaningful to local people.
What this deployment created was interest in IT (called ICT in Europe) among small regional businesses that were not previously benefiting from computer productivity gains. It also led to a strong preference for buying support from local suppliers who, these businesses believed, would give them more individual attention and service than larger companies. Here's the relevant quote from the report on how Free/Open Source software is easier to adapt to the need of local business:
"However, unlike their proprietary counterparts, FLOSS developers allow and encourage those with locally relevant motives to adapt their software. So local users -- and, importantly for building local ICT economies -- small local businesses are entirely capable of providing services and adapting the software to local needs."
But would widespread adoption of free/open source software unfairly damage proprietary companies and reduce the supply of high-quality software to the market? Here's another statistic from the report that surprised me:
"...it is important to recognize the relatively small share of packaged proprietary software in the total market. Packaged proprietary software accounts for 19% of the European software market and 16% of the US software market."
Again, having been in the software industry myself for too long put the blinders on. Most software written today isn't created to be sold in a box from a software store (or now more likely downloaded from the Internet). Most of it is written in-house, heavily customized, or used in embedded systems, and for most businesses is an external cost, not a source of profit. For almost all software users use of free/open source software will reduce costs in their business. Allowing complete control over the software used (such as customizing the source code) is of great advantage for many commercial companies. Nokia was able to create its "Internet Tablet" portable device while only having to write 1.5% of the total software used in the device. The rest is free software.
Here's another example in the report from an investment bank Dresdner Kleinwort, who ended up releasing some internally developed code as open source in order to cut their own costs:
The firm recognised the value of sharing to reduce maintenance costs: "We certainly didn't want to become a software company," explained Jonathan Lindsell, Global Head of IT Business Development. "We shipped source code to our customers all along, inviting them to fix bugs and contribute enhancements, and they did. That's one of the reasons why it's become so incredibly successful."
My own experience here is instructive. I'm forever fixing bugs directly reported by users on the mailing list for Samba, my own free software project. I used to think that most users were getting the software directly from our Web site. But the number of Samba users who are even aware of what software they're using is dwarfed by the number who get Samba embedded in a device or software product from another vendor. I began to understand this after wandering around Fry's electronics (the largest electronics shop here in Silicon Valley) and doing a survey of all the "network disk in a box" products on offer. I soon realized that if the product cost less than $5,000 then it almost certainly had Linux and Samba inside -- and that almost none of the users of these products are aware of that. That's why these days I give much higher priority to fixing bugs reported from vendors, as this magnification effect means I'm helping many more users from suffering from the same bug.
However, there are other ways then adopting free/open source software to create your own local software and service industry, even in Europe. Recently Reuters reported that the Romanian President Traian Basescu told Bill Gates in Bucharest:
"Piracy helped the young generation discover computers. It set off the development of the IT industry in Romania."
I'm assuming that's piracy of Microsoft DOS and Windows, as it's awfully hard to pirate free/open source software. But being a law-abiding sort of person myself, I'd much rather people not learn that the only way to bootstrap your own industry is to steal someone else's property, especially when there's a completely free and legal way to do the same thing. It does show that free or zero-cost access to software is definitely required in order for a society to develop ubiquitous computer literacy.
The most exciting possibility here is the "One Laptop per Child" project, which plans to get small portable computers into the hands of as many of the world's children whose governments will spend the $100 per machine. Most of the software in these machines is Free/Open Source and if widely successful, this machine has the potential to open up the ability to participate in the creation of software to anyone with the talent to do so, not just the economic means to buy a computer.
The report also comments on the use of various different kinds of licenses in the creation of free/open source software. I'm a big fan of the GNU GPL, the license under which Samba is released. The GPL has been criticized by some as a "viral" license (it requires you to share your changes to the code with others) and so I was very pleased to see the following in the report:
It should be noted that some large vendors are particularly critical of reciprocal licensing as being "anti-business" and preventing commercialisation, while approving of "permissive" licences such as the BSD licence, which allow FLOSS software to be appropriated exclusively into proprietary software. In fact, reciprocal licenses only prevent competitors from indulging in IP theft, by taking software written by others and "commercializing" it as proprietary software, preventing the original authors to benefit from modifications to their work. This is why the argument against reciprocal licences rarely comes from firms that actually release FLOSS software, but rather from firms that want to incorporate software written by others into their own works, without providing anything in return.
This pretty much matches what I've experienced myself :-).
There's too much in the report to comment on all of it here. All I can say is if you're in the business of creating, commissioning or even just using software in your business, I would heartily recommend reading it. There are some startling facts and statistics in it which will challenge many of the conventional beliefs held by the software industry. And, as it's written in the style of a typical government report, it makes a great nightcap, even for insomniacs.
Finally I'm off to FOSDEM, Europe's primary free software/open source conference taking place in Brussels between February 24 and 26. If there are any readers there, please come and say hello!
Jeremy Allison is one of the lead developers on the Samba Team, a group of programmers developing an Open Source Windows compatible file and print server product for UNIX systems. Developed over the Internet in a distributed manner similar to the Linux system, Samba is used by all Linux distributions as well as many thousands of corporations worldwide. Jeremy handles the co-ordination of Samba development efforts and acts as a corporate liason to companies using the Samba code commercially. He works for Google, Inc. who fund him to work full-time on improving Samba and solving the problems of Windows and Linux interoperability.
