Mr. Perens, what was your motivation as an Open Source developer?
Perens: Most Open Source developers start out to solve a problem. My first Open Source program, “Electric Fence”, was created in 1987 to find bugs in C programs at Pixar, which later became a famous animation studio. I sent Electric Fence out to the whole world on Usenet, hoping that it would be useful to others, but having written only partial documentation. A day after this, someone else wrote full documentation, which he needed to deploy Electric Fence at his own company, and sent it to me. So, I had gained a day of someone else’s useful work, just because I sent the results of my work out on the net.
Who contributes today to Open Source?
Perens: Volunteers, Linux distribution companies, companies with a single Open Source program as their main product, companies for whom Open Source software enables sales of hardware or solutions, service businesses, end-user businesses and their contractors, government, academics and scientific researchers. There is a continuing transition from volunteer to professional, as Open Source is used increasingly in companies. Former volunteers are gaining employment in organizations that support their Open Source work on company time.
What made Open Soource succeed in business?
Perens: From an economic perspective, there are two kinds of software: differentiating software makes the product of a business look better to the customer than a competitor’s products, and non-differentiating software might be essential to run the business, but doesn’t attract customers. Before Open Source, and especially before Linux, computer manufacturers spent a lot of money on non-differentiating software: for example, IBM spent Billions of dollars to develop its AIX operating system, which was very similar to the Unix systems of vendors like HP and Sun. AIX was enabling technology but it wasn’t a business differentiator.
As the Linux operating system kernel became more functional, IBM and other companies realized that they could save money by using Open Source for non-differentiating software like operating system kernels, and they could then have more money left over to spend on things that actually differentiated their business.
What is the long-term economic impact of Open Source?
Perens: Just like the major economic impact of Microsoft was the fact that they have enabled a great many businesses – their customers – to do business more efficiently, Open Source equally enables a majority of web servers today, a majority of e-mail deliveries, and many other businesses, organizations, and personal pursuits.Thus, its economic impact must already be numbered in many tens of Billions of dollars.
How can OS help companies with their software investments?
Perens: Perhaps 90 percent of the software in any business is non-differentiating: operating systems, web servers, databases, Java application servers and other middle-ware, graphical user interface desktops, and the general tools used on GUI desktops such as web browsers, e-mail clients, spreadsheets, word processing, and presentation applications. It wouldn’t hurt your business for your competitor to understand how every bit of your non-differentiating software works.
Consider the book-seller Amazon.com: they use a lot of Open Source software to run their web site, and it is no problem for them to share work on that software with other book sellers. But Amazon also has a recommendation system that differentiates their business: it uses data-mining to tell customers about books that will probably interest them. It would be a mistake for Amazon to “open source” that, because they’d lose a business differentiator. So, the challenge for the IT manager is to classify what software is differentiating, and what is non-differentiating, and to move as much money to business differentiators as possible.
Why did many consortia for collaboration fail?
Perens: Consortium product planning often devolves into irresolvable arguments among the companies, because each has a different marketing idea and marketing arguments between companies are subjective and difficult to resolve. Closed consortia generally are directed through pay-for-say, while technical merit would be the case for Open Source. With pay-for-say, a member can work to the detriment of the overall project when that is to the member’s advantage. Given the poor history of consortium development and, in contrast, the high rate of success for large Open Source projects carried out by the same groups of companies, it seems that the fairness imposed by Open Source licensing is an essential component of effective collaboration between a large number of parties with different interests.
How should business-differentiating software best be developed?
Perens: The critical thing about business differentiators is that they must be held close. So, a company must use internal programmers or external contractors for differentiating software, so that you own the result and can control who gets it. And of course you must bear the entire cost and risk of that development.
Another paradigm of software development is to buy retail software.This makes mostly sense for products that will have a mass market. A retail software vendor has a very large cost for locating the customer, for example advertising or packaging. Retail uses many customers to distribute this cost, so that nobody pays too much.
An Open Source paradigm of software development generally uses the internal programmers of a company, but only a few of them in each company. When many companies participate in a project, like the Apache web server or the Linux kernel development, they can build up to a team of thousands of programmers, but with no company paying for too many of them or risking too much if some development doesn’t work out. Generally, companies join Open Source projects when the software is already working – they start to use it first, and then they join the project to add features. For Open Source, it works best to release early and often, to attract more collaborators.
How can customers support their businesses by leveraging Open Source around their SAP core software?
Perens: SAP’s software can help bottling companies in their relationship with their distributors. It doesn’t make any drink taste better, but it helps make sure the drink is there in the store. Obviously, there is Open Source software for this niche, but none as thorough as SAP. SAP can continue to offer differentiating value to its customers as long as it continues to invest in software development and customer service, to keep its product easier to use and more functional than any other product, Open Source or proprietary.