Episode 192: Open Source Development: Perspectives From Management Science
18:10 Share

This is Software Engineering Radio, the podcast for professional developers. On the web at se-radio.net. SE Radio brings you relevant and detailed discussions of software engineering topics at least once a month. SE Radio is brought to you by IEEE Software Magazine. Online at computer.org slash software.

This is an interview on open source development with Professor Georg von Krog, who is a professor in strategic management and innovation at the Swiss Federal Institute of Technology, ETH, in Zurich. Georg von Krog has done a number of studies on open source development topics, including code reuse, innovation, motivation, and the social practice of open source development.

My name is Torge Dingsøer and this is an interview for IEEE Software. So working in management science, how did you become interested in open source development? Well, I was primarily interested in open source software development because it represents an entirely new way of innovating which turns out to be both effective and efficient.

And it is an interesting model that we, Erik von Hippel at MIT and myself, call private collective innovation. That means that private contributors, such as individuals and firms, provide work and software to a public good innovation. It's public good innovation because...

You cannot prevent anybody from using the software. One person's usage of the software doesn't really negatively interfere with another person's use of the software. So in that sense, it's public. It's a public good innovation. So it represents a fantastic incentive

a fantastic solution to an incentive problem of how to create public goods. That was the reason why I became very interested. Are there examples of similar phenomena previously?

Yes, I mean, science in some ways has some of the same characteristics, but science tend to be more of a collective action type of innovation. That means that society, for example, subsidizes scientists or, you know, not only society, but also, you know,

private fund holders subsidize scientists to also create public good innovations through the science that they do. So for example, if you write papers or if you do a laboratory experiment that you report on, this is also becoming a good for society, a public good.

But here the model is slightly different because salaries, for example, are paid by the state or are paid by private endowments that allow this science to happen. What's the current state of open source development? Open source software development has seen an incredible growth.

for the last decade, both in terms of the number of contributors as well as the number of projects being launched. If you just count the lines of code of open source software development projects, you see that it's been an explosive development. The number of people getting engaged, the number

public's involvement in open source has been has seen a phenomenal development if you look at the

such as development sites, such as SourceForge, for example, you see that currently there are more than 1 million people involved in open source software development projects with in excess of 100,000 software projects. That is not to say that all of these actually are successful in the sense of attracting contributions and attracting code and developing product or developing products

But they are part of a phenomenon which is kind of huge in society and that also changes the face of the software industry to some extent. Is it possible to say anything about how those successful projects are managed? I think that we start to know much more about the way that open source software projects are organized.

And in the past, we used to think of them as very, very open bazaars where everybody could contribute. Everybody could contribute comments and code and could fix bugs in the software environment.

And so on. And today we know that you need in most successful projects a coherent development team, small team that takes on suggestions from a large community of collaborators.

review modules of software before implementing them with the official release of a software. So you have typically a very large group or community of contributors and a much smaller group of core developers that secure the quality and sustainability of the project.

Are there any lessons from management science that could inform how to organize open source software development? I think that management science can learn a lot from open source. Of course, open source can learn from management science too, but open source shows the way for companies that want to open up their innovation process to the outside.

And that means that they motivate people to come up with ideas and identify technologies that they can use internally for their own product development. And we clearly see that, you know, open source software projects, they tend to motivate people to work.

to contribute in an outstanding way, in ways that, for example, companies sometimes find difficult to emulate.

But companies that do try this and do try to work with an open source model, such as IBM, for example, they are successful in using ideas and knowledge from the outside to innovate internally. Is it possible to characterize the open source developer?

The open source developer can basically be anybody with an interest in the functionality that the software offers or the code or some algorithm that can be implemented in code. I think that...

It used to be the fact that most people were volunteer contributors coming from universities, perhaps having a full-time job as a software developer in a company, doing open source on their spare time. But as this phenomenon has increased its impact and become more

become bigger. We also see that today many, if not even most, open source software developers work for companies and they do open source development in their work time.

In a recent study, you examined motivation amongst open source developers. Could you briefly describe what is it that motivates an open source developer? Yeah, that's of course the critical question, because if we don't understand what drives the open source developer to contribute, then we don't know if open source is going to stay a sustainable phenomenon.

The same thing goes for companies. Unless we understand how to motivate outsiders, we will have a problem to embark on these open innovation models that tend to propagate amongst firms or throughout firms and industries. Now, if we go back and think about what really motivates the open source developer in general, there are different perspectives on that.

What is clear is that you can pay the developer to contribute code. And here the motivation wouldn't be too different from what you would find in a company. So people are extrinsically motivated. They would write open source code to get their pay at the end of the day and to make a career within companies. And plenty of studies actually show that open source software developers are often extrinsically motivated. On the other hand side, you have intrinsic motivation.

In open source, you have self-allocational tasks, meaning that you can typically work on tasks that you find very interesting and gratifying to work on. So rather than

playing or rather than responding to an external incentive, you actually respond to the fun and the joy and the pleasure of doing a certain type of work on a certain task. This is intrinsic motivation. And I would say that open source software developers today very often have a combination of intrinsic and extrinsic motivation driving them. But sometimes

If you look more closely at what drives them to develop, you also see, for example, that peer recognition is critically important and learning. People like to be recognized for good software work amongst their peer developers, who they also respect. And

They also like to engage with very smart people in the open source arena who give feedback on the code that they have developed, whether or not it satisfies certain quality criteria. And lastly, and this is what we found in this recent paper in MISQ, or what we argued in this recent paper in MISQ, people might actually have a long-term drive to

to create something which goes beyond the short-term code or accomplishment of code that they do, but they want to contribute as a live project to a larger movement, which is called the open source movement, and thereby also create something that might be positive for the software industry and positive for customers and in some ways positive for society.

What kind of benefits do companies gain from allowing their employees to cooperate in open source development? They get many benefits. For example, they ensure that the open source software that they run with the commercial software, for example, or that they run on their hardware, that they can...

ensure that the development goes in a certain direction. They can ensure that some of the work that they are actually doing in-house can be tested and improved and refined when they take it into the open source project. Those are typical benefits. They can also benefit from, for example, a reputation gain in the field of software developers. You can have benefits such as

For example, teaching or learning amongst your own software developers. So some people, if you have a software development team that works with open source, they can actually experiment. They can test their ideas with the open source community. And that's kind of a low cost test bed for software development compared to what you would find internally.

What do you see as the main challenges now for future open source projects? I think that what we have seen is a very strong fragmentation of the open source landscape. And when I say fragmentation, I both mean the number of new products that have been launched, that have increased tremendously over the last couple of years, as well as a tendency to fork out

the code base of certain projects, making it sometimes hard for developers to choose which product to work on. And I think this increasing fragmentation of open source software can become an issue that needs to be resolved for the future sustainability of the phenomenon.

I think it's not impossible to resolve it in the sense that some of these projects simply wither and die if they don't attract sufficient development resources. But I'm a little concerned about the obvious waste of resources that you find in that process.

I'm also concerned that some of the projects that would not get sufficient attention and attraction of development time and resources, that these projects might actually have some innovation potential in the sense that they could really represent radical innovations of use to the software industry. So that I am also a little bit concerned about.

I think what is positive, and many people in looking at open source said that, you know, when companies start to get involved with open source, it is becoming increasingly difficult

to envision a future in which open source can remain a viable development model. But I was never that pessimistic, or I'm not that pessimistic. I think that company involvement in open source is very good for the phenomenon because it contributes to the sustainability of projects.

because the companies have customers who partly use open source and therefore they have an interest also in maintaining much of the code base that you find in some of these projects. What kind of impact do you think the open source development has had on the society as a whole? Yeah, I think that the impact has been largely positive in the sense that

It has allowed people to participate in software development and learn about software development who previously could not do so. I think that it has given a lot of companies and individuals access to software of a certain quality that they would previously not have, either for monetary reasons or because they were simply not present in the markets where these softwares were offered.

I think that it has shown society that it is possible to combine

traditional intellectual property regimes of patents and copyrights with novel forms of open intellect open regimes in which you create public goods so this private public combination private good combined with public goods that one is highly interesting and i think it

The model, this private collective innovation model that we see in open source software has and will continue to have a large impact on society. Good. Final question. What would you say are the open research questions now within management science regarding the open source phenomenon? Okay.

There are a couple of important questions and I think the first question is to understand the range of motivations of open source software developers across projects. We know a lot about individuals in specific projects, but we don't know too much about the full population of open source software developers and the full breadth of

that they respond to and the full range of motivations that drive them. We have had until now a very narrow focus on extrinsic and intrinsic motivation, but it will be important to take into account a full range or a much larger range of motivational factors to understand why they contribute. Thank you very much. Thank you. It was a pleasure. Thank you.

Thanks for listening to SE Radio, an educational program brought to you by IEEE Software Magazine. For more information about the podcast, including other episodes, visit our website at se-radio.net. To support us, you can advertise SE Radio by clicking the Dig, Reddit, Delicious, or Slashdot buttons on the site, or by talking about us on Facebook, Twitter, or your own blog. If you have feedback specific to an episode, please use the commenting feature on the site so that other listeners can respond to your comments as well.

This and all other episodes of SE Radio is licensed under the Creative Commons 2.5 license. Please see the website for details. Thanks again for your support.