Benjamin Mako-Hill on Open Source vs. Free, GPL and Prepping for OSCON

By James Turner
June 10, 2008

You may also download this file. Running time:

James Turner: We're talking today with Benjamin Mako-Hill; he has his finger in so many pies I'm just going to read his bio off the OSCON Website.

Benjamin Mako-Hill is a technology and intellectual property researcher activist and consultant. He is currently a Senior Researcher at the MIT Sloan School of Management, a Fellow at the MIT Center for Future Civic Media, and an Advisor and Contractor to the One-Laptop per Child project. He has been a leader, developer, and contributor to the free and open source software community for more than a decade as part of the Debian and Ubuntu projects; he is the author of several best-selling technical books and a member of the Free Software Foundation Board of Directors.

So you sleep somewhere in there too, right?

Benjamin Mako-Hill: Yeah; I try--try-sometimes more than at other times I suppose.

JT: First of all, what are you working on right at the moment?

BMH: Well I have a few projects that are sort of ongoing; I sort of--as I sort of have a couple constant projects that I'm usually working on--work on Debian and Ubuntu is basically something that I do in a sustained sort of way, and of course work with the free Software Foundation as well.

I'm working on a couple projects right now in terms of development. I'm working on a piece of sort of Election and voting software called Selectricity which I've been working on for a good chunk of the last year and then I'm doing some--I have a couple new blogs that I'm trying to keep up too which are sort of trying to explore some new spaces in free software. Obviously one of them is called Revealing Errors and it's looking at errors and the way that we can learn from them, and I've continued to do some research at the Business School at MIT on free and open software communities and how they work and how they can work more effectively.

JT: The distinction between free and open source software confuses some people sometimes; can you comment on why free software is so important as opposed to just open source software?

BMH: Great; so absolutely--so I think that--I think that so free and open source software are basically synonyms that point to the same thing, right? If you take a given piece of software that is open source there is almost a 100-percent chance that it's also free software. For the most part I use the term interchangeably; I use the term free and open source software because usually we're talking about the same thing.

Now there is an important distinction and the distinction is more about why people want to do it. Open source is designed with this idea that--Eric Raymond said that as soon as anyone in a suit hears the term freedom they turn around and run the other direction, right? And I believe that in some situations that's absolutely the case.

So open source is meant to communicate to those people by not talking about freedom but the fact is that talk about freedom is still very important because free software as it started out has motivated a lot of people to get open source where it is today and more importantly, it has a very different message that is also important that also needs to be heard. That message is one about autonomy; it's one about control over one's technological environment. It's one about building a social movement and it's one that has been hugely successful and has brought free and open software to where they are today.

JT: There seems to be an effort over time to make the GPL increasinglyly restrictive in terms of what can be done with GPL(ed) applications. Can you explain why the GPL would represent greater freedom than a licensed like BSD which lets the developer do pretty much whatever they want with the application and source code?

BMH: Hmm; so I wouldn't say that the GPL has changed to become more restrictive. I would say that the GPL represents a set of tactical decisions towards promoting user freedom, so free software and open source for that matter is a set of sort of freedoms or rights to do a set of things--to modify software, to share it, to collaborate with other people, and to use it. So those are sort of the core freedoms or the core rights that the--at the center of either open source or free software.

Now one can just throw a piece of software out there and say, do whatever you want, and obviously for the people that receive that software it will be --they will have all those necessary freedoms or rights--right it will be free or open source software. But one of their options--one of the options if it's been given under a do-whatever-you-want license which is more like the BSD license will be to close it off, to insure that--to make it so that the people down the line from them, the people who they give the software to don't have the same rights.

Now the GPL, the GPL and other copy-left licenses sort of prevent that from happening by saying, no; you need to share--you need to insure that the freedoms that you had when you received the work are something that can be passed on downstream. And that's a tactical decision because the FSF who sort of puts out the GPL and continues to modify it--modifies it believes that by insuring that the downstream users have freedom and continue to have freedom that there will be more freedom overall. The more people that will release their stuff as free software, the more people will be encouraged to keep things more free in the long run and that ultimately there will be more free and open source software as a result. So--and the GPL is designed to protect that.

Now the GPL has for example--the GPL is now talking about things like DRM and saying you can't use DRM to keep users from being able to take advantage of that freedom. Now I don't see that as something that is fundamentally more restrictive than previously; I think that it's just sort of updating the license to the more--to the current status of software because of course when the GPLv2 was written there was no DRM, right; people hadn't thought of it or at least the--it certainly hadn't been implemented and so it didn't make it into the text of the license.

So I think that by updating the license to take into account things like DRM and the current state of software patents what the GPL is trying to do is to create sort of protective--the GPLv3 is trying to create a more sort of protective space for free software because as we've seen, the free software and open source for that matter have flourished under the GPL and under the sort of more restrictive if you want to call it that--licenses that insure freedom. I don't think it's really useful to think of that as a restriction of course because the only purpose of that restriction is to keep people from imposing more restrictions.

JT: We've seen some open source projects like Firefox, Open Office, and Audacity which seem to be making a point of maintaining good cross-platform compatibility. Is making open source or GPL(ed) software readily available on Windows, Mac OS, and other closed platforms a good thing?

BMH: So that--that's an interesting question. I think that in many cases, probably in most cases it is; I think that insofar as there are a lot of people who are already using proprietary software right, like 90-percent--95-percent plus are people who are running proprietary operating systems. By creating versions of free tools that run on those operating systems the--it's unlikely that you're going to encourage anyone to stay on the proprietary operating system. So I mean I guess you can think of it as two choices, right; you can either not produce free or open tools to run a proprietary operating system with the goal of making the free or open operating system more attractive or you can create the tools to run there with the idea of giving them a taste of what the free and open tools can be like in a way of sort of enticing them to move to more free stuff in the future.

That's a tactical decision, right? Which one is going to be more effective and I think that intelligent people can disagree on this in general, and then I think that in specific cases people can disagree as well; and I also think that it's possible that while it might make sense to make--to spend a large amount of time and effort to make for example Firefox run well on Windows right now--that may serve the interest of freedom and the software freedom and open source, but in the future when a large--a much larger percentage of operating systems are free and open it might not make sense at that point.

So I think that it's a tactical decision; it's one that we have to make going forward, but I think that in general I think that it makes a lot of sense because I think that personally in my own advocacy to people who are using non-free tools, being able to say hey, yeah I like Firefox and--and so many people have used Firefox and almost all of them have good experiences with it. I think that for the sort of purposes of promoting free software and open source, I think that Firefox on Windows has been an unambiguously good thing.

JT: You're giving a keynote at OSCON about errors and how you can use them to talk about free software and open source. Can you expand on that a little bit?

BMH: Sure; so it's a little bit of a funny idea. So I've been thinking about--I've been thinking about free and open source software advocacy for a long time and I've also --I'm also a supporter of the Electronic Frontier Foundation and one thing I've noticed about the FSF and the EFF both of which are membership organizations and of people who are doing open source advocacy more generally is that if you look in the communities and the organizations that are sort of doing this advocacy and activism it's almost all technical people right.

And I think that part of the reason that it is all technical people--well there's two reasons; the first is that ultimately what we're talking about with the FSF or at the EFF or even in the open source initiative is about user autonomy; it's about power and control over technology and why that's important. Why is it important you be able to modify yourself or why is it important that you be able to change your software? Why is it important that you be able to share with the people and--and the answer to me is that well technology is powerful stuff, right? If I've got a piece of communication technology that technology is going to define in sort of very explicit terms what I can say, who I can say it to, and how I can say it--powerful stuff. As a result, because the technology is powerful it's important that people have control over it, right; there's politics in technology and it's important that we sort of politicize it.

Now technologists understand that because of course they do technology; they make it so they understand that it's powerful because they do that manipulation and they understand exactly what the capabilities are. Now most people don't see that; but it's actually even worse than that because most people don't see technology at all. Not only do they not understand that it's powerful; they don't see that it exists.

So I may say it's very important that this web-server be free, right and people say what's a web server? And you say oh well it's this thing that exists in there, right? The point is--is that people--they just--they go to a web page it sort of pops on their screen and they understand that there is this thing called the internet--that it brings them there but they don't understand that there is a whole set--whole set of different technologies that sort of run throughout and as a result it--because it's invisible they don't understand that it's something that they might want to have power over.

So a--there's a famous technologist named Mark Weisner who actually talked a little bit about this; he said--he described eyeglasses as the perfect technology, right because you don't see--when you wear eyeglasses you don't see the eyeglasses. You just see the world more clearly. And I think that that's the great example of the way the technology works in many cases. But of course if there's a smudge on the eyeglasses they become pretty visible indeed, right. It's the--it's through errors in technology right; when your web server isn't working you get that error from it. All of the sudden you know that it exists, right; you don't care what operating system your ATM runs until you walk up to it and it has a blue screen of death and you say whoa, my ATM runs Windows. Do I trust my money to it? So in that sense errors provide sort of the--a view of the tip of that iceberg right and the way that we can see technology, and as a result start a conversation about why it might be important that we have control over it, why it might be something that we can modify, we--we should be able to modify, or use and how it might be important to our own autonomy.

Hear more from Benjamin Mako-Hill and other leaders of the Open Source community. OSCON 2008 is happening July 21-25 in Portland, Oregon. Register today!

JT: Microsoft has been making a lot of forays into open source or at least their version of it lately. Do you think this is a genuine sea change in Redmond?

BMH: Well I think--I don't think it's a sea change yet; I think that Microsoft is legitimately interested because I think that they realize that it's something that is becoming increasingly hard enough to--for them--hard for them to ignore and I think that sort of openness and sort of open source, free software licenses in particular is something that they've sort of tried to right and tried to ignore and have realized that that's just not a winning--a winning argument. It also is increasingly important to them that they sort of be able to inter-op with a lot of these operating systems and programs that are out there, so I think that they are genuinely interested in it and I think that--I would describe the state they're in right now is more sort of exploratory.

I've met people at the--at the sort of Open Source Lab in Redmond and they seem to be people who were genuinely interested in understanding the free and open source software community. That said that's not where Microsoft's business is right now and Microsoft isn't going to forget that their empire has been built on development methodology that actually runs completely counter to free and open source software and I think that we still got quite a long way until we see them change in any sort of major, systematic, organizational sort of way.

That said, I applaud the stuff that they've released already under free and open licenses and I wish them all the success in that and I hope that they continue to participate more in the community. The hardest part for them is going to be learning not how to put software out there under free licenses, but how to actually encourage a community of user developers who are working together to help build stuff, which is something that lots of companies have trouble with--in the free and open source software space. And it's something that takes a lot of learning and it's something that will require un-learning some of the ways that they've previously done development. But I wish them all the luck in that and I'd love to support them in the places where they really are promoting freedom.

JT: You've been involved closely both with Debian and Ubuntu, some of the most popular Linux distributions today. What do you think makes them so popular and how do you build a really effective developer community that has such a focus on usability and the end-user experience?

BMH: Hmm; okay so I think that there--I think it's worth talking about Debian and Ubuntu separately because I think they've actually succeeded in a degree of different ways. I think that what Debian has done that other distributions have not is really worked hard to insure institutional independence of the distribution. Debian doesn't feel owned by any single company, any single individual, and as a result, lots of people are willing to contribute and work on it. And the result is that we've built in Debian probably the single largest voluntary free and open source software project. It's been hugely successful in attracting developers and continues to do that.

Ubuntu, I think the smartest thing that Ubuntu has done was a decision made very early on which is--let's be based off of Debian. Let's take the work that Debian has done as this huge institutionally independent space and in whatever ways are possible sort of try to support it and Ubuntu has done better in some places than it has in others in working with Debian productively and to giving back--but certainly better than a lot of other Debian based distributions in the past.

But by working closely with Debian, Ubuntu has been able to take a reasonably small development--paid developer team like dozens of people, early on--one dozen people, and to focus on just the area of sort of integration between existing pieces of software and on usability and on sort of really great user experience. So if that's all you're doing and if you're leveraging the work of the best volunteer community it becomes a lot easier to build something that is really compelling and it really solves people's problems.

I think that the other thing that Ubuntu has done really well is built a very strong user community and made great efforts to sort of stay in sync with people as testers, as bug-fixers, as documentation writers, and as coders, right. Lots of parts of Ubuntu are also maintained by the community as well--not as many people as are in Debian certainly, but because they're working with Debian that's okay, right because it's sort of a good interaction. So by focusing both on the user experience as sort of like the main goal of the project and by trying to leverage all the stuff that's being done elsewhere right--by promoting this type of efficiency, by looking and saying what are the things that Debian is doing well? What are the things that we can do to sort of create something that is very complimentary to it?

I think that Ubuntu has been able to succeed with a reasonably small amount of resources and has built the sort of community infrastructure necessary to insure that it stays in touch with that community and represents them and serves them moving forward.

JT: You probably have more of an inside view of what is going on with the OLPC at the moment than most of us. There seems to be a lot of flux and instability in the direction the project should take with the talk of Sugar OS being ported to other hardware platforms and closed OS(es) like XP being made available on the XO hardware. At the same time, more and more low-end devices are appearing, some running Linux. What is going on with the OLPC at the moment and how can it stay relevant?

BMH: Well so as you say OLPC is in a lot of flux, both the OLPC organization in there's been a few--sort of upper management, you know Walter Bender left and that's been sort of disruptive as I hope you can imagine and there's also been some disruptions in the community in that there's been for example the announcement about the Microsoft Windows being on the system which has made a lot of people in the community upset and has helped sort of create some issues and some problems for a lot of the contributors.

Now I mean it--it's not like--I mean meanwhile, OLPC is expanding the--has--is working only on pieces of software on the free and open source software; everything they're producing seems to be free. There is a growing community--they're doubling the size of the development team for the--for Sugar and for the sort of community based distribution and there's lots of great work continuing to happen there. So in that sense it's a little distracting. I mean I think that--I think that when OLPC sort of launched and went out there it was--we're going to do this; it's going to be all free software; it's going to be all open and I think that a lot of people in the free and open source software community saw this project, right. we're going to get computers into the hands of the kids; we're going to radically change education; we're going to do it all with free and open source software--I think a lot of people saw that as our --whatever, right, like our--this thing that will come forward and win for the free and open source software community.

And I think that with the decision about--with the decision about OLPC, in other words, the decision about Windows I think that to ship Windows as an option it doesn't change the fact that there's a free BIOS on this machine. It doesn't change the fact that all the deployments to date and most of the deployments in the foreseeable future are going to be--only running GNU-Linux and that in the future the versions will be GNU--will be GNU-Linux with a dual boot option for Windows. So if we had started out this way I think that a lot of people would be --would say wow; this is great. This is an unambiguously good thing for free and open source software; but because there has been what seems like a little retreat from a set--from a set of principled positions a lot of people are upset about it. Work continues and I really hope the project sort of gets over this and continues doing the great work that it's doing for free and open source software. And I think that the project should still sort of be universally and uniquely well-suited to doing that and so it's disappointing to see some of the--to see the decision about Windows but more importantly it's disappointing to see the negative impact that it's had on the community and I really hope that OLPC can pull through it as an organization and as a free and open source software project.

JT: One of the most recent projects you've been involved with has been an open voting software project. Can you tell us a little bit about that and why it's important?

BMH: Sure; so I've been involved and interested in free and open source software for quite a while. I've been--and free and open source voting systems for a long time as well--so it turns out that Debian uses a free voting system and I got very involved in it through that--Election Methods. They're--they use a different set of preferential system which allows groups to make decisions based on preference and based on sort of what makes the group most happy as opposed to the person who has the most number one votes.

So I've been interested in this; I've been interested in other Election technologies and at MIT there is a--there was this sort of Election Technology Conference that was held here and I sat through the Conference and towards the end of the day--it was a one-day Conference, I was sort of thinking to myself--man; people have been talking about Election technologies all day and not one thing that's been described here is being used. And the reason it wasn't being used is because it was all universally aimed for Government and State-based decision-making.

And ironically I suppose Governments and States are the last--probably the last likely group to adopt any new Election technology--sometimes for good reason and sometimes for bad reason. So Selectricity and there's a library that I wrote earlier called Ruby-Vote--were projects that were designed to take a lot of the best of this sort of Election technology community and academia and sort of the hobbyist community, right--to take a lot of the best new ideas for voting technology and to make them available not for Government and State-based elections but for absolutely everything else, right. We want something that non-profit organizations can use to elect their Boards of Directors. But more importantly we want something that everyone can use for everyday sorts of decisions, right--for deciding to where we're going to go out to dinner, for deciding when we want to have a meeting time, right; we have a huge amount of research on how to make decisions more democratically and how to vote and work more efficiently and we don't use them because we think of it as something--as elections as something that happens once every four years, six years, whatever, right.

So it's a project that aims to sort of turn that on its head, to create technology--voting technology for everything but Government decisions right--so to create voting machinery for the masses. And that's something that I've had a lot of fun doing and has been I think pretty successful up to date. Lots of people are using it for things like deciding when they're going to go out dinner or have a meeting and also MTV looks like it's going to be deploying it for one of their shows as well, so that's pretty exciting. The reality is that more people vote in American Idol than in Congressional Elections in the States. So having voting reform right is in many ways about getting new technology in front of voters and the best way to do that may not be referendums and gathering signatures; it may be just getting the technology out and getting people using it--at least that may be the first step.

JT: Okay; great. We've been talking with Benjamin Mako-Hill, who is an open source advocate and consultant. He is a Researcher at the MIT Sloan School of Management among many other pies he has his fingers in. And you can see Benjamin this summer at OSCON where he'll be speaking on several subjects. Thank you very much, Benjamin.

BMH: Hey, thank you very much; it was great talking to you.

You might also be interested in:

Popular Topics


Or, visit our complete archives.

Recommended for You

Got a Question?