Your host is Benjamin Mako Hill who graciously allowed this re-publishing of his original post from Copyrighteous for OLPC News.
In the last week, Nicholas Negroponte gave this unfortunate interview decrying "open source fundamentalism" and hinting the possibility of a warmer relationship with Microsoft. Predictably, this has elicited an ongoing response by OLPC News and on the OLPC development mailing lists.
Ben Mako Hill and XO laptop
Just a few days before Negroponte's statements hit the press, I gave a talk at Penguicon called Laptop Liberation where I talked about why I thought that OLPC's use of a free software operating system and embrace of free software principles was essential for the initiative's success and its own goals of education reform and empowerment. I've been saying similar things for some time.
My main point boiled down to something that, appropriately enough, Nicholas Negroponte was fond of saying back when the project was still called the $100 laptop: an extremely cheap laptop is not a matter of if, but of when and how. This technology will define the terms on which students communicate, collaborate, create, and learn. These terms are dictated by those with the ability to change the software -- by those with access to computers, the source necessary to make changes, and the freedom to share and collaborate.
Constructionism -- OLPC's educational philosophy -- is about putting powerful tools and control over those tools into the hands of learners. It is about learning through exploration and creation -- about shaping one's own educational environment. Constructionist principles bear no small similarity to free software principles.
Indeed, OLPC's stated commitment to free software did not happen by accident. OLPC convincingly argued that a free system was essential for creating a learning environment that could be used, tweaked, reinvented, and reapplied by its young users. Through these processes, the XO becomes a force for learning about computation and an environment through which children and their communities can use technology on their terms and in ways that are appropriate and self-directed.
We know that laptop recipients will benefit from being able to fix, improve, and translate the software on their laptops into their own languages and contexts. Much more importantly, however, are all of the uses for the laptops that OLPC has not -- and can not -- think up.
A future constructionist learner
OLPC is a powerful tool for learning, but ultimate power is only in the hands of those that can freely use, change, and collaborate in defining the terms of their learning environments. In its commitment to software freedom, OLPC chose not to be arrogant by assuming that it knows how its users will use their laptops. Flexible environments designed for constructionist learning and a free software platform protect against this arrogance.
Constructionism and free software, implemented and taught in a classroom, offer a profound potential for exploration, creation, and learning. If you don't like something, change it. If something doesn't work right, fix it. Free software and constructionism put learners in charge of their educational environment in the most explicit and important way possible. They create a culture of empowerment. Creation, collaboration, and critical engagement becomes the norm.
OLPC does not get to choose if educational technology happens. If we work hard at it though we might get to influence the "how" and the "who." Proprietary software vendors like Microsoft want the "who" to be them. With free software, users can be in power. What's at stake is nothing less than autonomy. We can help foster a world where technology is under the control of its users, and where learning is under the terms of its students -- a world where every laptop owner has freedom through control over the technology they use to communicate, collaborate, create, and learn.
This, to me, is the promise of OLPC and its mission. It is the reason I've been involved and in support of the project since nearly day one. It is the reason I left Canonical and Ubuntu to come back to school at MIT to be closer to the then nascent unincorporated project. It is the reason that OLPC's embrace of constructionist philosophy is so deeply important to its mission and the reason that its mission needs to continue to be executed with free and open source software. It is why OLPC needs to be uncompromising about software freedom.
As an adviser and sometimes contractor to OLPC, OLPC does not need to listen to me. But I hope, for all our sake, that they do.
Update: Richard Stallman and the FSF have published another essay on the same topic focused more on pure free software (i.e., less education specific) objections.
Your host is Benjamin Mako Hill who graciously allowed this re-publishing of his original post from Copyrighteous for OLPC News. Hill is an OPLC adviser and contractor and a director of the Free Software Foundation.
There is no such thing as 'free software'. Somebody has to give up their time to produce it. Of all the stupidities promoted by the FSF, the idea that software can be free is the dumbest. Software is free for X only because it is paid for by Y who produces it. It is never 'free' simply in and of itself.
The OLPC project has, from the start, sailed under a flag of convenience called 'constructionism' which effectively means that some other unpaid person is supposed to do the heavy lifting wrt to designing and planning the educational software tools so essential to the success of the project. It is a mark of the low esteem in which the applications programmer is held, that almost everybody involved in the project is drawing down a salary of some sort, but this work is supposed to appear for 'free' i.e. paid for by somebody else. This marks the OLPC project as a laptop project, not an education project.
Mark, you seem to be under the misconception that free in free software refers to price. It refers to freedom. I used the term "software freedom" in my article several times to help make this clear.
Obviously, the production of software is not without cost. That's the reality of technology production and cannot be a condemnation of the project.
Mark,
Odd that you have a tirade on free (as in beer) software, you link yourself to Lambda Assoc - providers of "advanced open source Lisp freeware".
I, for one, am very much a fan of paid Open Source software - I donate to or pay for what I use as I agree with the idea that money motivates. And the last time i checked, there are a number of people getting paid for their educational work on OLPC, though I will agree that there isn't enough focus on the educational aspect of the program.
Trying to separate the reality from the punditry, here all the quotes from the article:
Walter Bender: "Sugar is in a narrow place and it is ripe to be unleashed."
Negroponte (on Sugar): "grew amporphously" "didn't have a software architect who did it in a crisp way"
Negroponte (on Flash?): "there are several examples like that, that we have to address without worrying about the fundamentalism in some of the open-source community".
Negroponte (on open source): "One can be an open source advocate without being an open source fundamentalist."
@Charles:
"Negroponte (on open source): "One can be an open source advocate without being an open source fundamentalist.""
Would you say the same about Freedom and the Founding Fathers?
Those were Freedom Fundamentalists, weren't they?
Winter
I've never been a fan of OLPC for many reasons, most of them because of the fundamentalist attitudes and the close mindedness of much of the project and community.
It's refreshing to see someone looking at this project as a problem and fitting a solution to it, instead of having a solution and finding a problem for it.
I can't imagine that the kids receiving these laptops care at all about FSF or OSS or 'free as in beer' or 'free as in speech'. (I assume)They want a laptop, and one which facilitates learning. Why not just give them that, without all the BS fundamentalism in the way.
Mark
"Constructionism and free software, implemented and taught in a classroom, offer a profound potential for exploration, creation, and learning. If you don't like something, change it. If something doesn't work right, fix it. Free software and constructionism put learners in charge of their educational environment in the most explicit and important way possible."
I like this concept. However, what we (I'm a developer, too) distribute on XO does not provide this, actually. There is no way to "fix it", if it is GTK, X, kernel, etc on XO. In practice, the same is true for the stuff written in Python, as the developer uses other tools that are not available on XO. I feel that saying that "you can fix it" a bit unfair, as who says so isn't doing it on the equal basis.
Free software doesn't automatically put learners in charge. Properly architected free or open source software is needed.
(Incidentally, Etoys comes with its own development system and source code. There you can "fix it" almost everything on XO.)
Mark,
You are so right. All most of us want to do is get the XO into the hands of the kids and let them get on with their learning. I wish that NN and Brightstar had that attitude.
Their attitude and thinking is probably going to kill the project. They want to market under their rules and their time schedule. Customers get lost. I feel sorry for them. The kids are waiting. By the end of the year thousands of XOs could be in the hands of kids except for HH and Brightstar. It is going to be interesting to see how many will be in the hands of kids using HH's marketing rules.
I am sorry two mistakes are in my email above. HH should be NN. I should proof read.
"(Incidentally, Etoys comes with its own development system and source code. There you can "fix it" almost everything on XO.)"
And also, and not by accident, Squeak/Etoys/Scratch are available on Windows and other platforms - the creators of these, arguably most powerful educational tools available today, not only gave the kids "freedom" to "fix it" but also, and equally importantly, "freedom" to use them in the first place - places where, for whatever the reason, only Windows platform is available to them. There are millions of kids right now who would be 'locked out' from using these tools if the creators insisted that, in the name of the "freedom" of course, they will run only on Linux base (and a particular variant --Fedora base -- at that). Few people contributed more to practice of Constructionism than Alan Kay and others who developed those tools and for them kids always come first...
Hey Everyone,
Thanks for writing this Mako - powerful and convincing.
To those that think a belief in freedom is nutty fundamentalism, I think it is crucial not to fixate on today's static embodiment of durable culture. Instead concentrate on the ecology of this project - the software, the community, and the processes that bind them - that are supposed to enable (not impose) an educational transformation over space and time.
I elaborate this position in an essay I wrote over a year ago - Free Laptops: Creating, Producing and Sharing a Revolution http://pocketknowledge.tc.columbia.edu/home.php/viewfile/18259
"But how can we be confident that the OLPC will result in positive changes to the
social infrastructure where it is deployed? Perhaps it will merely capture and
export the worst features of our society. Will children’s innate curiosity motivate
them to crack open the hood of the laptop and master its inner workings? Or, will
it simply degenerate into the instrument which delivers spam and pornography
to the third world? Will teachers embrace the platform and craft their curriculum
to take maximum advantage of this device? Or, will they ask students to store the
laptop in a cubby for the duration of the school day? Will this technology improve
the ways children interact with each other, and the way they relate to information
and knowledge? Or, will it promote consumerism, competition, obesity, and
depression? If we don’t understand these dynamics in our own society, how can
we pretend to understand them as we unleash this technology upon rest of the
world?"
Our best shot at this projects success necessarily involves reorienting the political economy - nothing less. Otherwise this project will likely backfire and expose billions of the world's children to many of the worst aspects of our society.
Free software is integral to the larger educational mission of this project, and its values percolate through every layer of the project stack - hardware, software, and community. Every citizen on this planet has the right to control the means of their education, expression, and cultural production. There is no way to separate this effort from its larger political context.
I guess I am proud to be an educational fundamentalist. I actually thought this project was set up as a non-profit to allow for idealism to drive decision making. Luckily, Sugar is truly free and its legacy will long outlast the current leadership's short-sightedness.
OLPC is supposed to be an educational project - what exactly does NN think he is teaching us now?
If the laptops were 'free' this wouldn't be an issue, every country in the world would accept them. The problem is governments don't want to buy laptops with Sugar. The customer is willing to buy XO with XP (or so NN thinks).
Maddie wrote:
"If the laptops were 'free' this wouldn't be an issue, every country in the world would accept them. The problem is governments don't want to buy laptops with Sugar. The customer is willing to buy XO with XP (or so NN thinks)."
That's exctly the problem here, Maddie. I have said it many times before. Negroponte is in a no-win position: he can't please the Microsoft haters, because pleasing them means that he delivers a product nobody wants (except for the haters, and there are not enough of them, as we know).
The XO had an opportunity to be seen and operated without Windows, without Intel, without intervention of any of the "big, bad boys". people came, saw and said "No, thanks!"
Why?
Yoshiki, a full development environment necessary to build, modify and fix, any single package is fully installable with 1-2 commands. Each laptop can (AFAIK) modify any piece of software (with the exception of the wireless firmware) from the laptop. Because this still requires an Internet connection, full source for *everything* on the machine is being included in every school server build distributed alongside laptops.
You're right. That's not at the same as eToys or, for that matter, the majority of Python code on the machine that's in the situation. We're also missing lots of documentation in appropriate languages to help with this. But it *is* a fundamental difference in terms of realizable potential over the alternatives like a Classmate, a mobile phone, or an XO running windows.
> If the laptops were 'free' this wouldn't be an issue, every country in
> the world would accept them. The problem is governments don't want to buy
> laptops with Sugar. The customer is willing to buy XO with XP (or so NN
> thinks)."
Maddie - 'free' as in 'freedom'. The OLPC is currently free, but the green machines may not be for much longer. America is free, even though we all pay taxes (no snickering - this is a rhetorical point). This isn't about money, its about commitment to ideals. Consider the educational ones for a moment - what is the educational mission of this project? What values (or meta-values) is it meant to embody and express?
Perhaps the problem is not that governments don't want to purchase the XO w/ Sugar, its the the OLPC organization has been inflexible about its distribution models. Grassroots movements are bottom-up, not top-down. There is plenty of demand for the XO's with Sugar, just not (yet) at the governmental level. So what?
Partially thanks to OLPC popularizing these approaches, plenty of other corporate forces will be providing the big boy's take on educational technology. The XO can succeed to, on a realistic time frame, provided it commits to its principles.
uh oh,
again that pattern,
"...students communicate, collaborate, create, and learn. "
"Constructionism ... for exploration, creation, and learning."
"[Constructionism] is about learning through exploration and creation"
Look like when constructionism is in, learning is the tail end, a by-product, not a goal in itself.
Sometimes can be bypassed altogether,
"Creation, collaboration, and critical engagement becomes the norm."
Would it make sense to invert the order and say
"Constructionism promotes learning, creation, and collaboration"?
Of course not.
Dunno why we call them still "students".
Worries me.
For the record: I do not hate Constructionism. I love it. I think it's Great - as long as it's part of a plan that also incorporates proven methods that actually teach skills and competencies, and those methods are not just an afterthought.
I also find interesting how the Constructionist crowd today seem to be so happy to bite the hand that fed them. Without NN's commitment to Constructionism, they would still be an interesting theory among those whacked on the head of teachers everywhere. I admire NN in many ways, but that former allegiance, and the XP thing, of course, temper my enthusiasm for that otherwise brilliant man.
Mako,
"Yoshiki, a full development environment necessary to build, modify and fix, any single package is fully installable with 1-2 commands. Each laptop can (AFAIK) modify any piece of software (with the exception of the wireless firmware) from the laptop. Because this still requires an Internet connection, full source for *everything* on the machine is being included in every school server build distributed alongside laptops."
I was told that compiling some big compontents run out memory on XO (and again the stuff you can install with 1-2 commands wouldn't cover the tools that the developers actually use.); but it might not be the case when doing it from the console or such. (And, there are compilers with smaller memory foot print.) And you could maybe say that kids could do it on the school server. Computer is like its own mathematics and everything is *possible*.
But, if the effort to exercise the right and freedom is prohibitingly high, that is not too different from not having it at all. It is like the shortage of the voting booth in some minority concentrated areas in the US; "yes, of course you can vote, but wait in the line about 8 hours."
We are not in big disagreement; kids should be exploring the environment fully and change/break it to learn. But this is not in practice not true in terms of OLPC XO. When what we provide isn't so inviting, just saying "it is fundamentary fixable" doesn't tell the whole story.
Yoshiki, If you know of a single piece of software shipped on the XO that cannot currently be built on the XO, I consider that a bug and something that should be fixed. Please tell me and I'll file the bug. I do all of my OLPC development (for core shipped software, no less) on my XO and I don't know of any such software.
Benjamin, thanks for the information! I had the same impression as Yoshiki and felt that the lack of native development capability was a serious limitation. Not "eating your own dog food" can have a negative impact in program usability - at the Game Jams I was present the apps were only tested on the XOs at the last minute and of course the performance was not as nice as on the desktop development machines.
But I started out with Linux compiling the whole kernel on a 4MB machine, so what you say about it still being a resonable thing to do on the XO itself makes perfect sense. GCC and the kernel have grown quite a bit over the years, but a 256MB machine should still be up to the task.
I'm curious to know how you are doing it^^; I'm too young to only know when my PC with 16MB of memory compiled 386BSD. However, even my sugar-jhbuild directory doesn't fit on XO so I'd still think it is not simple.
Let us make it so that it is in the kids' confort zone (like it co-exists with the regular Sugar environment, and really fix itself, etc.)! Then, the statment "you can fix it" will mean much more.
Like someone has said, the OLPC is modern equivalent of sending Bibles to poor. Or... will they be preloaded on them?