Sugar Gets Sweeter: Former OLPC Exec Walter Bender on Netbooks, E-books, Blueberry, and Cloudberry

12/14/09Follow @wroush

Every so often, we like to check in with Walter Bender, the former president of software and content for the One Laptop Per Child (OLPC) Foundation. He’s always busy with something interesting—and lately, it’s been Sugar, the classroom-oriented software environment that he and a team of software engineers originally developed for the OLPC’s $200 XO Laptop. Bender left the OLPC Foundation in 2008 to start Sugar Labs, a Brookline, MA-based non-profit organization that continues to make improvements to Sugar.

The most recent, which Bender told us about back in February, is Sugar on a Stick, a version of Sugar that fits on a USB key. Insert Sugar on a Stick into the USB slot of your Windows, Mac, or Linux computer, and you can start up the computer in Sugar instead of the native operating system. The implications are exciting: for the first time, any classroom or consumer with a computer can try all of the educational software built into Sugar without having to obtain an XO.

Last week, Sugar Labs announced the debut of Blueberry, the second major release of Sugar on a Stick. I caught up with Bender just after he’d returned from the Netbook World Summit in Paris, where he says many netbook manufacturers expressed interest in putting Sugar on their devices. I asked him, among other things, for his views on the future of the netbook category, where Sugar fits in, and how Blueberry changes the picture.

I was particularly intrigued by two points Bender made. First, he says Blueberry includes a vastly improved e-book reader that makes any computer running Sugar into “a pretty darn good e-book reader” (his words), with built-in access to the hundreds of thousands of free books available from the Internet Archive and Project Gutenberg, as well as cool tools for shared book annotation. The implications for schools are obvious.

Second, Bender hinted that the next release of Sugar on a Stick after Blueberry, code named “Cloudberry,” will take Sugar in some very interesting new directions, bringing capabilities like cloud-based storage to Sugar users. That ought to make it easier for teachers and students to share documents and applications, for one thing. Click through to the end of the interview for the details.

Here’s the full record of our talk.

Xconomy: You just got back from Paris. What were you up to there?

Walter Bender: I was giving the keynote at the Netbook World Summit. This is the second year they’ve had it in Paris, hosted by Mandriva [maker of a popular version of Linux]. It was a small group of about 200 people, but the right people. A lot of people were there from the various manufacturers, talking about different approaches to netbook software. There was somebody from Google talking about Chrome OS, and somebody from Samsung. I also gave the keynote last year, and I talked then about the difference between computer culture and phone culture, and about how computer culture was going to give the netbook guys an edge. What happened was, I was 100 percent wrong, because Apple and Google wrested the phone away from the wireless operators and the netbook industry just started emulating the status quo. Netbooks today all look the same, and all do the same thing, and the innovation is really happening on smartphones.

So I challenged the netbook community to wrestle back their innovative lead, and to frame it in terms of what you can do with a netbook that you can’t do with a phone. At some level, they are all just computers. But a netbook has a bigger screen, it has a keyboard, and there is a certain level of expression and creativity that those affordances give you that you are going to be hard pressed to do on a phone. A lot of people shoot video on a phone, but not many edit video on a phone. On phones, a lot of people type text messages, but very few people write essays. On a phone a lot of people will play a game, but very few will write a game.

X: So are you still optimistic that netbooks can be a good platform for innovation, and a good market for Sugar?

WB: I do see them as a big market for Sugar, and I see them as a place for innovation. I think the problem is that they let themselves get caught up in thinking about netbooks as doing everything that [laptop and desktop] computers do, only less expensive. But there is some neat stuff going on between projects like Chrome OS and companies like Litl.

X: Litl’s John Chuang just presented at our Cloud3 forum on Thursday. What do you think of them?

WB: When OLPC let go a bunch of people, Litl scooped them up. One of the lead Sugar engineers went to Litl, and one of the designers, and a bunch of the other engineers. I think what they’re doing is absolutely great. And I think there is lots of room for those kinds of innovations. I like to describe Sugar as an example of bucking the status quo. We’ve got a problem to solve around learning and education, and rather than bringing our problem to the existing tools, we said we’re going to make tools to deal with our problem. And there are lots of other problems out there that aren’t education, where people are going to say “we can build x, y, or z for our problem,” and that is where the innovation is going to happen. That’s very different from what you see happening in the smartphone world, where there are a phenomenal number of apps, but most of them are pretty trivial. The netbook industry is talking about doing that—Intel is doing their Moblin app store, for example. I’m not sure an app store is the right way to do it, but it will at least get people excited about the kinds of innovation that Litl did and that Sugar did at OLPC and continues to do.

X: On the theme of Litl, it seems that they’ve decided that the problem they want to solve is media sharing, and they’ve built a tool that’s perfect for that. Is that how you see them?

WB: I think they have a couple of things going. That is certainly part of it. But the whole idea that it’s a cloud device, so you don’t have to worry about loading any software or maintenance or backing anything up, so that there is a certain simplicity to the end user, that is equally important. A lot of the collaboration ideas that we have explored in Sugar fit very nicely into the cloud model, and Litl has taken the ideas a lot further than we have been able to take them with Sugar yet. We’ll see what people are actually interested in doing with the Litl machine, but it’s certainly an example of the innovation we need in that space. But the big question is where things are going with Chrome OS, and what is a Chrome OS computer going to look like. There is an elephant in that room.

X: You just released Blueberry, the latest version of Sugar on a Stick. What’s new and exciting with Blueberry?

WB: Blueberry is the second spin on Sugar on a Stick. The original, Strawberry, was just meant to get the idea out into the world. It was relatively crude, and carried a relatively old version of Sugar. Blueberry has all the latest Sugar bits in it, plus a bunch of additional utilities—everything in Sugar, plus a little bit more. It’s running Fedora 12 [a version of Linux], which just came out. It’s got a utility that allows you to install Sugar onto your hard drive directly; with Strawberry, it was just for Sugar on a Stick, and there wasn’t any way from there to installing Sugar natively, but with Blueberry you can make that transition. That’s important because with a lot of schools, our strategy is to say “Try it, you’ll like it, and you might want to make Sugar native in your classroom or lab.”

The other thing, that’s more of a Sugar thing but it’s in Blueberry, is that we’ve really done a lot of work on our whole e-book offering. We’ve tried to make sure that we have the right set of tools so that you can turn any computer into a pretty darn good e-book reader. It’s great if you can afford to buy a Kindle. But there are some things we do with the e-book reader in Sugar that are in keeping with the whole Sugar philosophy. We’ve added support for the EPUB format, and we’ve refined our support for a bunch of other formats. We’ve built in some library utilities that allow you to access the Internet Archive books and the Project Gutenberg books directly. Then we have a bunch of utilities, like annotating your e-books as you’re reading and sharing the annotations with other people, that are in line with our pedagogical goals with Sugar.

X: There’s always been an e-book activity in Sugar, hasn’t there?

WB: Yes, we’ve always had an e-book in Sugar, but it was never really more than a placeholder. We basically put a wrapper around a technology called Evince, a PDF viewer, and we didn’t do much with it until Blueberry, when we really had an all-out push. The other thing we have done is to put a text-to-speech engine into the reader so that kids can have books read to them.

We also continue to put an emphasis on the notion of writing to read—that one way you learn to read is by writing and having some incentive to read because you’re communicating with other people. So we’ve been trying to expose some pretty neat tools for making multimedia books that were always built into Sugar but were buried or hidden. There is a technology in Sugar called E-toys that was originally done by Alan Kay and his team, and it’s got built into it a very nice multimedia document system. There is a teacher in New York who has been doing a lot of work having kids using E-toys to make documents for a science class. It’s a really rich environment for doing science—it’s the whole idea of a lab notebook and communicating what you’re doing and engaging in critical dialogue.

X: Brewster Kahle from the Internet Archive announced at the Boston Book Festival in October that the archive is making a million of the books they’ve scanned available for kids with XO Laptops. Was this new e-book software in Blueberry intended to take advantage of that?

WB: We’re trying to take advantage of Brewster’s work, certainly, but there are other archives of books that we’re also trying to take advantage of. We’ve been working with Project Gutenberg from the beginning.

X: Microsoft announced this week that it’s releasing its Windows 7 USB/DVD Download Tool, which was originally designed to allow netbook users to create a bootable USB drive or DVD that they could then use to install Windows 7, under the General Public License. What does that mean for Sugar?

WB: It’s good news for us because it means more manufacturers are going to take into account that people want to boot off the USB. The hardest thing about running Sugar on a Stick is wrestling with your system’s BIOS to get it to boot of the USB. You can change the boot order, but that’s a hairy, scary thing for most people. Everybody is going to start making that a lot easier, because they’ll want to accommodate the Microsoft tool.

X: I wanted to ask you about something Nicholas Negroponte at the One Laptop Per Child Foundation told us in our most recent interview with him. We asked him to talk about the advantages of separating the development of Sugar from the development of the XO. His answer was, “Sugar Labs has taken over Sugar and is doing what we should have done in the first place, making it an application not an operating system.” Do you understand what he meant by that?

WB: I never really quite understood that quote. Sugar is a desktop environment. It’s not an operating system. It sits on top of a number of different Linux operating systems, such as Fedora or Ubuntu or Debian or Mandriva. That was always the case. I think maybe what Nicholas was trying to say was that when you are going to make hardware, you have to worry about things like drivers, and a lot of the software engineering that OLPC had to do involved things like that, not Sugar per se. But a lot of that engineering had to happen whether you were going to run a vanilla Linux desktop or Sugar.

X: My interpretation was that he meant that he wanted the XO to support many different operating systems to increase its appeal to potential partners, and therefore it was important that there not be too much identification in people’s minds between Sugar and the XO.

WB: That could be what he meant. But that is pretty independent of the operating system. Right from the beginning, you could run Ubuntu on the XO. Somebody ported Ubuntu to it long before we ever shipped it. And Windows on the XO was going to happen eventually, if and when Microsoft put the effort into it.

X: Have you been involved at all in the planning of what Negroponte calls the “model 1.75″ XO Laptop, or the “model 3.0,” which he said would resemble a sheet of paper?

WB: I don’t know anything about that 3.0 machine at all. But I know a lot about the 1.75 machine. It’s using a faster Via processor. It’s a nice machine, and it runs Sugar great. It has the same form factor as the original XO, but they’ve upgraded a couple of components, most notably the touchpad, which had been fairly problematic in the original machines. We are involved in that, and I’ve also been helping them on a couple of the little aspects—for example, I had done all of the keyboards for the original XOs, so I am doing some keyboard work for them on this new machine.

X: Do you still think of the XO as the “reference” design for Sugar—the one machine that it’s most perfectly suited for?

WB: I’m thinking about it less and less that way. The XO is still our biggest deployment. But if I had to give a rough guess, I’d say that 10 percent of the instances of Sugar are now running on machines that are not XOs. And that number is going to grow rapidly. When I went to this netbook summit, I had a lot of interest from every single one of the netbook manufacturers about having Sugar be part of their offering. It’s easy for them, and it affords access to a different market segment.

X: Are you saying companies like Samsung might pre-load Sugar on their netbooks?

WB: There’s no reason why they couldn’t. It’s just a matter of whether they want to. But even if it’s not pre-loaded, it’s so easy to do, especially with the new facility in Blueberry for loading Sugar onto the hard drive directly.

X: Sugar Labs recently announced a partnership with a USB key duplication company called Nexcopy, where they’re going to take used USB keys and put Blueberry onto them. Is that part of your plan for expanding access to Sugar?

WB: That’s not so much about expansion as it is that there are some places where paying even $5 for a USB key is a lot. So this is really more in the spirit of trying to reach the have-nots. We’ll be bulk-loading these USB keys with Sugar and shipping them off to schools. We’ve already had quite a few schools asking about it. We don’t have enough in stock yet to make it worthwhile, but it’s a very easy, very low-cost thing for Nexcopy. At worst, we’ll be able to get these things out for the cost of mailing.

X: What else have you got up your sleeve at Sugar Labs these days?

WB: Well, one teaser I’ll give you is that the working name for the next release of Sugar on a Stick is “Cloudberry.” It’s a wonderful berry that grows in the Lapland region of Finland—a little orange-colored raspberry that they use to make a really nice liqueur. But as the name suggests, we’re really going to try to beef up a lot of the cloud features in the next release. Part of it would be things like using the cloud for storage, and part of it flips the other way, having the cloud locally on your machine, using Google Gears-like stuff. We have a bunch of extensions to Sugar already that we haven’t put into the main release that allow you to move your work offline and then redistribute it–so for example a teacher could go find something online at home, then bring it into the classroom, and have the kids work on it, even though it was originally a Web app. I had a Google Summer of Code student working on that last summer, and its been getting closer to being ready.

X: When will Cloudberry be ready for release?

WB: I think it will be in the March or April time frame.

Wade Roush is a contributing editor at Xconomy. Follow @wroush

By posting a comment, you agree to our terms and conditions.