January 4, 2007
I've largely been ignoring Nicholas Negroponte's One Laptop Per Child initiative. I appreciate the nobility of the gesture, but how interesting can sub-$100 hardware running Linux really be? Well, that was before I read about the novel user interface they're building into those small green and white laptops.
For most of these children the XO machine, as it's called, likely will be the first computer they've ever used. Because the students have no expectations for what PCs should be like, the laptop's creators started from scratch in designing a user interface they figured would be intuitive for children.
The result is as unusual as -- but possibly even riskier than -- other much-debated aspects of the machine, such as its economics and distinctive hand-pulled mechanism for charging its battery. (XO has been known as the $100 laptop because of the ultra-low cost its creators eventually hope to achieve through mass production.)
For example, students who turn on the small green-and-white computers will be greeted by a basic home screen with a stick-figure icon at the center, surrounded by a white ring. The entire desktop has a black frame with more icons.
This runic setup signifies the student at the middle. The ring contains programs the student is running, which can be launched by clicking the appropriate icon in the black frame.
When the student opts to view the entire "neighborhood" -- the XO's preferred term instead of "desktop" -- other stick figures in different colors might appear on the screen. Those indicate schoolmates who are nearby, as detected by the computers' built-in wireless networking capability.
Moving the PC's cursor over the classmates' icons will pull up their names or photos. With further clicks the students can chat with each other or collaborate on things -- an art project, say, or a music program on the computer, which has built-in speakers.
I'm interested now.
I've been disappointed in the lack of GUI innovation over the last decade. Sure, Microsoft and Apple take small jabs at each other every couple of years. And the Linux community apes both companies, occasionally throwing in a curveball of their own. But when was the last time anyone tried a radically different UI on the desktop? The Sugar UI featured in the OLPC appears to finally break from the well worn conventions of Windows and MacOS.
I wanted to try it out myself. I downloaded the emulated OLPC laptop image and ran it under QEMU. The documentation even warns you to prepare yourself for this alien UI experience.
Before you launch the emulated image, we strongly recommend reading through the Sugar Instructions on how to use the environment -- this does not look like the Windows or Mac operating systems!
They weren't kidding. It's nothing like any traditional GUI.
I was inclined to like Sugar almost immediately because it embodies a number of experimental GUI concepts I've talked about before:
Sugar UI development appears to lag quite a bit behind the challenging, sub-$100 design goal of the OLPC hardware itself. This doesn't surprise me, because developing UI is hard. And developing a radically different UI has to be especially difficult. Innovation and experimentation is much riskier than following the roadmaps from Redmond and Cupertino. That's why, despite the rough edges, I'm excited about Sugar.
The Sugar instructions offer an excellent basic overview of the UI, with many more screenshots. If you're a designer, check out the Sugar UI design guide. There's also a video walkthrough of the Sugar UI available.
I have to admit that I didn't find the Sugar UI particularly intuitive or discoverable, even after using it for 10 minutes and learning the basics. But I'm not a child. Maybe something unusual is necessary to get kids' creative juices flowing. Mr. Negroponte has strong feelings on this topic:
In fact, one of the saddest but most common conditions in elementary school computer labs (when they exist in the developing world), is the children are being trained to use Word, Excel and PowerPoint. I consider that criminal, because children should be making things, communicating, exploring, sharing, not running office automation tools.
He's got a point. I don't know many kids that want to grow up to be "Information Workers".
Posted by Jeff Atwood
Speaking as the parent of two young children (5 and 2 years old), I think the Sugar UI would work very well for them--*both* of them. To a large extent, it already matches many of the computer games they play on my PC and on their GameBoys. (Yes, the 2-year old has a GameBoy of her own.) The 5-year old can't read the instructions for all his games yet, so *all* the games he plays have mystery meat UI. But he loves playing them anyway. He tries things out until he figures out what works, and then remembers what the symbols mean. From then on, he experiments and explores how those symbols can be used in different contexts.
The important thing that the OLPC will introduce to kids who have never used a computer before is the very *concept* of a UI that allows them to manipulate objects in an abstract space. Once they have that, picking up Windows or OS X is just a matter of extending concepts they already have, rather than introducing new ones.
I think the amount of "useful work" a 5-year old will be able to do on a Sugar-based OLPC is going to be substantially greater than the amount they would be able to do on a MacBook loaded with OS X. That's "useful work" in a 5-year old sense, of course. I don't think anyone's suggesting that the Sugar UI should be adopted by web programmers or office workers around the world. (Hence the "One Laptop Per *Child*" name.)
'I appreciate the nobility of the gesture, but how interesting can sub-$100 hardware running Linux really be?'
Wow.... Is running Windows on sub-$100 hardware more interesting?
Is there any information wheter Sugar's comming to a linux distribution besides of OPLC? I'd really like to try this on Ubuntu or something...
Transhumanist: recall that the population of the Earth is roughly 6 billion people. Social skills are useful, and are only going to become more useful as time goes on. Don't make "hyperbolic generalized assertion[s]", or at least implications, of your own.
My concern is that whilst the interface is innovative - Why should we teach and encourage children to interact by computer when they are NEAR each other? Surely children still have things to learn about collaboration at the sand-pit, sharing a painting project over a single large sheet of paper, and so on? Or whatever equivalents they might use around the world.
There is an amusing and poignant blog entitled "7 Reasons the 21st Century is making us Miserable" by David Wong. (http://www.pointlesswasteoftime.com/misery.html )
I for one can honestly say that a fascination with the silicon world hampered my human interactions, even though I only used computers from my teens. The internet has allowed far more interactions with real people, but it still does not compare to the real world. And I for one have decided never to touch WoW or Second Life because, well honestly, I think I (and a good few other people) should really try and concentrate on making the First Life a bit better :)
While an avid user of ratpoison, I really wouldn't call it new or innovative. It has really just taken the old UNIX utility screen and made it work in X not just text consoles.
Usable, great bit of software, but not really innovative.
"But when was the last time anyone tried a radically different UI on the desktop?"
When and Where you searched for !?
This is like Windows, OS X, Gnome, KDE ?
Of course ratpoison has a different use case that Sugar, but it's innovative, don't ?
Computers are an integral part of our society nowadays and I would disagree that it is likely to impair children's social interactions. Why not let them learn about collaboration in the sand pit and with a painting project, while at the same time they can use their laptops as just another tool to learn. At some point they're going to be forced to use a computer, why not make it something familiar?
I do wonder, though, about the logic of teaching children to use a computer using a completely unique user interface. Sure, current UIs have tended to be based on whatever was easiest to conjure up and what was not too different from what everyone was used to. The solutions we have now are far from the best or most intuitive, however they are the solutions we have. Surely the children are going to have to learn to use a real-world user interface sooner or later and as much as we might not like the current state of UIs, how likely is it that we will witness any significant change in the near future?
While I fall into the category who say, "teach kids to think well enough to see through The Shrub; not be really good at video games"; I've marveled for decades that the structure of GUIs, at least mainstream ones, are rehashes of physical devices. Radio buttons for pity's sake. Sugar seems to have started from the two basic principles: a pointing device and a bunch of pixels.
"Largely ignoring" he says.
I'm largely ignoring 99.999% of everything that happens. If I wasn't, I'd be autistic. I'm not singling OLPC out here; it goes in the same discretionary bucket as everything else.
Is running Windows on sub-$100 hardware more interesting?
Commodity sub-$100 hardware isn't interesting, and using off-the-shelf commodity software of any type, open-source or otherwise, certainly isn't interesting. At that point, it's all a question of cutting and packaging.
But the Sugar UI is real innovation, real RD and thus quite interesting. I'm still quite concerned that they're not doing usability testing of any kind on this experimental work-- round up some third-world (or even second-world) kids with no computer experience and give them a task, an OLPC prototype, and see what happens. Until they do this, they're designing in an ivory tower, and they're probably destined to fail.
The idea that a UI has to be "simple" for kids to use it was destroyed by that guy who put a computer in a slum in India. It was a Windows machine and they took to it pretty much immediately. He didn't even hook up a keyboard and they figured out how to enter text on it! Kids figure out games like nobody's business, too, even when the UI varies from game to game. They're curious and love to explore, at least until they get that beaten out of them in high school, and they'll figure out anything if they have a good enough reason to.
Which basically means that it almost wouldn't matter what UI they put on the OLPC, and the lack of user testing is probably not a biggie, but it also means that the OLPC designers have the opportunity to try something new, and I salute them for grabbing the chance by the horns.
Is it just me, or does this UI remind anyone of the student tablets in "Ender's Game"?
Greg, I did not mean to imply that children should be prevented from using computers, but rather that 'we' (developers) should be careful what we let the user to do - especially as they are children. I see the 'Buddies' feature of the Neighbourhood to be the first step on a slippery slope; I've just finished a blog on the topic here (with speculative pictures of potential software enhancements - ooh aah): http://www.kebabshopblues.co.uk/2007/01/06/human-interaction-and-the-sugar-ui/
The whole subject of PC's targeted to school children is not one I'd really though of before. The wireless aspect raises the bar even more, as does the physical location aspect I mention in my blog. Even though the network is wireless, in a school environment some sense of physical location is probably useful... not to mention security as presumably we do not want 'undesirables' sitting outside the school gates yet virtually 'inside' the school because they happen to have obtained an OLPC laptop?
On the other hand, it is very difficult for me / us(?) to understand how these machines will be used in different countries, especially ones that are not first-world - so perhaps I just muddy the waters by thinking about physical location and security and sand-pits?
Everytime I have seen an article about the OLPC I have said the same thing, and for some reason I usually get slammed for it.
Anyway, it's all fun from our perspective to learn kids a weird interface that no professional computer will ever use but do they learn how to use computers ? Do they prepare for the real world ? Will they get jobs if they say "I have experience with OLPC" ?
The Windows / Mac / Linux user interface is standard, you shouldnt teach kids an awkward toy like user interface.
Just watched the video. It's too bad that it's just a UI and not something more ambitious. I can see a lot of non-intuitive features of the underlying OS shining through.
Why should there be a difference between starting a program and switching back to a program already running, for example? Why should they have to figure out how to close applications?
Why should a child have to figure out that they need to save in Abiword? Just always save the current document and let them switch documents when needed.
Congratulations on being so very Web 2.0.
If you have "largely been ignoring" Negroponte's One Laptop Per Child initiative because you dismiss it as "noble but not interesting", you are blinded by your specialization and your ego. How fortunate that OLPC has survived the crushing blow of being largely ignored by you.
Negroponte's team is doing something real for the people of this world. You are writing an oh-so-vital blog. One may make mistakes but is yet both visionary and aiming to do practical good in a corpocratic world.
The other takes almost no effort or discipline, is without any qualitative measure other than total number of visits by Digg zombies, and achieves nothing. Very Web 2.0 indeed.
This project is very interestings, but I have to wonder if education will really improve just because the kids have a laptop computer. I love computers as much as anyone, but there is a difference between learning how to use a computer and learning through the use of a computer. The first situation is imperative today, the second is very often wishful thinking. Every few years I fall into the trap of thinking that if I buy a new PDA I'll somehow become organized. It never works. PDAs are really just toys for me that always disappoint after the novelty has worn off. I'm curious, do we think laptops are helpful to education, or do we all just like toys?
ISurely children still have things to learn about collaboration at the sand-pit, sharing a painting project over a single large sheet of paper, and so on? Or whatever equivalents they might use around the world./I
Please refrain from writing "the" with the assumption that the noun being described is integral (in this case, a sand box/pit). Sand pits aren't much less artificial than computers, it is only likely that you have become so accustomed to seeing them that you associate them as a necessary part of social interaction. The notion that children should be socializing and interacting with other children is only useful if one wants to produce children that are devoid of intelligence; any child who learns how to reason objectively will not need much social interaction, as social interaction only furthers one's understanding of cultural subjectivities. I can only assume that parents want their children to interact with other children for the selfish cause of producing children that are more like they are.
"He's got a point. I don't know many kids that want to grow up to be 'Information Workers'."
Most kids don't dream of being mathematicians, either. Or historians. Should we not teach children math? History? Literature?
I'm trying hard not to blow a gasket over Mr. Negroponte's statement... the idea that anyone could visit the developing world and then use the phrase "one of the saddest conditions" to describe the use of Microsoft Office. Seriously.
It sounds... well, it sounds like something someone wealthy would say about poverty -- a sort of "Let them eat cake" moment for the information technology field.
Nij--more likely your human interactions are hampered by using the word 'whilst.' 'Whilst' makes people want to punch you in your mouth and call you a fucking nerd. :/
there are tens of thousands of noble causes at any given time, will you next castigate Jeff for not dedicating a blog entry to each and every one of them? Only so many things can interest us at once.
I believe the statement was simply offered as a reason it had not come up before. I can't see how such a statement could be taken as mockery of the project, nor how Web 2.0 and mockery of noble causes has become conflated, nor why you respond by mocking the blog in turn.
Whilst is a perfectly normal Britishism. This isn't an American-only blog. ;)
This whole thing seems pretty patronizing. I'm sure that any 8 year old kid could use the same interfaces I do to work with the computer. Sure there's a lot of room for improvement in those interfaces, but I think that's where we should be spending our time, not on some untested, half-baked UItard.
And I'd like to thank you, vic, for spending your time writing that visionary comment in this oh-so-vital blog. Very Pretentious Jackass 2.0 indeed.
vic, lighten up. WIthout the blogs and "web 2.0" resources you're so quick to dismiss, we'd be missing huge informational and collaborative resources, which provide much of the very educational value of the OLPC you love so much in the first place. Sure there are a lot of useless blogs out there, but this is not one of them.
However, a potential downside of OLPC that just occurred to me is that we'll suddenly have a lot more adolescents on the internet, making comments like vic's. I guess we survived the AOL floodgates, we'll survive this too ;)
He's got a point. I don't know many kids that
want to grow up to be "Information Workers".
Nor do I. The sad bit is how many teachers, school administrators, and politicians openly state that's exactly what they want. :/
I never really comment here, but the yelling in my head won't let me click 'close' after getting to the bottom.
A point I think need sharp underlining is the impact that web resources such as wikipedia and google could have on the impoverished (and much of the developed) world. Maybe the UI poke-and-prods are up for grabs, but some crucial technologies that would benefit the third world are glaringly obvious to those who look online.
Imagine the boom a third-world farmer could reap in googling 'Crop rotation' or 'how to double my crop output'. Think of a struggling working-class revolution with access to U.S Army field manuals (globalsecurity.com) or maps (Google Earth). Think of how much of that fighting may not be necessary with access to dreams such as the U.S. Constitution, the Magna Carta, and the latest advancements in the study of game theory (John Nash, not Nintendo). I speak three languages and am learning a fourth, mostly through the power of online info/resources.
I think the global potential for progress would explode given instant online access to those still struggling with lack of power, water, and food- hands down.
Good article, but ... I'm rankled by the arrogance of bloggers.
"Largely ignoring" he says.
Smacks of the mom's basement movie maverick at my local rental place who haughtily and loudly dismissed "The Matrix" (the first one) to attempt to claim some podium of elite film appreciation by dissing something that was generally and popularly held to be a blockbuster.
Not at the author, but in general: Thanks to the net, dorks can feel like they're standing shoulder to shoulders with the Negroponte's of the world. Then mom comes in and bursts the bubble threatening to turn off his computer until his chores are done.
Hmm... this GUI suffers heavily from mystery meat navigation: http://en.wikipedia.org/wiki/Mystery_Meat_Navigation
I know the X is probably close, the two squares is probably minimize or maximize, and the spyglass is search, but the rest of the icons are a complete mystery.
The children are already being taught too much reliance on the computers and it ~is~ affecting their interaction with each other. Although computers have their place in school, I am not sure that the lower grades are the appropriate grades. This should be the time to learn interaction, people skills, and tactile skills. We as a society are ending up with people that are incapable of doing math without a computer; we are "forgetting" how to spell or even write by hand. People are forgetting how to deal with one another on a personal level. How often have you seen two or more people walking together and are all on cell phones to other people or even each other!
The other issue. Teaching a student how to use the "office automation tools" isn't neccesarily training him/her to be an "Information worker" as much as it is training the child to use programs that they will in all likihood use their whole lives (in particular in high school and college).
That said, I think this GUI looks quite clever. A whole new concept that seems like it might also have great potential in third world countries.
"I am not sure that the lower grades are the appropriate grades. This should be the time to learn interaction, people skills, and tactile skills."
What "people skills"? The only thing I learned from lower grades were basic things like arithmetic and basic spelling, which were redundant and dumbed-down. Instead of engaging my brain, I had to be forced to be patronized and "play" for the waste of time that is "recess". Learning "people skills" is equally an equally imagined notion as "street smarts", and contributes to the dumbing down of society by having people believe that there is something else that people who are obviously not intelligent can make up for by having. Such egalitarian ideals only contribute to regressing society into retrograde subhumanity.
"We as a society are ending up with people that are incapable of doing math without a computer; we are "forgetting" how to spell or even write by hand."
Do not use "we"; I am not a part of your hyperbolic generalized assertion.
Mystery meat is easily internationalizable. And what about kids who are still learning how to read?
But when was the last time anyone tried a radically different UI on the desktop?
To my knowledge, Chuck Moore around '99-'01.
I think most anyone would say that colorForth has a 'radical' UI.
Back when I was in high school, the most sophisticated piece of technology we each had was a TI-81 calculator. But we also had textbooks and teachers. I imagine for a poor child in a poor nation, they won't have many books or many teachers. So devices like the OLPC will enable them to share the information that would ordinarily be carried by textbooks. And if the teachers are clever, a single teacher will be able to monitor and grade hundreds of students, by having questions be turned in electronically and automatically sorted and collated.
So, anyway, the OLPC isn't going to be used to teach computer science (though it would work for that), it will be used to teach history, literature, reading, writing, arithmetic, and anything else that requires information and exercise.
Jeff, while the Sugar UI still hasn't had any direct usability testing, it is built on the theories of people like Seymour Papert and Alan Kay, who have more than twenty years of experience building UIs and computers to help children's learning process. So this isn't really a shot-in-the-dark kind of effort.
If you've ever actually tried to automate Word, Excel or PowerPoint, you would know that they are not office automation products. With every version, Microsoft tells IT departments: "There. Now they can be used as office automation products." And IT departments complain that it still doesn't really work. Infopath, Access and Sharepoint are office automation products.
It doesn't take much thought to realize that Word, Excel and Powerpoint are COMMUNICATIONS products that are popular in offices because communication is very important in offices.
Now that said, the office environment is different than the third world school environment in a variety of ways. In the office, they have printers and an infinite supply of (sporadically available) paper and toner. Thus, Word's dominant metaphor is words on paper. Powerpoint is designed primarily for display with projectors, not small computer screens. Excel is optimized for the kinds of calculations done in an office and not a classroom.
The bit that bothers me about the Negroponte quote is that it is so illogical and elitist. How could a product named "Word" be about anything other than communication, sharing, making things and exploration? How could a product that projects one person's ideas onto a large screen at the front of a room full of human beings be called "office automation software"? If PowerPoint did not exist and were invented today it would be hailed as amazing, uplifting collaboration software. But it comes from Microsoft and therefore it must be evil, dull office automation software (common sense notwithstanding).
Let's try to look at this stuff with clear eyes, based on requirements and empirical testing, rather than knee-jerk reactions and biases.
The venom and unwarranted hatefulness displayed in some of these posts is a hand-in-glove demonstration why putting a "neighborhood" computer in every child's hands will never be the end all-be all to helping them improve their lives. A helpful tool, perhaps. But a replacement for textbooks, pencils, paper, and even recess? Hardly. Whether some people like it or not, the world is about socialization, not "collaboration" behind a keyboard.
Another thought occurs to me. I wonder what will happen when all of these innocent, newly-empowered children start Googling and run across screen caps of the Paris Hilton video, or worse. The vision tarnishes when it meets reality.
@ Luciano's ratpoison UI
They might have a hard time justifying the use of something called "RatPoison" on the One Laptop Per Child project. :)
I also agree with Mr. Negroponte's comments on the way we bottleneck youth into using "one true system". I hope this SugarUI gives a bit more freedom and creativity back to the user(child), even if it is just rudimentary. If anything it will hopefully give them a different perspective to add to the later enforced "standard" UI training they will endure, (whatever that may be at the time).
Word, Excel, Windows and other "real world" tools were designed to be used by people who didn't grow up with computers with minimal amounts of training. I doubt kids who cut their teeth on something like Sugar are going to have much trouble learning them. Whether they'll want to use them is another matter...
The single biggest reason I suspect I should continue to mostly ignore OLPC is the messianic pretentious blow-hards here defending it from imaginary attacks. If people like that are involved, the project can't end well. You're doing a heckuva job, zealots!
Sugar, on the other hand, looks to be interesting for all the reasons Jeff has given. In addition to user testing, I think it needs a team of fit-and-finish geeks. (Those icons are horrid, especially the "person" which looks like branding for Cingular.) But it's still interesting.
Maybe you don't want your kids to be information workers - but for the last couple generations parents wanted exactly that: for their kids to go to college and get a white collar job. But somehow the average while collar job seems to have become more menial while the blue collar jobs are less so. While I was brought up to believe I'd be better off working in an office, I see a lot more commonalities between being a software developer and being a skilled mechanic or tradesperson than I do between my job and the white-collar jobs of the past.
so basically...you likey very much?
pardon my french ;0
And the Linux community apes both companies, occasionally throwing in a curveball of their own. But when was the last time anyone tried a radically different UI on the desktop?
Ever heard of Compiz/Beryl? try it!!
The system was meant to encourage development of the next generation by making this entire OS different than anything they've seen before, so that their minds won't be pidgeonholed into thinking that a GUI "must" act or look in a particular way.
These are the future leaders, programmers, mothers, fathers, etc. of our world... they should be given as wide a range of experiences as possible. This could be the experience that leaves them thinking, "Hey, this open source stuff isn't so difficult after all!"
On that note, the people who criticize it for being difficult to adapt to can't be all that bright; I was a fairly competent user of the emulator after only about 15 minutes - AFTER having used Windows, Mac OS, and Gnome/KDE. It's not hard, it just takes a little extra loosening of the shackles that the commercial software companies have placed on your mind.
While the sub $100 is great stuff, not sure if the dumbing down of the OS was required at all. As Jerry Kindall mentions kids (that too kids in a slum and hardly with education) have managed to learn to use windows. What is required is the access and a sub $100 pc goes a long way. Here is a link to the experiment and the guy who installed the pc in the slum
even an idiot can learn how to click on icons/menus in windows/osx/linux or any other standard system/gui.
the point is to get the information to the children, to give them a tool to help them learn new things they wouldn't be able to without it, to support maximum amount of interaction in the information exchange.
I started out on an Apple II as a kid. The radically different interface on that computer did not impair my ability to learn windows or OS X in the slightest.
I have several XO's now, and IMHO, the OLPC mission is in trouble! The thing ships without instructions, not even any somewhere in the machine. the OLPC people say that this just reflects the entire design of the XO, that children will discover how to use it. Well, I have watched children of all ages just look at it, fiddle and give up...quickly.
After further trial and error, I report that this thing is just broken. The touchpad doesn't work, the OS crashes (what a thing to bring to the Third World!), applications don't run, and even basic word processing files don't conform to any standard (for sharing with others). What six year old needs to learn Linux OS commands, Pippy programming language, or ADSR envelopes for tone generators? They need to learn basic literacy skills and the XO doesn't come with any software to help with that learning. Even if it did, a child couldn't get it to work because the OS and hardware is so buggy. Children of the third world are not stupid. A few minutes with these and they'll know instantly that time spent collecting firewood will be way more helpful to them than one minute spent with an XO!
Jim, you must be stupid, because the OLPC comes with a complete suite of UI applications built on top of the sugar interface. There is no need to learn Linux OS commands unless you are a developer.
If the OLPC doesn't remain open source then it is going to simply become a watered down corporate desktop. This would be a worse case for the potential third world users, and will completely alienate the open source community surrounding the project.
I think that the ability to develop on the OLPC is as important as assisting an increase in basic literacy skills. Soon, there will be a million of these in the wild, so I would expect a significant community of developers will form.
Do you argue that a start menu and desktop is a better educational tool than sugar?
Can you run the most recent version of visual studio on the OLPC?
@VG: as much as I like and use Compiz, I'd hardly call that innovative. You still have a desktop, windows, icons... It just lets you manage your windows in a fancy way. (that's why it's a window manager duh).
"Will they get jobs if they say "I have experience with OLPC""?
I don't think this is really a concern. From what I understand, the driving idea of OLPC is that technology can be used for learning, which is quite different than believing that technology itself needs to be learned. The goal is to teach kids math and science, how to work together, and how to think about solving problems. We should be more than happy if the best interface to meet that goal ends up being poor training for Microsoft Windows.