The XO laptop gets a Windows makeover
On the outside, the Windows version of the XO laptop looks just like the Linux model. But simply booting up the device shows that the Windows version bears little resemblance to the original One Laptop Per Child device.
With the Microsoft version, you get Windows, for all the good and bad that entails. It's full-on Windows--XP Professional, in fact--and can run basically any software that can adjust itself to the mini-laptop's diminutive screen and modest processor.
Microsoft has managed to slim down the OS enough to boot up off a 2GB flash memory card and has written drivers for a number of the XO laptop's unique features, such as its scratch pad, game controller, and built-in camera.
But what's missing in the Windows version is the personality that oozes out of the Linux incarnation. The Linux model comes with an integrated suite of educational games, programming tools, and other software, all built around a kid-friendly OS shell known as Sugar.
The Windows version of the XO doesn't have much of that built-in spunk, although a child-oriented programming tool known as Scratch did survive the Linux-to-Windows switch.
At the same time, having Windows allows students to take advantage not only of Microsoft's dominant Office suite, but of all the educational software that has already been written for Windows.
For the past week, I've had two XO laptops on my desk, one of each OS variety.
I've taken them both to coffee shops and let myself explore each machine. I'll save my thoughts for a later post.
But to really get a sense for each device and its unique appeal, I turned to an expert--an 8-year-old who's far more representative of the target market than a reporter who has to dye her hair.
Ella Taggart, the daughter of one of our editors, happily volunteered to put each of the devices through its paces. She spent an afternoon at CNET's offices on Wednesday exploring the built-in software on each, looking up her spelling words on Wikipedia and attempting to visit her favorite Web sites.
In the end, she found each option had its challenges and each its benefits. She had a great time using the built-in speech synthesizer on the Linux version, while the Magic School Bus game that was on the Windows version was also enthralling.
Web browsing was slow on the Linux model and the pointer and menu system somewhat complicated for someone used to Windows. Still, when it came time to borrow one for the night, she opted for the Linux model, in part because it had more built in than she had a chance to explore in her brief time at the office.
In the end, she said she liked the XO no matter what software it was running. It was fun and just the right size for her (even if all the adults complained about its small keyboard).
From my perspective, her experience shows not that the software doesn't matter. It matters a great deal. But it's all about how a school chooses to use the laptops. Used properly, as part of a well-thought-out curriculum, both models offer tremendous opportunities for students to learn about technology and how to use technology to learn about many other areas.
At the same time, I don't think either model simply dropped in the hands of children will do the trick. That meshes with the experience I had touring through the Bradesco Foundation school in Campinas, Brazil. It wasn't the fact that all the students had Intel Classmate PCs that made the program stand out. It wasn't the use of Windows over Linux.
What made the the experience so dramatic was how well the teachers incorporated the laptops into their teaching. It was the fact that the art teacher used the PCs for research, but had the students put them away and use their hands to make wax sculptures.
It was the fact that while each student had their own laptop, they still worked in groups as often as they worked alone.
Some countries have demanded that their students work on Linux, arguing open-source computers offer a chance for an independent software economy not tied to Microsoft. Many others, such as Peru, have demanded Windows, arguing that that's what their students need to get good jobs. Ultimately, OLPC hopes to offer a dual-boot option, though that is still being developed.
And while developing nations will now have their choice of operating systems, those in the U.S. won't enjoy the same flexibility.
OLPC plans to reprise its "Give One, Get One" program this holiday season, which lets Americans pay for two machines--one of which they keep and the other of which they donate. However, in all cases, the one they get will be of the Linux variety.
In part, that has to do with the fact that Microsoft offers a cut-rate version of Windows for large educational programs that doesn't apply to the consumers here that buy the machines.
Microsoft general manager James Utzschneider notes that it's also not comfortable with the lack of support that comes with the OLPC for those who take part in Give One, Get One.
"This is Windows," he said. "People want to be able to pick up the phone and call us if they can't get something to work."
During her years at CNET News, Ina Fried has changed beats several times, changed genders once, and covered both of the Pirates of Silicon Valley. These days, most of her attention is focused on Microsoft. E-mail Ina. 
"The computers are for educating a future workforce in a Windows dominated world."
That's horribly short-sighted. Technology tends to have a lifespan of about 5 years, and is completely obsolescent within 10. If we're targeting 8-year-olds, then, the particular technology they learn with will be moot long before they hit the workforce. Windows would seem to be an exception but looking back on Windows ten years ago and Vista today ... they have roughly the same differentiation as Vista, the Mac, or Ubuntu Linux. People adapted to the changes in technology as necessary.
The XO was not intended to be a workforce training tool but rather a discovery tool. It's greatest possible benefit should be the vast amount of information that can be accessed with just a web connection. Books, reference material, etc. that would otherwise be very hard to come by in poor or isolated communities. The operating system on which it is built is largely moot except that Linux is royalty free and far easier to strip down than Windows. It was an excellent choice for a base.
I think the failure in the XO was tossing out literally all of the existing technology for UI and writing it again from scratch. A stripped-down X11 plus simple window manager with just about any widget toolkit would have given them a lot more maturity, and applications, right out the door. Above the OS they've had to build a complete ecosystem and in so doing have had all the usual growing pains. It has matured a lot in the last year, but it still has a long way to go.
This is particularly true when it comes to information access. With internet access and a terrific display system they should have been able to give immediate access to the web and tens of thousands of books. In practice, though, the web browser is painfully slow and there are no "standard" e-book readers so you can't take advantage of huge libraries like Project Gutenberg. The books that have been available were poorly formatted for the device and there just weren't many of them. I think if they had built a "library" tool capable of hitting several common book repositories and just ported Mobipocket they would have had a huge winner just as a reading device, never mind the other things it can do. More on this idea in a minute.
If you want an information access device with educational games and whatnot a rather cheap and very effective solution right now is an iPod touch. It's one hell of a lot more usable than an XO and not much more expensive. So, highly functional mainstream commercial offerings are already getting down to the same price-points. I think we can expect that progress to continue.
In the long run, though, I anticipate that the big learning tools are going to resemble the Kindle much more than a PC or handheld. The principal issue with learning is information access, and a PC is just overkill for that. Tries to do too much and is too expensive in terms of both capital and power costs. But a device that tries to be an especially capable book and basic web browser can be made very cheaply, can have superb battery life and durability. With e-ink displays coming in at as low as a few dollars now (see the October Esquire cover) I think we are no more than a few years away from mainstream devices that hold libraries of tens of thousands of books, access the web, and yet cost only a few dollars. And that, my friends, is going to change the face of education worldwide.
- The mission statement for the XO project is "To create educational opportunities for the world's poorest children by providing each child with a rugged, low-cost, low-power, connected laptop with content and software designed for collaborative, joyful, self-empowered learning."
- There is nothing in that mission that says "dethroning Microsoft" or "killing proprietary software."
- The implosion of the XO project in the past year is the direct result of Free Software ideologues demanding that this device be imbued with an ideology that takes priority over its true mission.
- This project is about learning, collaboration, and children. It is NOT and SHOULD NOT be about ideological battles over software development. That sideshow is getting in the way of achieving the mission. If educators/politicians/etc. want a machine that runs Windows because its maps better with their view of economic and educational progress, let's not tell them "we know better than you..." If they want Ubuntu instead of Sugar...we should not tell them "No. We know what's best for you."
Ideologies are fine, but there is a time when you need practical solutions for serious issues. That is the time in which we need to stop preaching and just find flexible solutions. Mr. Negroponte realized this...it's too bad so many others have not.
Unless you install virus protection software on the Windows version of XO, it will become infected quickly, especially if you are running Explorer. And if you do install virus protection software, expect the XO to become a slow pig of a computer. You are damned if you do, and damned if you don't.
/P
penguinisto....you only demonstrate your ignorance.
When you take good works - like the Gates Foundation - and you smear 'em so you can fit whatever cause you have -
As for the article, what I did not see was how much of a footprint was used by each OS... memory consumption, drive space usage, etc. How much is left for applications? And since this is for children, why don't we let them and the teachers choose rather than all the adults argue over which is best.
Please present your facts and dcoumentation for review. I call you on your accusation.
Back up your claim or retract it.
When you accuse a person of a crime, you better have some evidence to back it up, Penguinisto.
Twenty-five years ago a study was done as to why various "reforms" "pedagogies" "new methods" etc worked or didn;t work in schools. The end report was that good schools got good use out of any tools. Schools where the culture was defeatist, exhausted, or lacking vision got no utility other than a fast bounce out of any new methodology.
this device either "stands for" a linked in, communicative, operative world....or an another attempt to play us/them..which strategy is working sooooo wellll everywhere else. children know better, and they will decide.
But don't blame MS for this. This has as much to do with OLPC failing to establish a serious foundation and framework for deployment and support of their system as it does with possible bribes. They put this thing out into the world and expected people to rejoice. But there is no technical support for buyers. And there's no real evidence that just putting a computer in a child's hands will fundamentally spark creativity and a desire to learn. If OLPC can't figure out a good way to manage their own ecosystem, then don't blame someone else for seeing an opportunity and jumping on it.
You're presuming that Windows even exists in anything like the current form some ten-plus years in the future. Windows programming today, using C# et al, is vastly different than MFC programming was ten years ago. Learning to write code to the Windows API as a teaching tool is only marginally useful.
But I wasn't even thinking about writing code, I was thinking of using productivity tools -- the "workforce" tools you're implying make Windows more valuable as a teaching system.
MS Word has changed quite significantly in ten years, at least as much as the difference between, say, MS Word and Open Office right now. Similarly the Windows user interface has changed at least as much in the last ten years as differences between any of the modern UIs, possibly excepting Sugar. If teaching the specifics of those tools was important than most of the workforce wouldn't be able to use them. Similarly by the time those children enter the workforce the tools will be very different no matter what technology you pick today. They'll look different, they'll have different UI idioms, different underlying operating system, different applications.
This is not a problem. My eight-year-old daughter, for instance, has had no particular trouble learning to use the XO, Windows XP, Windows Vista, Mac OS X, and the iPod/iPhone. (Some of the particulars of Sugar escaped her, like the neighborhood display since there are no other XOs around us, but that's not a fault of the XO interface -- it was intended to be used in group environments which I don't have.) Browsers in all of them act pretty much the same, wordprocessors work pretty much the same everywhere, etc. WordStar users managed to learn Word. Mosaic users still get along with Internet Explorer 7. Mac users managed to learn to use Windows, and vice versa. The concepts carry even though the specifics don't.
(continued next post due to size limitations)
Over time the devices will likely bulk up to the point where it's possible to run Windows comfortably, but not with a substantially reduced price point. Moreover, there is a huge reason NOT to pick Windows: cost! Windows is expensive!
To bring that point home, I had dinner with the head of Microsoft's HPC project back in I think 2003. We were discussing Microsoft's odds of doing well in the HPC market. I said I though it was unlikely given the licensing costs. To use Windows in that environment forces you to use the Server variant, which at the time was $800/license. If you had a 4000 CPU machine that was $3.2M in Windows licenses alone. If you used Linux instead you could buy a hell of a lot more CPUs. His response was "we'll cut you a deal" but even at a 90% discount that is still a lot of money you could otherwise spend on CPUs -- every four or five Windows licenses is another CPU. This explains quite well why Microsoft has done effectively zilch in the HPC market in the intervening four years. If the Microsoft isn't given them the OS for free then Linux is the overwhelming choice.
Now, I grant that we're not talking about $800 licenses for XO PCs but the point is the same: Even if Microsoft were selling Windows for $1 on the XO PC, approximately a 98% discount over bulk Vista Home bundles, across a million XO PC installation that's $1M in software licenses that could be used to buy more than 5,000 additional XO PCs running Linux. Windows really needs to provide a hell of a lot of added value for that to be sensible. It doesn't offer anywhere near that kind of value.
Your Win98 argument really doesn't hold water: Sure, there are a lot of those still out there in absolute terms but not in relative terms. In the US the number of Win98 installations has been shrinking by half every three years since 2001. It's slower elsewhere in the world, but not a lot slower; hardware failures force the obsolescence cycle. By 2011 Win98 will have a very small market indeed and the odds are very good that anyone trained on Win98 as a child will never see it when they hit the workforce. It doesn't really matter since the concepts they learned on Win98, or XP, or Vista, or MacOS, or Ubuntu, or even Sugar carry across to the others just fine.
So: Education necessarily revolves around some generalities since the specifics change on a much more rapid time-scale than the education process. Given that this is the case it makes little difference what the underlying OS is in terms of training. It makes a difference in terms of applications and it makes a difference in terms of software cost, but in both of those cases Windows fairs poorly on an XO PC given its limited hardware and its market's high price sensitivity.
Personally I think the XO PC is doomed to fail no matter what OS it runs; it is just not possible to make PCs cheap enough to compete with special-purpose devices we expect to see over the next few years, particularly web-enabled e-books. That isn't the case today since the technology is still ramping up and volumes are low, but even today it's possible to find devices at price points very similar to the XO PC that do all of the really interesting things an XO PC does without so many of the limitations. Cellphones and music players are already encroaching on the price and capabilities of these PCs, but are often more usable. E-ink tablet computers start at $300 today, should be $200 by 2010, $100 by 2012, $50 in 2014, and less than $20 by 2016.
I don't believe PCs in any form can match those prices. I notice that PC prices have flattened out a lot in recent years; economies that can be made have already been made. The XO and Classmate PCs really push the envelope of what is possible but still reasonably useful. They will never reach $20 price levels, and will instead get gobbled up by other technologies that can be so cheap that they can be given away.
(continued next post)
Some twenty years ago I came up with what I call Jim's Law: The cheapest thing that gets the job done wins. I had noticed that the best technology in just about every domain wasn't the one that won in the market. Why did minicomputers beat out mainframes? Workstations beat out minicomputers? PCs beat out workstations? UNIX beat VMS? MS-DOS beat CP/M? Windows beat MacOS? Linux beat UNIX? Simple: Each was the least expensive one that did the job well enough. No matter that each successor was inferior in major respects to its predecessor. People put up with all kinds of hardships to save a few dollars. Over time those cheap technologies grew more capable and their prices grew with them, until the next cheap-but-good-enough technology came along to wipe them out.
This is how I know for sure that Windows is not the place to bet the farm right now. It stopped being cheap. It has lasted as long as it has on momentum and not a small amount of monopoly leveraging but it's easy to see it losing pace with every passing year.
The cheapest thing that gets the job done wins, and it's hard to beat zero. That in a nutshell is the problem with betting on Microsoft even if you're one of those people that believes education needs to focus on specifics rather than generalities.
jim frost
jimf@frostbytes.com
Why not invest the already scares resources in things that would make real difference.
Mass produced textbook only cost as much as the ink and the paper it prints on. With less than few dollars we can cover the cost of an entire year worth of textbooks. With the money saved from not buying OLPC, we can provide much more needed school supplies such as textbooks for kids who might otherwise afford it.
Let not forget that almost all the K12 classes in the U.S. are still taught by teachers in front of students with textbooks.
That's not entirely true. You also have to pay printing costs, transportation costs, storage costs, and replacement cost when damaged to the point of un-usability or lost.
There is no way to upgrade a printed book when new information is available, you have to replace the whole book.
The future of education is electronic readers/computers. An entire year of schoolbooks can easily be stored on a hard drive or DVD. Updates to books will be a easy as a download, anywhere in the world. With electronic transfer, homework can simply be transferred to a teachers computer. No more unreadable reports!
Very soon printed school books will be an anachronism.
The true solution is to buy everyone Korean parents.
Korean parents for sale
You say you're not all
That you want to be
You say you got a bad environment
Your work at school's not going well
Korean parents for sale
You say you need a little discipline
Someone to whip you into shape
They'll be strict but they'll be fair
Look at the numbers
That's all I ask
Who's at the head of every class?
You really think
They're smarter than you are
They just work their ***** off
Their parents make them do it
"This is Windows," he said. "People want to be able to pick up the phone and call us if they can't get something to work."
Everyone that has had any luck picking up a phone and calling "Microsoft" for help on Windows raise their hand.... anyone...? anyone...? The answer is doesn't happen lets move on... ;-P
My name is Carlos Mauro Cardenas Fernandez. I'm studing systems engineering into deh National University of Engineering. Now working in my thesis about the
"Evaluation of the OLPC with the Usability Engineering", and i made some usability testing pilot and testing process with the Sugar into Classmate, OLPC, and Desktop. I would compare the usability of WindowsXP with Sugar by testing.
But... I could install windows into OLPC... I think make this with adding a memory SD but have problem with the boot and the install the windows into the SD.
Can you help me install the Windows in the OLPC. I have the original program or Where do I get one?... In the ministry from Perú is very hard.
My work:
[1]http://unimauro.blogspot.com/2008/06/usability-pruebas-piloto-con-las-olpc.html
[2]http://unimauro.blogspot.com/2008/07/usability-thesis-segunda-sustentacin-de.html
[3]http://unimauro.blogspot.com/2008/08/proyecto-presentado-en-intercom.html
[4]http://unimauro.blogspot.com/2008/06/usability-prueba-piloto-de-usabilidad.html
That is what happens when you move from Linux to Windows.
- by pauljweighell October 26, 2008 3:48 PM PDT
- As the review said, Linux is slow to browse on the XO and that's a killer for modern kids. Windows is the world's standard opsys so not to allow the XO to use it would just be dumb. If Linux lovers want to push their personal preference then that's fine too, variation is how things evolve.
- Like this Reply to this comment
-
(40 Comments)