I Don't Want a Dual-Boot Sugar/XP XO Laptop

   
   
   
   
   

In The Sugar daddy for future generations we learn that One Laptop Per Child is farther down the Windows XO dual boot path that expected. Nicholas Negroponte says that OLPC should start shipping laptops with Microsoft Windows XP in April.

olpc windows xo
Negroponte selling OLPC

That would be a huge shift from earlier assertions that OLPC would not offer Windows XO - it would only be a custom modification made by Microsoft. But Negroponte feels that by offering XP, he can remove and objection to the XO: Sugar.

"It gets you into countries where the minister of education or whoever else is involved are themselves Windows users, and find something like Sugar or Linux so alien that they think their kids will be compromised in the future. They can't argue if the machine does both. It's just like the Apple laptops, which companies didn't buy because they didn't run Windows. Now they do run Windows, they're more widely purchased and used - it got them into the office."

Now as a recent Apple convert, I have to strenuously object that the ability to run Windows is why Apple is gaining market share. The amazing maturity and stability of the OSX operating system, the quality hardware, and even the iPhone cool factor had greater impact. In fact, if anything, the rampant viruses on XP and the abject failure of Vista would be specific reasons not to trust hardware that can run Windows.

But bad analogies aside, Nicholas Negroponte then goes on to suggest that a Sugar/XP dual boot machine would be more desirable than a Linux-only model:

olpc windows xo
An XO laptop marketing flaw
"I think everybody will want the dual-boot version," and he expects all XO laptops to ship in that format. "We've had nobody say 'we don't want the dual boot' so it will presumably be the default. We boot Windows on open firmware now, which is a major breakthrough.

Congratulations on getting XP on open firmware - that's a good move. But again, Windows XO is an OLPC marketing flaw beyond any Open Source vs. proprietary discussion.

The "Does it run Windows?" question shows that OLPC's marketing is flawed in a very basic way. Negroponte has positioned the XO as a cheaper version of a business laptop. In this comparison, the XO will always fail.

The XO is a specialized learning tool specifically designed to empower education for primary school children in the developing world - 4-12 year olds in need of an educational experience suited for their developmental level.

As such it should be compared to other learning tools; chalkboards, libraries, and textbooks - then the comparisons are much more interesting and compelling. Once you look at the XO in a pure educational context, the question of "Does it run Windows?" becomes irrelevant.

Or think of it another way - when was the last time you wanted a Leapfrog to come with Windows? Me? Never. Not even as a dual boot.

Related Entries

1 TrackBack

I pose an interesting question to the esteems Open Source software development community - do you get better FOSS code, more functional software if the developers are paid or unpaid? I ask this because Nicholas Negroponte put forth a very counter-intu... [more]

22 Comments

The quoted “marketing flaw” article was written 8 months ago and has its value, but is there anything that can make the XO-1 marketing position any worse TODAY???

The developed world is totally indifferent to the XO-1/Sugar by now, no new large-scale deployments from the developing world have been announced the last 6 months (or more) and there are (apparent) requests from interested parties to run Windows. Should we just ignore all these and stay the course?

Besides, it is not for the seller to decide how the purchasers view the XO. Learning tool, production machine, communication device, is their choice (by the way, XO-deployment is way more expensive than chalkboards and that’s when chalkboards are available…). Would you accept the seller to define the use for any of your purchases? Why should they? The “I’ll tell you what and how” approach was attempted already and fell flat on its face.

Actually this was/is one of the major problems of the OLPC mentality. Unless you are suggesting that this mentality is OK as long as “I agree with what it proposes” or "I define the what and how".

Like it of not (and I don’t) M$ has >90% of the world market and even for the Mac that you mentioned the most popular non-Apple software is M$Office (I assume that you have it on you Mac…:-) followed by the Parallels/VMWare virtualization combo and millions of Bootcamp downloads that allows you to run Windows (and I would guess that one of the 3, if not all 3, are already on your mac). I do not think that any of these decreased Mac value and appeal or jeopardized the future of OS X. If anything it strengthen it.

We can discuss for hours past mistakes and alternative approaches that should have been taken, but this is not changing where the OLPC/XO-1 is NOW. Where it is now boils down to the choice “dieing as a virgin or leaving as a (explicit)”. We can not ignore the fact that the OLPC/XO-1 IS dieing. Any option that provides even a slim chance of survival should be considered. XOs that can run Sugar and Windows are better than no-XOs-at-all. No?...

Realistically, the alternative is a nice funeral service for XO-1 and, assuming that development funding is there (?), wait for the XO-2 arrival while trying to organize in a way that will not repeat the OLPC/XO-1 mistakes. This means an immediate commitment on the architecture, FURIOUS software development addressing the needs from the field and a management scheme that can navigate the complex realities of the developing world. But even if you succeed in all these, we can ONLY HOPE that the world will be still interested when you are ready a couple of years down the road.

These things MUST be done, wouldn’t it be nice if we can do all these without the funeral? Wouldn’t it be better if the XO-2 arrives without an 1year+ void? Wouldn’t it be easier to deploy XO-2, if XO-1 is still screaming and kicking under Sugar or any of the available and forthcoming OSs?

A marketing flaw 8 months old and never corrected is a current flaw today. The XO laptop should not be positioned where XP is thought to be a good OS option, if in fact, we're talking about an educational project for primary school children.

Windows XP is a fine business OS, and may be fine for middle or high school kids, but I don't see how it could compete with a software experience specifically designed for young minds, unless:

1) that software doesn't actually create (or at least perceive to create) a better learning experience, or

2) the buyers are thinking of the wrong target market - teenagers instead of children

So either OLPC has failed in making Sugar the right tool, or they failed in educating the buyer on what the tool is best used for.

OLPC may have failed in optimizing Sugar or educating the buyer (and probably few more) but this is not the issue.
Does the buyer have the right to use the "product" any way they want? If this is a prerequisite for them to buy the "product", should OLPC facilitate them or better stop to exist?
Do the recipient countries have the right to test/experiment with Windows or maybe another Linux flavor if they want?
Does "education" means "MY kind of education"?

I have no issue with a country buying XO's with Sugar, testing it, and then deciding it will go with XP. Once bought, they can do what they want.

I am saddened that OLPC, which billed itself the education project, is willing to think of themselves as a neutral hardware vendor willing to put any OS the customer wants.

Maybe I am the only one who remembers the OLPC's original mission.

I’m sudden to but I do not think that OLPC’s demise will serve these goals better. I’m not convinced that OLPC has taken the road of no return and that its dissolve will better serve its original goals. Do you?
I also do not see anybody that could pick up the slack. Sugarlabs can definitely not do it. Besides, in spite of the mission statements, OLPC is perceived as the “$100 laptop” organization. A hardware project for the affordable, power conscious, outdoors usable, interconnected laptop. Sugar or a SPECIFIC educational scheme was never its strong selling point.

"Any option that provides even a slim chance of survival should be considered. XOs that can run Sugar and Windows are better than no-XOs-at-all. No?..."

No. What would be left of the project if it will be controlled by the people who only need it to create more Windows-dependent users? What do you think, Microsoft is going to use it for -- education?

Wayan,

I think the whole XP-on-XO is a scam, a diversion. XP simply is useless on the XO, even if it boots. Everyone seems to require that others should use XP on the XO. But these same people will NEVER use the laptops with XP themselves.

You will sure know:

"Netbooks are NOT made for Windows XP or Vista"

http://www.helpmerick.com/netbooks_are_not_made_windows_xp_or_vista.htm


Because netbooks run on tiny batteries and less powerful processors, trying to run a large operating system like Windows XP and especially Vista is completely impractical. Windows XP with all its updates and service packs and security software requires a strong processor and a minimum of 512mb of RAM or 1 GB of RAM on a slightly older processor. Also, Windows XP, and especially Vista) require above average graphics processing to function with any oomph.

And the following might explain why there are so few solid state netbooks sold running XP:

Flash Drives and Operating Systems

http://www.esis.com.au/Adv1CPU/Adv1CPU2SSD.htm

The following operating systems are well suited for flash drives as they do not constantly write data to the system drive:
· Windows XP Embedded when set up to boot from a read-only partition;
· Windows Embedded CE;
· Many variants of Linux can also be configured to run from a read-only partition, or to minimise data written back to the system drive;
· plus all the lesser known OS’s (many derived from Unix) that designed for embedded systems.

Winter

Winter wrote:

"I think the whole XP-on-XO is a scam, a diversion. XP simply is useless on the XO, even if it boots"

;-)

Irvin,

I am glad you too have come to the conclusion that XP on the XO is a joke.

Winter wrote:

"Irvin,

I am glad you too have come to the conclusion that XP on the XO is a joke."

I have said it from the beginning: Negroponte's promises of the XO *successfully* runing XP are as legit as a 3-dollar bill.

I have no doubt that they are capable of releasing some aberration that PRETENDS to run XP, but we know that it would never work with the current hardware specs.

On the other hand, Negroponte is in a real tight spot: nobody believes that mere possession of a laptop will result in an education and Sugar is buggy, slow and devoid of any serious educational software (yes, they have an assortment of crap, but it is not what *potential buyers* would consider useful).

What can Negroponte possibly do?

He offered the XO plus Sugar to the entire world and very few people showed interest.

What can Negroponte possibly do?


:-(

Windows XP does work on the XO, and quite nicely too. I've seen it with my own eyes and used it a bit too.

So XP on the XO is real. Very real.

Now the question remains if its a better learning environment than Sugar. I fell it does, but as I am quick to admit, its a feeling based on anecdotal evidence. But so, mostly, are claims that any computing systems improve learning outcomes.

-- Correction: This comment should read that I feel Sugar is a better learning environment than Windows--

Wayan wrote:

"Windows XP does work on the XO, and quite nicely too. I've seen it with my own eyes and used it a bit too...

Now the question remains if its a better learning environment than Sugar. I feel it does, but as I am quick to admit, its a feeling based on anecdotal evidence."

So, why do you object to XP on the XO?

According to you, XP works "quite nicely" on the XO and you also "feel" that it provides a better "learning environment" than Sugar.

Puzzling, to say the least...

Correction - I feel that Sugar provides a better learning environment than Windows. Not the other way around as I wrote above. Typing too fast...

Wayan,

Windows XP does work on the XO, and quite nicely too. I've seen it with my own eyes and used it a bit too.

No doubt about it. Two questions:

1 How long will the Flash memory last?
The XO's are expected to be used 5-6 years. As the links I posted pointed out, XP writes too much to disk, and there are running too many processes. I understand you can cut down XP to half. But can you really use all these "wonderful" educational applications and games without virtual memory? I m afraid XP will wear out the Flash memory fast. (maybe that is why SSHD netbooks with XP are so rare)

2 The XP GUI is designed for "large" screens. Look at any XP user's desktop and there are dozens of links on it (hardly usable on 7"). Somehow, the menu system in XP is not enough. So how productive will people be on 7"?

I know perfectly well that these issues can be resolved. Intel uses a modified XP GUI. MS can pimp up XP embedded. But all these changes will require changes in the applications. And in the end, the whole point that "children should learn Windows" becomes moot as the MS OS they will "learn" is incompatible with what MS sells.

@eduardo montez:
O, so the XO-1 is going to run XP, but what about Windows 7? It's coming out this year, is it being ported? Or will olpc stick with XP, and will Microsoft let them?

I expect that MS will rebrand XP embedded as "Windows 7 netbook edition". So users will have the benefit of running the latest trademark of MS (although the OS will be 8 yo).

Winter

I expect that MS will rebrand XP embedded as "Windows 7 netbook edition"

I laughed out loud when I read that. But you might be right. Microsoft is desperate to stop supporting XP, but maybe it will be forced to keep it up anyway.

"It gets you into countries where the minister of education or whoever else is involved are themselves Windows users, and find something like Sugar or Linux so alien that they think their kids will be compromised in the future."

It's almost funny to hear that quote from THE MAN himself. :)

I loved Dr. Negroponte's vision from the start (although I was a bit skeptical when I first first heard about OLPC). But I never once believed that going with Linux would be a good choice in trying to sell this vision to 3rd world countries.

It's just not practical.

Running some sort of Linux based OS (Sugar, Ubuntu, you name it) on XO is very cool for computer enthusiasts around the world. But there is no single incentive for 3rd world countries. Especially if you are selling the vision as an educational tool.

I know this is one of the often debated topic here. But if you strip away all secondary issues (such as if running an XP on XO is fast enough), you can't deny that one of the major failure factor was alienating world leaders to recognize XO as a machine that runs apps that most of the world uses in their work force.

O, so the XO-1 is going to run XP, but what about Windows 7? It's coming out this year, is it being ported? Or will olpc stick with XP, and will Microsoft let them?

And if the X0-2 uses ARM, will Microsoft port Windows 7?

I was interested in what the article said about how they are going to open source the hardware design for the X0-2, and hope people will copy it.

Maybe we will see 10 variations, some focused on education, others for personal use, or field work, or who knows what. And the more different models, the more competition and the lower prices will go, all of which will be good for the developing world.

Here is a nice article on what MS is willing to do against companies, even the size of Intel, to prevent users from getting their hands on Linux:

http://boycottnovell.com/2009/01/30/microsoft-intel-anti-linux/


"On the OEM side I am thinking of putting hitting the OEM harder than in the past with anti Linux actions, in addition I will stop any go-to-market activities with Intel and only work with their competitors (something which is easy to do because they normally put crazy demands on us). "

So, maybe, putting XP on the XO is a kind of self defense against very real MS retaliation?

Winter

With defenses like this, who needs attacks?

Seriously, there is one thing Microsoft can't do, and it's to affect people who don't use Windows and don't make anything for Windows users. OLPC wasn't supposed to do either of those things, so it's Negroponte's own damn fault to ever talk to Microsoft. What would Microsoft do, try to keep AMD from selling embedded CPUs?

@teapot:
"What would Microsoft do, try to keep AMD from selling embedded CPUs?"

Actually, this is not as far fetched as it looks. Over the years, it has been shown that MS have "forced" hardware makers to abandon technologies that it deemed unprofitable for itself.

Cutting off supplies of smaller competitors has been a VERY old strategy (eg, used by Coca Cola). Furthermore, AMD are desperate and would buckle at the slightest pressure.

Winter

I agree with those who say that 256MB just isn't enough to run XP well. I recently had to upgrade RAM in a desktop XP system which only had 256MB because it had slowed to an agonizing crawl due to a large amount of swapping required to run Firefox with all of the other bloat that has crept in. I can't run Opera with the Flash Player 10 plugin on my XO because 256MB isn't enough to load the plugin once Opera is loaded, with no other activities running. Even if XP is stripped down as small as the Fedora in the XO, what will run well in it?

Close