OLPC urges Microsoft to make Windows 7 for ARM

Yeah right like that's going to happen. It failed with NT back in the day so I don't think Microsoft is going to go through the hassle of releasing Windows 7 on Arm, it's not as if any x86 apps will run anyway!

I think the closest they're going to get is probably Windows Mobile.

Personally I think the OLPC project lost the plot when they dumped Linux and should just roll over and die. Maybe if they hadn't shot themselves in the foot by dumping Linux they would be able to get an ARM port up and running. Oh well.

Rob

I was going to say that MS will most likely do this out of the heart of their bottom .... if only to spite penguins like the mug above.
OLPC lost the plot when likes of DELL and Co wanted to make even cheaper laptops to make money on the developing world and then everyone just forgot about it. BASTARDS! This is almost as bad as JP Morgan robbing Nikola Tesla cus he wanted to give the world FREE< WIRELESS electricity ..... but you cant make money on FREE.

That would be a great move, particularly as some of the ARM parts are pretty damn fast now (Qualcomm's Snapdragon is a dual core that runs at 1-1.5GHz). It'd also introduce more competition in the market for Intel and AMD, particularly so because of .NET and Java applications which can be cross-compiled for multiple hardware architectures (yes, .NET does this, with XNA you can compile applications straight to the 360 without tweaking anything, there are also options to compile for Windows mobile devices). More competition is always good, so I hope this happens.

Even if they're able to convince MS to make win 7 available for ARM (which is not going to happen anyway, it'll just cost MS a lot of resources without any chance of return), there aren't any apps out there for the platform. You'll be limited to paint, wordpad and notepad.

Linux may not be exactly user friendly for the 'normal' user (moderately computer literate), but I feel it's good for the highly savvy crowd or the computer illiterate. Latter because it's open source and can be customized into a custom distro. Just put a few big icons on the desktop and teach people to click on them. That's it. If they're computer illiterate, they're not going to try to install anything or make changes or tweaks to the system settings. (That's the part about Linux that the 'normal' users hate the most).

Porting Win to ARM is not enough.
You have to port all windows based apps to ARM.
let me guess ...
It's not gonna happen :)
Linux have little advantage.
Ports done by distributions.
It means that they create binaries from source provided by developers.
It is working, but you don't have anything of this kind/workflow in M$ world.

+10 :)

I'm sure an OpenSolaris port to ARM would not be too difficult, especially if the hardware is controlled by OLPC.

Ahh, I remember the days when we ran RISC OS on our Acorn machines at school. Maybe Negroponte should have a talk with the keepers of RISC OS - he might find a bit of a winner on his hands.

"He has 18 months to convince the world's most profitable software maker that it's in the company's best interest to create what, to all intents and purposes, will be a free of charge operating system for some of the world's poorest people."

This is just to funny. Intel couldn't even get Microsoft to port windows xp to the "Itanic" so why would they make it for some low cost laptop that no one is going to buy? I have an idea... They are going to port it and everything and proclaim how much they love those poor people and then they are going to sell it to arm for their netbook and give olpc a chance to bid on some licenses.

Seriously I thought it was supposed to be an advantage that it was using Linux, since it isn't tied down to any proprietary format.

Maybe if OLPC were to show interesting in OSX then M$ would be a bit more motivated. But to be honest, linux would be in everyone's best interest. It would allow developing countries to create their own apps which would provide a return to the community that's reached out to them. They don't need MS Office or IE or Photoshop since there is open office, firefox and Gimp.

OLPC became a joke when it abandoned sugar and nix, now it's abandoning x86 too. Stick a fork in it, the OLPC project is done.

As the poor nations gain wealth, they will one day be able to afford things like computers. Now is the time for M$ to get their foot in the door by advertizing.

Also, I'm frusterated by the epic failure of OS developers in general. An operating system should be able to take an app made for that OS and run it on the hardware the OS is installed on. Software made for Mac OS 8 should run on OSX, forget about differences in IBM vs Intel. Likewise, a program made for Windows/x86 should work on Windows/ARM. It's what an OS is supposed to do. Seems to me that all OSs suck in that respect.

Didn't AMD do the dumping of the OLPC when they canceled the geode cpu in 2008?

Why does he even want to bother with Microsoft and Windoze/Orifice, when there is all that nice, open source, free software running around that does just as well, if not better?

I use Ubuntu and OpenOffice at home; I have no trouble exchanging documents with my office PC (windoze XtraPerverted Pro and Orifice 2007), and I didn't pay one nickel for the software at home.

Negroponte definitely has his head up where the sun can't shine.

Remeber Sata Drives io-Dual? It came out yesterday. Well Io-dual high speed sata has restriction, works only On iA 64. So as Low Cost Searches Lower. High End Seeks Faster. XP as 16 bit core/32 bit O/S thats ENDED by Summers End, How'd School Kids Wind Up Work on Vista? Maybe Microsoft Is Too Furturistic in Fast Lane, 32 bit core vs. xp 16, Less Packaging for Higher Quality Hardware, Yet Even Little Ones Have their Needs.Probably Be Linux Machine ?if it wasn't backed by another.Small Windows, like starter or ?xp equivelent be best. drashek

The only convievable reason I could think that Ed McNierney would want Windows 7 on the thing is some evil contract from Micky Mouse. Like others said, the OS might run on it but none of the software will. I doubt you'll see office on it or anything really. Poor sap.

Good point Rich. Once you get a bit of the MS Kool-Aid there's no going back. It's just like crack and Negroponte needs his fix is all.

Linux is already ported to ARM but they won't pay Negroponte or his project or "favourite charity" anything.

MS on the other hand has lots of cash and they already paid to get on the OLPC once. Now that Negroponte has sold out we can expect MS to leave him out to dry. You served your purpose now run along, there's a good little UN official. Don't you know we are MS and what we do is buy things and wreck them?

MS decided to implement an emulation layer into the platform.

This has been done before with Windows... DEC created FX!32 that allowed native x86 apps to run on the Alpha version of Win-NT... without any need for a re-compile.

There was a big performance hit, but given the massive overkill of modern CPUs, and code written by the people who are more intimate with the OS than DEC was, it's possible they could get away with it.

They're using ARM processors. I don't think they're powerful enough to do that.

Microsoft should just get on with writing their own Linux distro, along with an emulation layer for Windows compatibility. I'm surprised they haven't done it already...

It needs a 2GB RAM machine, as a realistic minimum, and about 10GB of disk. It's just not built to run on ultra-low-end systems. Getting MS to open-source Windows 3.11 and Office 4.2 would be far more realistic, but that isn't going to happen either.

OLPC would do far better to pick some small, simple and powerful OS, like Android. That has a modern user interface, excellent connectivity, and requires far less resources than Windows. If you want to bring mass-market IT to the developing world - which already has mobile phones - that's a more realistic idea.

The emulation layer would have to be even bigger than W7 would be.
I think Negroponte is backtracking and wants to give the illusion of being on MS side but knows theres no way they'll get a megalitre into a pint pot.

I don't know what sort of Kool-aid they serve in the UK, though in the US it comes from the Wall Street Journal and Business Week, which hate Free Software.

OLPCNews.com, on the other hand, reports that in the rest of the world there are over a million OLPC XOs running Linux and Sugar, and fewer than 5,000 running Windows XP. There will be no more Windows-only XOs. It is either Linux alone, or dual-boot.

I can't tell you how much I am looking forward to trials of Linux + Sugar against Windows + educational shovelware on the same hardware by the same people, almost none of them on the Microsoft payroll.

There are more than a million other Linux systems on order for several countries, including Mobilis systems for Brazil with X-SCALE (Intel ARM) processors. Sugar Labs is porting Sugar to their Linux distros, including ARM/X-SCALE.

We all agree, however, that Nicholas Negroponte's mouth is his own worst enemy. ^_^

FLOSS Manuals has brought out Free manuals on the XO, Sugar, and several Sugar Activities (applicatons). We are getting started on textbooks incorporating Sugar software. You can get Sugar packaged for Fedora, Debian, Ubuntu, and other distros. Try it! Some adults hate it, but kids love it.

"Like many, we are urging Microsoft to make Windows – not Windows Mobile – available on the ARM. This is a complex question for them," said OLPC head honcho Nicholas Negroponte.

... doesn't that just leave you laughing so hard you can't breath?

pretty please microsoft? pleeeeaaaase!

Bet my house it doesn't happen.

Microsoft still thinks ARM is intended for mobile use only. It is more complicated to porting x86 software binary that have been developed for years like autocad and coreldraw

people forget that OLPC wants to make money. they are not 'nice guys' - never were. At the beginning they tried to play the tech-loving open source crowd. A laptop for every child! And with linux! And free software! Support us! Give us your money!
But it was less money than they imagined - so they are dumping them - instead of asking themselves why other corporations can make a netbook for less that doesn't suck as hard as that olpc crap - and are available world wide and not only in the states because OLPC fudged on quality and is afraid of RMA's.

OLPC needs to die. The earlier, the better. It was a stupid idea, it still is, and it was not started for altrustic reasons. That was only a card played to get the groklaw and slashdot idiots into buying the crap

asynchronous chips and memories. OLPC could be an interesting platform to innovate for some outside the grid action.

XO-1 created the whole Netbook market revolution. Where Intel's profit margins were forced to shrink 95% between 2007 and 2008.

With XO-2, now OLPC is going to go for the kill. Intel and Microsoft are going to die from it.

XO-1, just as XO-2 is produced by laptop manufacturers in China. OLPC doesn't manufacture anything themselves.

The focus with XO-2 though will be to make it easier for many different manufacturers to copy the designs and the technologies on following points:

1. Thanks to the OLPC XO-2 project, all Laptop manufacturers will use ARM processors instead of Intel.

2. Thanks to the OLPC XO-2 project, all Laptop manufacturers will use Google Android embedded Linux OS instead of Microsoft.

3. Thanks to the OLPC XO-2 project, dozens of ARM processor manufacturers (TI, Qualcomm, Marvell, Nvidia, Samsung, Broadcom, Freescale...) will compete on pricing for the delivery of processors for the bulk of the future Laptop market. Intel and AMD will be gone from most of the consumer markets.

4. Thanks to the OLPC XO-2 project, Laptop power consumption will all be below 5W, towards being below 1W in use. All will have Kindle-like E-readable screens for seriously reading a lot of digital text.

5. Thanks to the OLPC XO-2 project, all Laptops will start being available at $100 each, and will reach the 5 billion people who still don't have access to the Internet to improve all of their lives and organize society much better.

All you naysayers and open source fundamentalists need to take a walk and learn to think. OLPC shipped over a million XO-1 laptops so far to third world children, they are ALL running Linux Sugar OS so far. The Windows talk is only about satisfying the politicians who are the ones deciding to order the laptops and they only know about Windows so they require it to be on the laptops.

In my opinion, some mix of Windows CE, WinMo and Windows 7 for ARM will be good enough. Nobody really wants any of the usual Windows applications. What people want now are applications that do basically the kind of stuff that Android applications can do, just for bigger screens that are on laptops. And if you need multimedia or 3D acceleration, then that can perfectly well be available as well as DSP cores working with the ARM architectures systems on chips.

There have been some comments on the relation between OLPC and netbooks by manufacturers such as Asus, MSI, Acer and Dell.

In the first place, the OLPC project was a great help in starting developments of Netbooks. The interest from the western world in the OLPC was one of the aspects which prompted Asus to develop the EEE PC. It probably also mattered that the OLPC was made by one of the major rivals of Asus.

Some complained the netbooks have done much better than the OLPC and have made the project redundant. While it hasn't always been a smooth ride with the OLPC project, I would have to strongly disagree.

If you just look at the western market, the OLPC is indeed made redundant. I don't think it ever actually stood a chance, even without netbooks. But the OLPC was developed for use in Africa, which means there are quite a lot of requirements to think of which are irrelevant for a product designed for the western world. And some aspects highly valued in the western world are less important for laptops in Africa.

The OLPC is, for example, designed to be a lot more resistant to extreme environments and to have better energy efficiency, while performance is less of an issue.

Netbooks manufacturers were considering another market, our market*, hence they are much better suited to our needs than the OLPC. They wouldn't last for long in harsh environments, but they don't have to. Energy is less of an issue. There are plenty of power outlets available. Performance on the other hand is more valued and running regular software smoothly is a requirement made by the majority of the consumers.

The OLPC XO-1 still matters there where it was developed for and I expect the same from the XO-2.

Charbax, the revolution you describe is not going to happen, because it involves MS, Apple, Intel, NVidia, VIA and all other parties involved with netbooks (and by your vision all computers in general), doing nothing.

This article itself is proof to the point MS isn't sitting still. It sought to involve itself with the project.

I'm certain OLCP XO-2 will influence the netbook market, but not as much as the XO-1 did. The fact remains that the western world has different requirements, where performance plays a bigger role. The main influence from the XO-1 was to initiate the netbook market. The netbook market is now already in place and following its own path, which has similarities with the OLPC project, but isn't exactly the same.

The current trend I see with netbooks is that the demand for better performance and more features increases and is (going to be) met by several manufacturers who you discard as having no future in the netbook market.

There is also a healthy interest in better battery performance. There is indeed going to be an ARM netbook** with well over 10 hours of battery life, but there is also going to be the MSI U115***, which is sporting an ATOM CPU and also provides well over 10 hours of battery life. The first has the price benefit and its modular design. The second has the production capacity, the branding and the ability to run Windows.

Part of the western market values Linux OS, but there is also a much larger part that can't do without Windows. At the same time MS, Intel, etc. have a healthy influence with manufacturers, which they no doubt will be using to keep this as it is.

Another point is that while Android does have the required backing of a large firm, it's not exactly ready for the market yet. There is also the danger of the 'divide and conquer' effect, which has been troubling Linux.

Basically I strongly doubt points 1 to 3 and 5 will happen. And while point 4 stands a chance, I don't think it will be because of the OLPC XO-2, even if it will play its part.

One of the points where I do think the XO-2 will have its influence will be the screen. Or rather, I think Pixel Qi screens**** will be applied in both OLPC projects and netbooks. Pixel Qi is a company started by ML Jepsen, who successfully developed the screen for the XO-1. Their site makes great claims. In reality those are (disputable) aims for the future, but their upcoming product would add a few hours to battery life nonetheless, while offering better performance and keeping a low price. Using her name and earlier performance with the XO-1 screen, Jepsen managed to interest several manufacturers in taking one of their screens into production. Based on their descriptions, I would say they stand a good chance at getting the consumer market interested in their screen as well.

In simple terms, for an innovation to be successful, the new product or service needs to have the advantage over the previous product, in sense of costs, performance and these days environmental effects too. Further, there needs to be a market interest in the product, or at least a potential for creating market interest. There needs to be an interest from producers in making the product. Market and production barriers from existing products need to be overcome. Legislation also needs to accommodate the new service or product, although that is not much of an issue here.

Most innovations fail at one of these points and never make it through. The screen stands a decent chance at success, because things look positive on all those aspects. The other points you raise are met with a lot of resistance on one or more aspects. The more resistance, the smaller the chance of success. Some of the Lock-In's in this world are very strong. The OLPC XO-2 won't be able to break them and to be honest, with some I don't think we should want them broken, not by the XO-2.

Solinx
www.msiwind.net

* Initially not the same market group they are serving now, but a western market nonetheless.
** http://www.alwaysinnovating.com/touchbook/
*** http://msiwind.net/msi-wind-u115-roundup/
**** http://www.pixelqi.com/
***** I have the feeling this place doesn't accept returns... well done on reading it all if you've come this far.

No, the binary translator/emulator is just a program loader and small run-time to host the running x86 app. It allows it to use the underlaying win32 environment as though it were a natively compiled app. DEC wrote one for x86 on Alpha (FX!32) and it's not a big app... in fact, you can still download it if you want to see. The downside is the perf hit, but it's only for the running app, and in the real world, for apps like office etc, even a 50% hit is often transparent to the user.

What utter nonsense. I have the Win7 beta and Office-2007 running on an EEE 900 and it performs as well, and in many cases better, than XP.

Why not just use windows ce?

Or the HPC version of CE?

it works fine. I know its not compatable with windows programs, but a lot of programs are made for it. ANd its pretty efficient (more then windows).

How about instead of just windows for arm you get down to the real world question. If XO wants to succeed it needs windows for the nay-sayers, but for the applications. There is plenty of open source support for windows. In fact every major open source project has a win32 port. Just have open source run all major functions like, firefox, mplayer, gstreamer, Open Office, aptitude....
Then when people change user land, they have all the major landmarks.

"We all agree, however, that Nicholas Negroponte's mouth is his own worst enemy."

And all the continued good work on Sugar and the rest of the stack is not exactly thanks to Negroponte - it's because the community stepped up to the plate when Negroponte last played his "Davos card", belittling the efforts of the open source community, and trying to rub shoulders with Gates and company.

For all we know, Negroponte might be playing a political game, but he does so incompetently: who wants to work for a guy who publicly trashes your work and what you believe in? Even if the XO-2 arrives, the action will be elsewhere. Perhaps that's why Negroponte wants someone else to play with him so badly.

This feels like OLPC trying to use MS to further his "Intel is Evil" campaign,

In doing so, he vastly over-estimates the importance of his own company, and the extent to which anyone else - especially Microsoft - is likely to care.