If ARM based mini-netbooks (ARMbooks?) are done well, I think they will sell like hotcakes. They could be what many accuse current netbooks of being. i.e. web surfing & e-mail machines. In the case of ARMbooks, this will not be a bad thing. A cross between the old CE based pocket PCs and current netbooks is a good analogy. $150 for a machine that serves up the web better than a phone and is smaller, lighter and has better battery life than a netbook. One can hope.

Again, the Pixel Qi screens will also be key to this ARM migration (and the ability of netbooks to go to lower power (only power as needed as described in these videos about portable devices being all about the screen, low power, sunlight readable, and batteries that recycle into fertiizer:
http://bigthink.com/maryloujepsen/ideas ) will be both revolutionary and evolutionary in the computing and environmentally friendly needs of the computing world as we go forward.

Thanks for posting that, I hadn't heard about Pixel Qi. It sounds like they have exactly the technology netbooks need the most -- getting away from conventional backlit LCDs might be even more important than getting away from X86.

How could it be "underpowered" if it "capable of handling most desktop apps"? Only certain marketing departments want consumers to shy away from the netbook because they've heard it's "underpowered". It is an insidious meme not worth spreading.

Quote: "Or would you rather have a full computer?"

This question doesn't make sense in the context you are asking it What is the definition of a full computer? Something with an X86 processor? A computing device that runs Microsoft Word? A computer with 3D graphics capability? A computer with at least an 85 key keyboard? Define the term "full computer" and then decide whether an ARM based system meets that definition.

I have a suspicion that you mean a system that will perform all the the same tasks that a person might use a desktop computer for right now including any random Windows based application. My opinion is that the lack of Windows compatibility doesn't by itself define the term "full computer" Rather I'd say Windows compatibility defines the term "Windows computer"

The question you should really be asking is, "Are you willing to give up Windows for 20 hours of computing in a 1 pound package". To that, I'd answer with a resounding "yes".

Regards,

Hans

Yes x86 == real computer to most people. The question is whether that perception is changable. And real computer == Windows is just as much a perception that must be changed. ARM is going to have to seriously deliver on it's advantages to have chance against those two mental blocks. The potential is there though.

And to address a bigger question raised in the article about Microsoft porting to ARM. Not likely. Microsoft has attempted ports in the past and found the ISVs totally uninterested. Windows NT ran nicely on PPC, MIPS and Alpha. XP was ported to Itanium and X86_64. Zero ISV support. Not even Microsoft's own Apps division bothered. Digital finally did a x86 emulator called FX!32 (I think) so Microsoft Office would run on NT/Alpha at speeds any off the shelf Pentium could smoke, thus creating the impression the Alpha was a wimp. Adobe STILL hasn't released a 64bit Flash or Acrobat plugin for windows, which is why you get both a 32 and 64 bit IE with Win64 and everyone runs the 64 bit copy because it has plugins.

Now consider all of the previous platforms have been faster and better than x86 while ARM is slow and rather limited. Native compilation will be a serious problem and I don't know if Microsoft's developer tools have ever been configured as a cross compiler. Just where does one buy an ARM developer's workstation? Something with a few GB of ram, big drives, DVI port to drive a big assed monitor, etc.?

How many copies of Quickbooks would Intuit actually sell on Windows/ARM? Probably fewer than they would sell on Ubuntu(or RHEL)/x86 and they haven't bothered with that one yet. Thus it would be likely that even if Microsoft ported to ARM the ISVs wouldn't, or wouldn't before Microsoft pulled the plug.

Not sure I agree. The small instant on CE devices of the past were not as popular with consumers because they were at almost the same price point as a full laptops not due to lack of features. Back in the Jornada 720's day (late 90's) we were talking serious amounts of money.

I for one will be getting a cortex netbook when they become available (would love one with either 6" screen or 7" with small bezel so it will go in the coat pocket). Being it will last 10+ hours I only need the device it-self, no massive extended battery, no bag with charger etc.

Still regularly use a smartbook G138 from 2003 with a 200mhz strongarm and CE 4.2 (it is reliable, stable, has long battery life and is actually surprisinly fast for a web browser - in fact it is considerablly faster than the latest crop of Chinese netbooks with 400 and 533mhz arm cpu's and ce 5). Can easilly turn on and check a webpage quicker than a netbook returning from suspend.

It is also a lot more convenient to take contacts, tasks, jot down ideas etc. as they come to me due to instant on.

There always seems to be loads of people mumping they don't run Windows can't see why that makes the device a bad device. If they need windows then it is not right for them and not automatically "rubbish".

The latest crop of cortex devices will be able to make a device that will fit in a coat pocket, will run all day, play 720 HD, run 3d graphics etc. etc. does not seem like it is too bad.

If they are cheaper, nicer looking (slim and sexy), quick for day to day tasks etc. but also come with loads of battery I can easilly see a good number of people swapping from a netbook with 6 and/or 9 cell battery for one.

Another reason they may appeal as price point almost get's disposable/impulse buy teritory which is ideal for a device for kids.

People like the current "netbooks" because they are, for most intents and purposes, just a tiny laptop and offer the core functionality of a normal laptop (with better battery life and a lower price). People want the ability to perform everyday computer tasks (email, web, office apps, basic photo editing/storage) wherever they go; size, weight and battery life are key.

I believe the successful netbooks will be 10" screens with 10+ hour battery life and thinner profiles than most of the current netbooks. Basically, people want a computer like an 11" Sony Vaio TZ, but with better battery life and without the $2k+ price.

As for the "ARMbooks", it seems that the basic niche they're aiming for is already pretty full of devices that are capable of email and web browsing (iPhones, Blackberries, Nokia N8000 and similar devices).

Funny cos you just described what a Cortex netbook will be best at in the list of wants

email, web, office apps, basic photo editing/storage like a Sony Vaio TZ

So which of those tasks can't linux do? And the cortex cpu devices are going to be slimer and lighter than the vaio whith 10+ battery???

To confirm you list a list of wants (all boxes ticked by cortex netbook) then say they are for a diffrent market???

John

I'd say that this years Cortex-A8 ARM netbooks won't have quite enough kick to them for most people, although they're an important step.

By the end of next year we'll see Cortex-A9 dual core ARM netbooks, and that along with more mature desktop software support I think they'll be appealing to lots of people.

As for porting Windows or beefing up Windows CE, there's an important factor you might be missing. A big part of what makes up a Windows computer is third party drivers and software, by a gaggle of different developers. These have all been produced for x86 for the last 15+ years.

For a Windows desktop on ARM to be viable, the third party stuff has to be ported as well, and Microsoft can't do that themselves.
This is where a Linux desktop has the advantage - because everything is open source, when you move to ARM you don't just get the operating system - you get every device driver and piece of software too, all re-compiled for ARM ready for use.

I'd say Microsoft will want to avoid getting into a fight with Linux on ARM because it will be much weaker there than on x86. And so I would expect them to avoid "lending legitimacy" to ARM as a desktop platform by making a version of Windows for it.

Palm's WebOS is a good candidate: it's basically just a webbrowser.

My issue with that report is that it assumes Intel is going to remain static on power usage while battery makers will not improve their batteries. If it means risking losing market share, Intel *will* respond, just as they did when Transmeta introduced power saving modes and when AMD started smacking them around on benchmarks. 2012 is a looong ways off in the tech world.

They can try. But ARM's low power abilities are way out ahead and come from features that Intel could only emulate by no longer being x86. You won't build a a fast x86 core with under a million transistors. Intel already has the advantage of several processs shrinks over most shipping ARM product so there isn't much advantage possible for Intel on that front. They will have a heck of a time producing a fully static design, which means idle power is going to be orders of magnatude higher.

As for better batteries, that helps both camps equally because there isn't such a thing as Intel batteries and ARM batteries, either can use any new battery tech.

Right now the other subsystems in a notebook tend to mask just how bad x86 tech is on the power front. The CPU gets mixed in with the display, drives and wireless power consumption. But flash is pushing drives to an asterisk on the power chart, the OLPC display tech and LED backlights are whacking away at display consumption. Wifi is a problem because it was designed at at time when portable computing was P4 laptop, but 3G and WiMax are newer wireless designs and newer WiFi tech is working on power consumption as well.

All these trends tend to make x86 look worse. Intel has to fight upstream against all of those forces. But no, don't expect Intel to just lie down and die; they will attempt something and they just might have the resources to pull off a miracle.

My last issue is re. the options in the poll. Linux is a full desktop OS so where it the option.

Arm Netbooks (well cortex) "the hardware and OS is fine - as long as price is right".

So there is no suitable option for me to vote

"Wrong they are" my Yoda brain says.

Look at the progression of netbooks in the last 24 months.

- 7" screens to 9" screens to 10" screens and soon 11" screens
- Low end Via processors to Atom to slightly better VIA Nano and AMD Neo
- Tiny paperback size cases to cases the size of a sheet of paper
- Cramped 85% keyboards to nearly 93% full size.

The trend is to make something that is just powerful enough, just light portable enough and just cheap enough. I have no doubt that ARM processor units running slimmer OS will be a part of the netbook market, but I think they will be the the niche end. Netbooks seel because they DO things, currently they bring nearly the full computer expericne down to the smallest most affordable form factor.

The other issue is that Win 7 will likely run 'well enough' on most netbook made today and will run even better on the next series of machines. Microsoft seems to have decided that the low end will be current Atom....super smart planning or dumb luck? Who knows. However, I actully think Win 7 will have 70% of the netbook market and the other 30% will all other OS platforms in 18 months. Netbooks are becoming more mainstream and people will look for the same old OS they have come to know.

What would make bigger ARM devices hot won't be the hardware or OS tech...it will be the service backend. Can someone make a "fat iPhone" experience. Can these machines set up an Applestore to support these machines. If 'hot apps' are not stressed, then who cares? I think that is where the disconnect will happen, the "service end" will be lost from the concept and that will make these bigger ARM devices stall. Meanwhile, Win 7 will still do that "computer" experience just fine, old thinking, but viable for another decade.

"However, I actully think Win 7 will have 70% of the netbook market and the other 30% will all other OS platforms in 18 months."

That prediction only holds if teh netbook market remains pretty much like it is today, i.e. 10 inch screens, 160GB hard drives, $400-$600 pricetags. Probably not a safe assumption under any circumstances but in a down economy it is especially unsafe. What if the average selling price is $300 and most of the machies don't have an Intel CPU?

There is a whole world under the minimum Windows machine waiting for someone to exploit it. Dozens are rushing into the ARM space and all must fail for your rosy scenario to come to pass. If just one gets the price/quality/featureset in a sweet spot where people start buying in quantity it will open up a huge land rush to own the newly opened market segments and the losers of this Xmas season will be back for round two in '10. See ASUS and the original eee and what happened in the following months.

I'll throw my hat into the "Linux is plenty capable of being a full, real PC" ring.

I'll also thank you for throwing in an idea I hadn't really thought of more fully until just now: maemo based laptops. The Nokia internet tablets really have lit a fire under some developers and hackers out there...I can't imagine it being that hard to make it work more like a traditional desktop. It could also solve the perceived "app store" issue: just have nokia run it.

I'm all about two things: base usability(Web, maybe emugaming, writing the great american novel, and communication with friends), and battery life. Low price and portability are important too, but honestly, at this point aren't they simply assumed to be part of the deal with these machines?

I just wish the future would get here already. There needs to be a *continuum* of these devices, not simply more atom based boxes in a 10 inch form factor with a gig of ram and 160 gig HDs running XP....

Your poll is skewed. If ARM runs Linux, then it is a full service computer. To put the "No I want a full desktop operating system" in your poll is misleading.

Belatedly I realized I probably should have said "full desktop OS
experience". While you can run various Linux builds, including Ubuntu on an
ARM processor, to date I have yet to find one that isn't either stripped
down or extraordinarily sluggish when compared to similar operating systems
on x86 processors.
I'm just going to leave the poll as it is because it's easier than pulling
it down and replacing it. But next time someone remind me to write a poll
question that can truly be answered with a simple yes or no. :)

you run linux on a cortex then?

or you talking 206mhz SA or 400Mhz xscale?

not really the same is it?

I mean the demos I've seen on machines at trade shows with new CPUs like the Qualcomm Snapdragon 1GHz CPU.

what in a linux build not optimised for the cortex so probs falling back to a version of arm v4 command set.

and being it is a prototype I would have thought the drivers although workable are not fully optomised for gpu etc.

You seen how quick the Nvidia Tegra displayed the youtube homepage? and it is an ARM11?

John

Yes, ARM can run a "full OS" but I don't think that is where the money is in the niche it can make for itself. Netbooks are already about as small as you can get a "real" computer and still be usable. The ability to run Windows if you like is just icing on the cake. My vision of an ARMbook sits right between the netbooks of today and the smartphone/PDA market. Yes it could do more than I stated, but would you want to? Netbooks and laptops can do those tasks, but they can't do it at 1 - 2 lbs and run many hours with a tiny 2 cell battery.

Yeah, a replacement for devices like jornada 720, mobilepro 900c, Sigmarion 3 etc. but no reason for it not to run right up to 10" or laptop size.

Think all the professional writers/reporters out there where 10-20hrs battery would be invaluable. There are still some writers and reporters use J720's and Psion 5MX's etc for this very reason.

Also it will allow you to make a netbook with 10" screen that is so thin you could slip in a thin pocket in a ringbinder and hardly notice it is there. Add to that they will come in cheaper.

I think the cortex/arm based netbooks that are due out soon have a place in the market and will have good reasons why some would choose them even if the device is filling the same market as current netbooks.

So I take your idea further and suggest the arm units will fill the old Handheld PC size right up to large netbook with a reasonable crossover. There are also enough linux builds out there with a software repository large enough not to worry.

The cortex or nvidia tegra (arm11 with media processing and GPU) is not a step down. Anyone who has seen the 3d grfx or 720 HD demos on these devices will be happy to attest it is more capable than current atom.

So far we havn't even seen one with linux fully optimised for cortex.

I am looking forward to them

What about some other industries. Corvertable touch screen tablet/netbook great for surveys handheld units as the comms are onboard and battery will last all day without breaking your arm.

Industrial/medical use? Again all day battery and the cortex machines have no heat issues and require no venting holes so can be sealed to the environment better?

There are lots of pluses. but with all these things. they are neither crap or the best thing ever. they just have strong points and each customer will need to weigh them up.

however think it is diffucult to draw meaningfull results from a survey until we see hardware ready for sale. Sometimes manufacturers cost cut and rush things out the door and the finished article isn't all it could be.

Fingers crossed

The People wants computers, full power computer not tablet or pocket pc with big screens and they also want cheap computers.. _That's all If someome bring me a new "netbook" ARM based computer saying that's enough for me since I don't use powerful apps I will send him to the hell.

So don't buy one.... doesn't make it a bad computer. If it runs ubuntu why would you feel the apps are not powerfull enough or limited? esspecially if it can perform as well as atom machines (though we need to wait for production models before we can say for sure if power is up there).

ARM will never replace x86 - at least not by 2012, or 2020 for that matter.

Will there be ARM based "netbook" devices? Sure. There already are clamshell devices using non-x86 CPUs, (like the 3K RazorBook - which runs Windows Mobile) - and there are plenty of MID/UMPC style devices that use ARM, (the Nokia N-Series, for example). The hopefully won't be vaporware for much longer "Open Pandora" gaming UMPC is ARM based, clamshell, and pocketable. But it costs more than the average ATOM based netbook these days (again, assuming it ever actually exists).

One of the huge selling points for the majority of netbooks is that you can have ultraportability, low cost, and a full desktop experience. I mean look how many people put not just Windows - but Mac OS X on their systems. Unless they somehow hack the iPhone OS to run on non-iPods/iPhones, no ARM based netbook will be able to do that.

Now, if these ARMbooks could retail for under $100, and still have a decent screen, 90% or better keyboard, WiFi, BlueTooth, etc. - then they might have a shot. But it's going to be hard to justify giving up all the OS options you get with Atom based netbooks without a very large price difference.

I agree with Chad.

I had a HP Jornada when they first came out, an Asus 701eee and now a HP Mini 1000 with a 10" screen.

I would buy any inexpensive system that (1) has a 10" screen with similar resolution to the Mini (2) has a 90%+ keyboard (3) lets me do routine word processing and spreadsheets (4) lets me email and surf (5) ideally is 1 - 1.5 lbs and thin (6) is inexpensive and - here's the biggie - (7) does NOT require me to learn a new OS or to rely on highly techie friends.

So... if someone can put out an OS that is intuitive enough for a solid mid-level users, which runs downloadable apps when I need one (for example, when I realized that Works and Word are incompatible with HTML while I was traveling), then I'm interested.

Let's all remember that a survey not that many months ago said that Linux netbooks were being returned at far higher rates than those running MS. Remember that, once you get to the mass market, familiarity with an OS matters. There are plenty of people around who aren't super-competent even in MS, so don't expect them to muck around with one that requires a real learning curve.

And my super techie friend, who did the Linux mods on my Asus eee stopped using Linux on his own system because it was just too much work.

Come on folks, be realistic about the masses who would need to purchase these systems and realize that they HAVE to have mainstream OSes.

Yes, the easy of getting the apps you want is vital. That's one of the (IMHO, the only) things that makes the iPhone better than the rest - the App Store. Make an easy to use ARM based OS with an App store (Maemo, Andriod, iPhone OS), and put that on a sub $100 un-contract'd device with WiFi, qwerty, BT, and decent screen - you'll have a winner. But, it's not going to kill the netbook market.

And think about this, (another commenter brought it up, but still), do you really think an Asus Eee PC 1004HE is going to still cost $400 in 2012? Do you think it's going to cost that in 2010? The Atom/Via Nano based netbook market is only going to get *better* in the next 3 years - CPUs clocked over 2.0 GHz - 4GB of RAM - 320 / 500 GB hard drives - maybe even dual core, all without raising the price - that and/or the price of the 1.6 Ghz models will drop. I mean you can easily find a 1.6 Ghz Atom netbook with WiFi, BT, 120 GB Hard drive, webcam, card reader, a 3-cell battery, and Windows XP for under $250 these days. I've seen the old school Eee 701 on sale for $199 at Best Buy.

Again, unless the price is drastically lower than the x86 netbooks - I don't see any justification for ARM "taking over". Even if you could pick one up for $75 - $100 - it's not going to take over, just add yet other tier.

The higher return rate FUD has been debunked. Yes if you ship a preloaded Linux that won't access the webcam you include you are going to experience a higher rate of return. ASUS reports the same return rate, but they did a good job on their preload.

As for apps, again, ASUS is the one to follow as they have a download site where you can get a boatload of apps with a 'click here to install.' and if one knows enough to enable the Xandros and/or raw Debian repos you can get all sorts of things with a few clicks in a package manager.

For ARM to dominate we will need to see one of the upcoming models ship with the same polish and ease of use. Put a nicely polished system on fairly pretty/solid hardware, price it to open up a market and watch the fun. If it doesn't happen it will be evolution in action. Get stacks of them in stores for Black Friday and use em as loss leaders and come Jan 1, 2010 we will be in a new world where x86 is in a war for survival.

Because Windows and x86 are in the same boat, they are both long since past the point where we NEED them, people just don't realize that life is possible without both of them. Let that idea ever break out into the mainstream and putting that genie back in the bottle will be a fun fight to watch. But it has to be both, it is now pretty clear that any x86 system is going to get Windows cramed down it's throat. WinTel lives or dies as a pair. If customers can't be broken out of the notion that all computers are Windows (except those rich wierdos on Mac) x86 also wins.

Forget the current crop of arm netbooks from china like the razorbook. They are cheap rubbish that should not be on the market. My StrongArm 206Mhz device from 2003 totally outperforms them and not by a small margin.

The new cortex netbooks are looking at a price point of $200-$300 to start and that will be on devices hitting the market Q3 this year and mark my words they will be considerablly more sexy than the current crop. When the mass market is presented with larger, less sexy, less battery etc. but familiar OS and the cortex option which is sexyer, takes less space, runs all day - the choice becomes interesting and familiar OS isn't the only consideration when the other devices look so sexy.

A lot of ARM units look like they will also come with a touch screen (not all but a lot of them) making them smaller for same device and on a 7" screen a touchscreen is a very nice thing to have.

Lots of years ago I used to think a small device with x86 and full windows would be great but after going through 10 or 20 devices from laptops to PDA's to tablets to Handheld PC's I found it was the CE device that actually became the most productive due to portability and silly long battery.

Doesn't take many people to buy them and start using them for people to notice and re-evaluate their requirements.

Actually was quite funny; the number of times I was on a aeroplane and as soon as the fasten seatbelts light was out I was up and working in about 2 seconds while some corporate person was still getting subnotebook out of the bag. Then we hit some turbulence and I had my device off and stowed in about 2 seconds while he was still saving the doc he was working on. Ended up so many times in a conversation where they asked what I was using and every time they actually took device details to look for one.

I am just looking forward to seeing some retail hardware so I can see the speed etc. as it is about time my venerable StrongArm device was moved on to give me a boost in speed.

John

There's an important market for ARM-based netbooks that I haven't seen discussed much: K-12 students. First, any price difference is hugely important in this market, since savings are multiplied over the thousands of students in a district. Secondly, battery life and weight make a big difference in schools. Kids need to use a laptop all day without recharging it, and they also need something light. Finally, the hassle of switching to a new operating system doesn't mean anything for an elementary school student, when the IT team for the district is managing the machines.

Currently, the student-computer ratio in U.S. schools is only 4 to 1. It's only a matter of time until schools jump on the netbook market, and ARM-based netbooks might just be the ticket (especially if they can run video).

Intel always intended the Atom to eventually compete with ARM processors for speed, price, and battery life. This market segment will get VERY interesting if they accomplish that goal.

I have at my disposal 5 portable devices: Nokia N800 Internet tablet, Irex Iliad reader, Asus EEE-PC 900-Linux Xandros, MacBook and 6 years old elephant of Fujitsu-Siemens type, about 3.5 kg and 2.9 GHz clock (rather dragable than portable). The first two are ARM based (RISC; see the end), the rest is Intel (CISC). All these things run mostly unix derivates, mainly Linux, except of the triple-booting elephant which in addition to Ubuntu and SciLinux CERN also has XP-home from the shop. With this broad specter, I can say one thing: I could not use any of the things out of the box for any serious work without doing much adjustments. Without internet information, I could not have used any of them for anything useful for my work. All of them came with a certain functionality which the manufacturer considers important - not for the user (customer) but obviously based on some marketing magic.

This magic seems to propagate even into the so positive offerings as e.g. Ubuntu, which also starts to converge to the general idea of a user as a half imbecile who screams in horror only one line of typing instead of clicking is exposed. Particularly on Windows, there is this fantastic option "hide known file extensions", which also propagated to OSX GUI on Mac, and seems to gain foothold even in some linux applications. Not to speak about filenames with spaces and national characters, which are a true joy for any scripting and mass operations.

Well, after you switch off all the rubbish which was supposed to make your life "easier" (whateverver that means), you must get hold of all the software which you need. For the three last boxes, either it exists in various repositories, or you can get the source and compile - build yourself. I had a disk crash on my MacBook and the new adjustment took about a day since I had most of the packages on the back-up (next time that would happen - it would be 5 minutes - I have a bootable back-up disk now)
For the two small things with 128MB (Nokia) and 64MB (Iliad) boxes you need an SDK running on something else to compile your sources. In both cases you must add yourself the command line ability (x-terminal and shell), for this you get a certain help from the manufacturer.

But am I also talking about the windows machine? Oh yes. When it came, it was completely useless. It was the first XP I have ever seen. Fortunately I had all my editors and my customized version of Cygwin on some CDs, which I previously used on win98SE, and after some days I could use the elephant that winter 6 years ago.

The Xandros on EEE PC is basically Debian. Some of the things I compiled for the older Ubuntu on the elephant I could simply copy to the EEE-PC and it works. Matlab works (with department license over the net). The whole GNU system was on the EEE PC and modification of the IceWM which appears as the windows-like GUI just happend by copying the configuration file into the users directory. EEE PC is thus a full blood workstation of the 2004 standard (also after about 1 days work).

All of these five are underpowered. Some real work I must do on national supercomputers or desktop machines for some graphics (all run Linux or Unix). Many things I can start or control from terminals on the portable devices (in principle also the Iliad, but the screen is very slow and I gave up on installing a full internet connection on it - but I still have the package).

The bottom line - what makes windows to dominate: During the years anybody who does any work or has any hobbies transferable to computer has collected a personal collection of software and habits, better or worse assembled on CDs and USB sticks and other storage means. Some of this is quite old, sometimes better newer versions exist, sometimes the ancient versions are better (at least for the person in question). All this is "installed" in the new netbook when it runs XP home and the person is happy.

It does not take much to figure out how to make that person happier: first of all, the person must be educated. The person must stop reading the nonsense published by PC-journalists, people who do no real work. The academic and professional communities, people who really use computers for work must come out from their professional dark holes or ivory towers and tell the people what IT really is and can be. There is not much place for windows then, nowhere.

But it also requires an easier building of applications on linux. The frenetic use of newest libraries should be avoided. The influence of the windows philosophy, which definitely exists (just see otherwise excellent Google chrome) should be avoided. These things should be discussed by more than the present reader.

RISC : reduced instruction set computer (PA-RISC, PowerPC, Sun SPARC, ARM; supercomputers and cellphones)
CISC : complex instruction set computer - now mainly Intel (Desktop computing - WHY not RISC?)

Funny as CISC chips as you call them (x86) actually have more in common with RISC now than the way CISC is meant to be :)

I entirely agree there is no good reason for X86 over ARM cpu's it all depends on software and how it is implemented (speed of the GUI for want of a better description).

there are very few people actually need or use software that is only available on windows on a netbook sized device.

Look at some of the multiple ARM devices announced recently I think you will likely find ARM cpu's will power one of the big Supercomputers soon. And it will likely power more and more Servers (less heat, less power but it will give same computing power ok so maybe over multiple cores' or cpu's but does it matter how it gets there in a server).

Once ARM is in servers and Supercomputers (One of MS's big markets is in server software - you think it won't make a version of windows for them?). So many servers are using Linux of BSD these days a switch to ARM or any other processor architecture is no big deal.

Most industries computer included suffer a 10x force once in a while that just changes everything completely in a short space of time. Computer industry has had a few so far.

IBM Choosing a 4bit intel chip for it's first PC??? when there were multiple choices that were so much more powerfull. Intel were as good as bust till IBM made this strange decision. Another was IBM as good as handing DOS to Mr Gates.

A lot of this is just me thinking out loud but if ARM gets it'self as big in the Netbook market and Big in the Server market (this is where cores optomised for power usage and chipsets to go with fast RAM etc will get developed) then it has every chance of coming back into desktop machines.

With energy getting ever more expensive there actually may be other reasons for ARM making inroads to server markets cost of the electricity is likely to be a big reason.

Some people are saying ah but ARM it's a PDA processor. Nope it is just a processor. And the cortex is quite a jump forward.

Disagree if you like, opinion makes the world interesting but doesn't mean it will or won't happen.

John

I don't see ARM powered clamshell devices taking off.... unless they're also mobile phones. Why get a tiny device with long battery power when you can already carry around a small device with instant on and long battery life that does internet and email, eg. an iPhone, Pre, Blackberry, etc.

I think netbooks (most of which will run Windows, whether you think that's good or bad) are going to stay popular, because they are a full computer experience that's also very portable and very inexpensive, and they complement a desktop and a good mobile phone by being right in the middle and offering the best of both worlds, without being complete competition for either.

As long as netbooks keep improving like they have been while keeping the low prices they are at now, they will always stay popular, and as long as mobile phones keep improving at the rate they are, small ARM devices won't take hold unless they are part of that mobile phone market.

That's my opinion, anyway. I certainly have no intention of getting a small instant on computing device if my cell phone can already fill that role.

try typing a document on your phone :D

If I'm in a situation where I didn't have a netbook handy, yeah, a Pre would be good enough for typing a document... The only way I would carry a small ARM device is if I had a bag with me, like a bookbag, and in that situation I would just as easily take a netbook with me. If I don't have a bag with me, I'm not going to carry anything more than a phone with me anyway. And I'm sure I'm not the only that feels that way, which is why I see netbooks with full laptop features taking off and smaller, weaker devices staying a niche market.

Ohh I will have to say that you are off base on this one. Fist off, the Atom and other x86 compatible processors will increase battery efficiency and capability as the Atom and Nano processors continue to push for decent performance and lower wattage. ARM isn't that much cheaper when you take into account developing or porting an operating system on a specific ARM based platform for any device, and application base will be to broad for ARM to truly shine in this category. I would venture to guess that we will see next generation net-books only increase in battery life while maintaining the "Desktop Computing Appeal". We have seen a negative reaction to small not quite notebook devices already.. Remember the Palm one? What was that? It was a flop. Besides Phones will dominate in that market, the consumer is already familiar with them, and as you add programs, web browsers with more capability and decent resolutions, thats really all you need. Besides the performance on smaller ARM's is pretty abysmal with larger screens and better browsers. They are not going to be happy with ARM architecture performance on a package or upgradable browser plugins, and if anything this will create a market fad. Just my two cents.