Slashdot Mirror
Intel Relying On Ice Cream Sandwich For Tablet Push
An anonymous reader writes "Intel thinks tablets live and die by their software, not their hardware. So as they get ready for a big push into the mobile device market, they're relying on Ice Cream Sandwich to provide competition with Apple's products. From the article: 'The company has largely watched from the sidelines as mobile device makers have used processors based on ARM's microarchitecture to power their products in recent years. This despite the fact that Intel actually predicted the rise of what it called "mobile Internet devices," or MIDs, several years ago, and built a chip, Atom, for such gadgets. For all that [Intel CEO Paul Otellini] touts the software over the hardware when it comes to tablets, Intel knows it's got a lot of ground to make up to wrest design wins away from ARM. The Medfield System-on-a-Chip (SoC) is a promising but still uncertain step in that direction.' Otellini thinks the tablet market will get much more competitive over the next year as ICS devices mature and Windows 8 devices arrive."
Microsoft didn't wait and see.... Windows CE was around on tablets (including ARM and MIPS-based ones) for a long time before Android ever existed. They were typically called Handheld PC or Palm-size PC devices. Windows CE 2.1 was actually pretty tolerable on the HP 320LX and Sharp Mobilon HC4100 I had. Never liked releases much past that.
Apple also had "tablets" long before Android, iOS, etc. The Newton MessagePad of which the 2100 was actually really nice and the eMate 300 bit slow but cool nonetheless. NewtonOS 2.1 certainly didn't suck.
Linux and NetBSD have also been capable of running on such devices for a long time as well. I owned a few WinCE devices over the years and a couple of Newtons.
There have also been x86 tablets since the early 90's. Dauphin DTR1 and there was a tablet Thinkpad as well.
Did they have goofy oversized widgets for sloppy finger-based simple computer usage by retards? No. They were pen based. You know..... for functional useful software in a professional environment instead of a web browser on steroids for morons that can't type or write legibly anyway.
What? Intel has been announcing they are just about to release a mobile chip for phones for at least 5 years. And the chip will totally dominate and kick ARM to the curb. And with Apple's recent reorganization of the computer hardware industry with the iPad, they're complete failure to actually release such a chip just hurts more.
Finally, both Intel and MS deciding that they mustn't hurt their margins entering this new market means their less functional, more power-hungry systems will still cost more than existing Android/ARM and iOS devices.
Yup, it's a recipe for success!
Sleep your way to a whiter smile...date a dentist!
Just like microsoft had a wait and see attitude with mobile phone OS then iOS and Android swept the market and then they released windows mobile 7 to a world that didn't care.
I don't think that's a fair comparison. Microsoft entered the mobile device business long before Apple and Linux, the problem was that they sucked. Really bad. Until Windows Mobile 7 I always told people "whatever phone you get, make sure it doesn't have Windows on it". Now nobody asks me anymore... the question is "should I buy an Apple or something with Android on it?". So you're right about the not caring bit.
What makes you think Windows 8 will be terrible?
Even if Windows 8 is terrible, personally, I think Android will end up losing the tablet war. The reason is that windows 8 will be able to leverage its existing base of "software capital", and bulldoze its way into the tablet market. Android simply does not have certain critical software (e.g. - MS Word) running on it.
Think of it this way. The mass market desktop pc will die. For the vast majority of users, a simple tablet like device, with word processing capabilities, and media/internet capabilities, is all that's needed. Bulky laptops will disappear too, turning into tablets with Asus "transformer" like capabilities. Eventually, a multitude of device will be consolidated into one single tablet device - a single personal computer. People will want to do everything they did with their desktops, on their tablets. This will include word processing.
What answer does Android have to this?
If they don't fix this, and have their software base ready to rival MS-Word etc. I believe the ending will be very unfortunate, and MS can continue unhindered with their nasty monopoly.
The one consolation might be that Android will continue to thrive in the mobile phone segment, since a tablet form factor is too bulky to replace a phone, unless tiny phones become powerful enough to run Windows 8. Then, it might be curtain's for Android's there too. Why bother with several devices, when one single "personal computer" will do?
And this is exactly the attitude that landed MS on last place on the mobile market. Calling its potential users morons and retards for wanting a sloppy dumbed-down UI, when in reality they were just average users who wanted a simple interface.
It doesn't matter when WinCE was around when it didn't deliver what people wanted.
What?
Bad move, Intel. I used to rely on an ice cream sandwich. Then Häagen-Dazs stopped making 'em and everything else in my life went to complete shit for about three months. Take it from someone who's been down that road: if you're going to rely on ice cream sandwich, do not commit unless you have control of the supply chain.
Thank you, Edward Snowden.
"Arguments from authority are worthless." —Carl Sagan
But what am I talking here... tablets have no point, other than being an e-penis, anyway. Same as SUVs. They are expensive, impractical, slow, and basically all the bad things in one, and the best in none. ^^
In the long run, they will be replaced by mobile phones. Or from my p.o.v., they never found a place that a real computer or a proper smart phone (one at least offering what S60 offered 10 years ago, like a file manager, communications tools, media playback, Internet surfing, install whatever you like, 3d and video acceleration, big display, full-featured hardware) hadn't already taken.
You're entitled to your opinion, but I think trying to read in bed with my mobile or my laptop is kinda sucky, and I think a tablet might work much better for this. I don't see it replacing either of them anytime soon, but that doesn't mean it's not a good form factor for some use cases.
Il n'y a pas de Planet B.
having looked at windows 8, i would have to say that it is what makes me think it is (not "will be") terrible. granted it was the developers preview but even that was enough to make me run screaming in the other direction i have been unable to sleep due to the nightmares its memory has brought me ever since.
---Saying gnome 3 is better than windows 8 not so much a compliment as it is damning with light praise.
What makes you think Windows 8 will be terrible?
I'm using the Developer's Preview right now and I can tell you it's annoying as Hell, and it's not gonna change. Why? Because the things that make it annoying are things Microsoft wants to push.
There's no real Start menu. All programs have to be launched either from the Metro interface, an Explorer window of the program folder, or having the app docked on the taskbar. The Metro-enhanced apps look great on the Metro launcher, but regular apps just get their Start menu files added as tiles.I have several tiles labeled "Uninstaller" but I have no idea what program they uninstall because they aren't grouped at all with their parent programs like they were in folders on the old Start menu. Same with those apps "Read Me" files. But if Microsoft put a regular start menu in people would likely jump right into the Desktop and not even bother with Microsoft's Metro at all, continuing to use their PC like they did in Vista/7. That would threaten Microsoft's plan to steer everyone into using their Metro app store and taking a 30% cut, like Apple does on their App Store. There's also a lack of regular menus in Explorer. It's been replaced with the Ribbon interface. Microsoft sees Ribbons as the future: usability or customer preference be damned.
Why is everyone so down in Intel? More competition is good for us, the consumer! Although software is very important so in hardware, no one wants a tablet that needs to be recharged every 2 hours.
Thanks for that clear answer. Sounds pretty bad to me although I think non-technical users might not care, but what interested me most was the bit about the 30% cut with the Metro App Store. All of this serves to highlight why Microsoft shouldn't dominate the tablet space. But my fears that they will, are I I believe, legitimate. Android very badly needs to think of itself as a proper OS, not just a mobile OS running toy applications. It needs to rethink of itself as being able to run serious software - everything from a full fledged word processor, to Photoshop to Crysis.
You missed the mark... What made the iPhone break thorough was the fine integration work. I had a WinCE PDA a decade ago, too, and it was an absolute nightmare to use for anything... Little things like the power button putting the device into standby (instead of just shutting off the screen) made it useless as an MP3/Ogg player.
The apps were all massively crippled. Pocket Office was inferior to Wordpad. Browsers were all crap, crippled compared to desktop versions, and nobody had figured out how to render full sized web pages on a 240x320 screen. They were still massively dependent on desktops. And worst of all, WinCE was just unresponsive crap. It was laggy as hell on 300MHz+ CPUs when Palm and others were snappy on 30MHz CPUs. The start menu model was never a good idea. And I despised having to go download a REGISTRY EDITOR for my PDA first thing to fix insanely stupid default settings...
Now if you actually wanted to get stuff done, Psions were awesome. Slide-out keyboard. Office suite that allowed composing pretty full-featured documents, even embedding charts and drawings into documents, and printing them out directly to the nearest IRDA enabled laser printer. There, some of the limitations were avoided just because the portrait display eliminated side-to-side scrolling with web browsing and whatnot. Even had a PDF reader, but you'd have to squint to read the tiny fonts, or deal with side-side scrolling every line... the software that makes smartphones tolerable today just wasn't even a dream back then.
Slashdot gets worse every day... Pipedot: News for nerds, without the corporate slant
They were also all based on resistive screens, as capacitive screens weren't around then (I think - at least I never saw one). Resistive screens are rather uncomfortable to use due to the need to apply considerable pressure, but they do have the advantage of allowing for greater precision with a pen than a cap screen and a squishy finger.
how about that on a basic level, you don't try to have a consistent interface between portable devices and desktops/laptops. That's just a no, and a bad idea.
Granted you can turn off metro, but even approaching this idea in 2011-2012 when it clearly has been demonstrated by the market to be an unconditionally horrible idea and to go ahead with it anyway? Is this really a hard thing to figure out?
Even for "android on laptops" via chromebooks, you don't have them setting it up as the exact same interface all around. That and nobody fn cares about windows 8. They can use android dollars to continually revise the project but if the starting point is the metro UI then they've already got one foot in the grave.
"Did they have goofy oversized widgets for sloppy finger-based simple computer usage by retards? No. They were pen based. You know..... for functional useful software in a professional environment instead of a web browser on steroids for morons that can't type or write legibly anyway."
I love you.
*hugs Thinkpad tablet* (no, not the slow-ass Android version - a REAL Thinkpad tablet!).
maybe MIPS in the form of the Chinese-derivative Loongson?
That's already happening, and they're selling like hotcakes. http://www.mobilemag.com/2012/01/12/79-ainol-novo-7-paladin-tablet-does-ice-cream-sandwich/
The problem for Intel is the price of these SoCs:
http://rhombus-tech.net/allwinner_a10/
They may not be as capable as the Atom, but they're good enough to make very usable tablets at 1/10th the price.
"I've got more toys than Teruhisa Kitahara."
Yeah, it's a good chip. In performance-per-watt, it'll outdo any other Intel chip with ease. By x86 standards it sips power, even if you include the northbridge. But that is by x86 standards... by ARM, it just can't compete. If Intel really want to succeed in mobile, they'll need to take a big risk: Abandon the thirty-year heritage and backwards compatibility of x86/64.
It didn't deliver what 14-year-olds wanted. It delivered what business users and field techs wanted quite nicely.
The mantra for the future should be "It's not the hardware or the OS, it is the content stupid". If you can create a device that enables the end user to easily access content of their choosing then you are on a winner. The issue has always been ticking all the boxes, apple was the first to do so and android soon followed. So yes Intel is dead on the money, the hardware is a minor player but at the same time important as it is one of the many tick boxes that must be implemented correctly.
No I'm just tired of having to pay double for a machine useful as a tool instead of a locked-down funnel for paid entertainment content.
I tried to use several types of PDAs etc for a decade in business environment. It was always more trouble than it was worth. Until iphone came along in 2007.
I don't know about that. WinCE was way, way ahead of competitors for some time in many ways, right up until they abandoned Windows CE in preference for "Windows Phone". Things like:
* Mobile access to commonly emailed documents
* A good integrated mail client
* Thorough integrated contact management
* Excellent 3rd party mapping/GPS software
* Streaming and local video support
These are things Apple still lacks, and which Android devices are just now coming around to doing well. WinCE's biggest shortcomings were that:
* From a technical perspective, it had horrible support (for multitasking). Loading things was slow. This was probably due to its flash memory support and multiprocessing model, I suppose.
* Vendors fucked up the installs. Third party ROMs were usually much, much better - stable, fast - than what shipped on them. Think of the slowest, most hideously unstable Android device you've seen so far: that was a good day for a WinCE phone, and it was almost invariably the fault of the carrier.
* It was physically ugly compared to wiz-bang products like a Blackberry
* Microsoft didn't know what to do with it or how to control their IP (as with pretty much every 'mobile' venture they've made before or since).
~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
so, I should shell out between $500 and $830 to read in bed? It's going to need to replace more than a paper back book to separate me from a whole paycheck. I mean, my phone doesn't do that either but I think you need to come better than "reading in bed"
You can get kindle fire for about USD200. And yes, for book reading on your bed, it's overkill. So, aside for reading sci-fi novel in bed, I use my Archos 70IT to browse, read books, newspaper, comics, mangas, watching videos, ssh-ing to my NAS box at home, etc etc on my 4 hours daily commute
The above claim - of Windows 8 being able to run legacy Wintel software - will only be true about Windows 8 tablets based on x86, not ARM. Windows 8 on ARM will not run legacy Wintel software any more than Windows NT on Alpha or MIPS ever ran legacy Wintel software, which given the lack of native support, ended up being their undoing. Windows 8 has not been attracting ISVs the way Apple or Google have, and so that platform will depend mainly, if not solely, on support for legacy apps. If Medfield can trump ARM implementations like Apple A5/A6, Qualcomm, nVidia, Freescale, et al in terms of power consumption, Windows 8 may have a chance against Android.
In fact, even if Windows 8 can run legacy Wintel software, I'd think they'd at least need some sort of a rewrite to enable those applications to recognize touchscreen inputs, as opposed to keyboard/mouse inputs that they've been used to until now. There will also be the question of how to do it. Assume that a Windows 8 tablet has 256GB of flash memory, it could be possible to install it from a PC using a USB cable while treating the tablet as another hard drive. But the question of how the application will behave in a new environment is still open - unless the tablet can have USB keyboard & mouse inputs, in which case, why prefer it to a netbook?
MS Word or Office ain't gonna make or break Android - I believe LibreOffice can be installed on it. Moreover, versions of Office since Office 2007 have had those new file formats and ribbons, and are very different to use from previous versions. I worked w/ Excel 2003 very smoothly, doing pivot tables, vlookups and so on, but in subsequent versions, I'm totally lost, even though some of the previous keyboard shortcuts work. Besides, Apple, which depended on MS Office for its desktop offerings, came out w/ their own native tablet apps like Numbers (for spreadsheets), so it's not difficult for Google to do the same, even if they chose not to use LibreOffice as is.
The apps that matter - be it things like home budgeting, games, etc are there on iPad and I think Android as well. The only advantage a Windows 8 would have is if it could run native PC apps, which the other 2 can't. Let's see whether they can or not. But it won't happen if Microsoft insists on going w/ ARM, or Intel tries to run against ARM w/ Android.
...I think Android will end up losing the tablet war. The reason is that windows 8 will be able to leverage its existing base of "software capital", and bulldoze its way into the tablet market. Android simply does not have certain critical software (e.g. - MS Word) running on it.
What's to stop anyone from using all the existing FOSS? Is there something I missed about it being especially difficult to port (e.g.) OpenOffice from Linux to Android? And this is something of a strawman in any case. See below.
Think of it this way. The mass market desktop pc will die. For the vast majority of users, a simple tablet like device, with word processing capabilities, and media/internet capabilities, is all that's needed. Bulky laptops will disappear too, turning into tablets with Asus "transformer" like capabilities. Eventually, a multitude of device will be consolidated into one single tablet device - a single personal computer. People will want to do everything they did with their desktops, on their tablets. This will include word processing.
Hello there. I'm a writer. Writing is what I do. I do it 8-10 hours a day. I am going to do it with a separate, horizontal, *physical* keyboard and a vertical monitor. I am NOT going to do it using a virtual keyboard on the same 10" touchscreen I'm trying to read my work on. Ain't gonna happen.
Sure, you can modularise it--I can see having a tablet that one could drop into a dock/stand/whatever with a keyboard (and possibly a larger monitor and other peripherals)--but at the end of the day, see above: for any serious work involving manual entry of complex, nonrepeating alphanumeric data, you're still looking at a laptop/desktop configuration of some sort.
Il n'y a pas de Planet B.
Thanks for that clear answer. Sounds pretty bad to me although I think non-technical users might not care,
The Metro interface itself needs more work to show it can cut it as a launcher interface, too. In its current incarnation, it's scrolls left and right. You can move it with the scroll wheel on your mouse or drag a scroll bar that appears on the bottom, but it feels like a kludge way to navigate through all your apps. It's like someone up in Redmond suddenly realized "oh yeah, you have to have a touch-screen to swipe! That might be a problem for people who don't (like almost all desktop users)". Besides the "extra files" getting tossed into the interface in separate tiles I mentioned, there's the problem or navigating large collections of apps. On a tablet this isn't so much of an issue because of the more limited storage space on tablets, but on a desktop machines you could find yourself getting a bit weary scrolling through all those tiles to reach something at the other end. A full install of the Adobe CS4 Master Collection adds 25 new tiles to the Metro grid. There needs to be a way of dividing the "Full" Metro launcher view into sub-screens, like you can on iOS when you pull up specific "genres" of apps (or like you had programs and their support files segregated into folders named for the publisher on the old Start menu). Non-tech users will feel this as well once they have a healthy collection of free games from the Metro store or traditional apps from other places installed.
The Metro UI (and the included apps that come with it) are also obviously written under the assumption your screen is 13" or smaller. The Metro apps all run full-screen (and can't be changed to windowed) and their controls are all oversize for a desktop environment (I have a 1920x1200 display -- waste. of. space. ). I hate it when I hit certain links in the Desktop zone (usually in control panels) and for some reason instead of Firefox launching IE is coded to launch instead. And not the Desktop (normal) IE, but the Metro full screen version that whooshes everything else I'm working on out of view. I've also been unable to find a way to actually Exit any of these Metro apps, either. I can click out of them and back to the Launcher, but I cannot stop the process without actually End Tasking them from the Process Manager. They eventually go into a "hibernation"-like state instead if you don't use them. Also amusing: the Process Manager keeps track of network utilization and data usage on a per-app basis (obviously written with tablets and metered 3G data plans in mind).
but what interested me most was the bit about the 30% cut with the Metro App Store.
Correction: It's 20% for above $25,000 in sales. But it is exactly what it appears, Microsoft finding a way to take a cut from application sales revenue on programs they ha nothing to do with writing, just like Apple's store.
Not only PC vendors, but even most Android manufacturers ain't gonna prefer Intel to ARM, unless Intel can demonstrate lower power consumption AND greater performance @ the same time. And they'll have no reason to - all the apps already there for Android are Android on ARM. Plus you have a rich ecosystem of ARM manufacturers - Qualcomm, Freescale, TI, et al.
If you're not going to consider a Wintel tablet, there is really no reason to look @ Intel. The only thing Intel brings to the table is w/ Windows 8, where there is at least the theoretical possibility of running legacy software on it.
Problem for both Intel & Microsoft is that software for PCs are still pretty pricey, while software for tablets is really cheap, thanks to the repository stores. People who would mull over whether to spend $30 on a game would have no hesitation spending $1.99 or even $4.99 on it. Since the tablets have such inexpensive software, people can get a whole bunch of them, and Wintel can't have as many to offer there. So their trump card would be to offer Windows 8 tablets based on Medfield, and hope that it sticks. That's the only thing I can imagine bailing out Windows 8 from a fiasco in the tablet marketplace.
Tablets ain't gonna replace office laptops or servers, so there, both Intel & Microsoft are safe. But as far as home usage goes, tablets - particularly once they go head to head w/ PCs in price - will be seen as more and more attractive. As it is, the elimination of VGA and DVI from monitors is going to make a lot of monitors outdated, even though they are functioning just fine, while there will be so many affordable software titles available that at least on the home front, it has a good chance of heavily eroding the home PC business.
WinCE delivered professional grade suckiness - frequent crashes with data loss, and no upgrades to fix bugs at all. It looked like a scam to me. Professional non-tech people who used it against the advice of IT people (because it was "windows") were horrified at how crap it was, and could not wait to ditch it.
Sent from my ASR33 using ASCII
How weard is it for somebody to claim that the hardware it not important, the important thing is the software, and go on talking on how they'll create a product with the same software everybody else uses.
Yeah, the hardware is not important... I'll belive it when you stop using Android.
Rethinking email
You have a uid in the low 700k and you should know better. Why bring the discussion down to such a low level? Obviously many people find capacitive screen tablets useful and they are used by many serious businesses in a functional capacity. I wrote the catalog app framework my company uses for salespeople in the field. Our sales have probably doubled since putting it into production last year. We have a fleet of Acer A500 Android tablets btw. What should we do? Stop using the tablets and not worry about making money because of some anti-capacitive screen fan boy crusade? That is flat ludicrous. Not everybody is like you and is willing to torture themselves using a mouse centric os on a touch screen. The fact that you think so just makes you look like an ignorant pompous child. Seriously, the kind of low brow flaming you're laying down here is what I would expect from a 15 year old.
The soylentnews experiment has been a dismal failure.
If Office were that critical to touch screen devices, windows phone wouldn't be languishing at sub 2 percent market share. Furthermore, there are windows tablets on the market right now and there have been windows tablets on the market every day for the last decade available to general consumers. They all run office. The "real" office. Yet they collect dust while I pads fly off the shelves. Office outside of work, e.g., the desktop is not as important as you think it is.
The soylentnews experiment has been a dismal failure.
Maybe it's strictly a geek thing, but I'm not using any computer with a processor named "Allwinner".
Sorry, it ain't happening.
Faster! Faster! Faster would be better!
Let's talk about the other two typical problems with WinCE devices.
Until 2003, portable Windows CE devices nearly always had volatile storage. It was nearly part of the spec.
Windows CE's interface made no concession to small devices whatsoever, meaning they had to be operated with a stylus. Who cares about ugly, let's talk about usability.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
What fucked WinCE was MSFT trying to make it act like Windows, full stop. Look at WinCE 5.0 and what instantly comes to mind...its a fricking hybrid of Win2K and WinXP, which is precisely the WRONG message you want to give on a device that don't run x86. A local retailer over the holiday thought he was gonna make some extra scratch selling "Windows tablets" with WinCE, hell the way the box was laid out it looked enough like XP if I didn't know what WinCE was i wouldn't have known. The customers didn't know either as they didn't know WTF WinCE was and i'm sure looked at the picture and thought it was just another version of XP. Well the guy ended up having to cover the Windows part and sell them at a loss as generic tablets thanks to all the returns he got.
What Apple and Google got right that MSFT STILL hasn't got right is you can't run the desktop on something you poke with a finger, it just don't work. You'd have thought WinCE would have goten that through their heads but instead we're getting the desktop fucked so Ballmer can try to hoodwink enough developers to get some apps for WinTab. if any developers fall for this let me be the first to give you a Nelson HA HA because that shit is gonna cause confusion just like the WinCE WinTab and its gonna bomb. you need to keep mobile and desktop separate, not jam all that shit together.
ACs don't waste your time replying, your posts are never seen by me.
You are implying that businessmen and "professionals" are all adept at electronics. And conversely that if you are not an expert in IT, then you must be living in a trailer park, collecting unemployment.
And yes, a tablet has some severe restrictions. But the things it can do, it does with a lot more ease.
What?
Gee, so Compaq made iPaqs for a decade and never sold one. And HP never sold any Jornadas either. Never mind Linux also ran on the same hardware for years using either the GPE or OPIE environments or even sometimes plain X11.
Complex apps are near impossible to write for touch-based devices with small screens and the Android API and standard widget libraries suck. I would MUCH rather use OPIE on small screens. They could have done touch UI's back then but people who work for a living (the target market for $500-$800 expensive toys like these in the late 90's-early 2000's) wanted pointing precision and the ability to write/draw directly on the screen. Fingers are useless for precision pointing or complex usage of a system. For iFart apps they work great.
Palm Treos also sold well but I'll freely admit PalmOS is even more of a joke than WinCE. I never said WinCE was great but it certainly wasn't LATE. I was more of a NewtonOS fan personally.
My original point was MS was NOT late to the game. They were an early player alongside Apple and Palm.
Not saying mobile linux sucks but in reality, Android is just a mutant slow JVM running on top of Linux.
Most people want a lame portable web browser with text messaging and an MP3 player. Android does that ok I guess. Me, I want a highly portable computer with a cell phone bolted on. Different target market. For a portable COMPUTER, touch is not necessarily the right way to go. Win8 is going to be a joke because of this as well. For a portable content-consuming ENTERTAINMENT DEVICE (aka a toy), touch works quite well.
Recent mobile OS's are also a giant step backward when it comes to user freedom even though Android is "open".
I think the OP is confused about what should be called 'retarded'. There is no virtue in having a bad interface. WinCE had a bad interface. The stylus wasn't a better tool in most cases. It was what we used because the interface failed. WinCE also failed at offering "business" use. Android, and even iOS have far more and better business tools than WinCE did.
Another big failing with WinCE was that it's compatibility with itself was horrendous. If you bought an application that said it was WinCE compatible, there was a very good chance you couldn't run it on your device. Apple solved this by having stricter APIs and a very limited set of hardware. Android solved this by visualizing the processor. When WinCE was released, the Apple path of limited hardware was really the only path MS could have taken for compatibility, as the hardware wasn't up to snuff yet for emulation. I suspect the didn't do that because it was in direct opposition to how they made their fortune.