Just think of the halo effect of having the mammoth App Store incorporating the Mac! It'll be so cool when you can buy a universal app that works on your iPhone, iPad and iMac.
Eventually they'll finalize HTML5. Web-apps will be the way forward IMHO.
Not that old canard.
Web apps have been the way forward for seven years or longer. There's a place for them on the Mac, but every single app I've used on the iPhone or iPad that makes use of HTML5 in the interests of cross-platform compatibility (ie to save money) has been a dog.
Sometimes big companies just don't get it! For Intel read IBM/Microsoft etc.
Of those you mention, I am actually finding myself a lot warmer and rooting for IBM. I have long got over IBMs outright attack on Apple in the early 1980's with the removal of VT emulation support from main frames and minis. IBM was punished ironically by Microsoft when they stabbed IBM in the back with the clones. I am very hopeful the iBM-Apple partnership could be huge going forward and IBM have no reason to stab Apple in the back. In fact I even have hope as well as iOS that partnership may also benefit OS X in the future.
All those Apple/IBM iOS apps are being written/maintained/supported on Macs running OS X ... I suspect that there will be a comparable flesh out of Macs in IT departments who will write their own apps to interface the Apple/IBM offerings.
And, I just gotta' keep emphasizing this -- the lingua franca is Swift which currently runs on iOS and OS X ...
But, Swift is robust and scalable -- it can easily easily be used to serve up web pages, or feed database/Big Data to mobile/enterprise ...
It's true that sales of the iPad have declined in the past year, though I think they will return to year-on-year growth.
However, they have declined from an incredible sales rate. Remember the initial estimates for iPad sales? I recall that every analyst had them marked for the low single-digit millions. They were all made to look foolish by the extraordinary growth of the iPad. It grew faster than the iPhone or iPod. As such, the relatively modest decline in the past year is from a huge base.
If one compares iPad to Mac in sales, there is no contest. It seems that the refresh rate for iPad will probably end up being somewhere between the iPhone and the Mac. I used to buy an iPhone every two years, but have changed to every three. That is the same as the iPad. I will probably hold onto my Air 2 for even longer, as it's likely to keep speedy for a long time. A big iPad would entice me to upgrade, however.
Thanks for another great DEDitorial. It's going to be very interesting to see how Intel's fortunes play out. My feeling is that they deserve to be punished by Apple for being so slow with their chips and for not working more closely with Apple to produce optimisation.
I think that it won't be too long before we see ARM Macs running some form of OS X. The ultimate endgame will be to have a universal App Store that combines iOS and OS X in one place, with Swift leading the charge.
Seems like i see this over and over. Unless Apple knocks it out of the park every quarter and every year, that they're slipping! Being hugely profitable isn't enough, you have to be insanely profitable!
That explanation may be correct, but does it fundamentally change anything except, perhaps in the most generous sense, to say that it should not even be included in the comparison. If they never intended it to be a commercial success and it still failed to compete significantly with the iPad, what should we conclude?
That Nexus devices fulfill their stated purpose pretty well.
That is certainly the message that Google would like to project, but I'm not sure that it holds up to scrutiny. So we have a pure Android, testbed device intended to showcase the platform and its capabilities. Presumably, to that extent, they fulfilled their purpose. But, sales were not production limited, and the resulting sales were much lower than the iPad. So - again - what should we conclude?
That is certainly the message that Google would like to project, but I'm not sure that it holds up to scrutiny. So we have a pure Android, testbed device intended to showcase the platform and its capabilities. Presumably, to that extent, they fulfilled their purpose. But, sales were not production limited, and the resulting sales were much lower than the iPad. So - again - what should we conclude?
I suppose anyone can imagine a different scenario but unless you have some realistic evidence that Google isn't really using the Nexus program for the purposes stated (and stated as such for several years now) it's a moot conversation. We're not talking "what-if's" are we?
It's only a matter of time before Intel is on the scrap heap.
I was under the impression that without Intel x86 architecture there is no Thunderbolt. It appears that Apple has pinned the future of Mac and OS X on Thunderbolt. For that reason alone it doesn't seem very likely that they would bail on Intel, at least not anytime soon.
That is certainly the message that Google would like to project, but I'm not sure that it holds up to scrutiny. So we have a pure Android, testbed device intended to showcase the platform and its capabilities. Presumably, to that extent, they fulfilled their purpose. But, sales were not production limited, and the resulting sales were much lower than the iPad. So - again - what should we conclude?
I suppose anyone can imagine a different scenario but unless you have some realistic evidence that Google isn't really using the Nexus program for the purposes stated (and stated as such for several years now) it's a moot conversation. We're not talking "what-if's" are we?
No, not "what ifs". Simply - if the Nexus tablets are good, why do they not compete, in terms of sales, with the iPad, irrespective of their "stated" purpose?
It's true that sales of the iPad have declined in the past year, though I think they will return to year-on-year growth.
However, they have declined from an incredible sales rate. Remember the initial estimates for iPad sales? I recall that every analyst had them marked for the low single-digit millions. They were all made to look foolish by the extraordinary growth of the iPad. It grew faster than the iPhone or iPod. As such, the relatively modest decline in the past year is from a huge base.
If one compares iPad to Mac in sales, there is no contest. It seems that the refresh rate for iPad will probably end up being somewhere between the iPhone and the Mac. I used to buy an iPhone every two years, but have changed to every three. That is the same as the iPad. I will probably hold onto my Air 2 for even longer, as it's likely to keep speedy for a long time. A big iPad would entice me to upgrade, however.
Thanks for another great DEDitorial. It's going to be very interesting to see how Intel's fortunes play out. My feeling is that they deserve to be punished by Apple for being so slow with their chips and for not working more closely with Apple to produce optimisation.
I think that it won't be too long before we see ARM Macs running some form of OS X. The ultimate endgame will be to have a universal App Store that combines iOS and OS X in one place, with Swift leading the charge.
" think that it won't be too long before we see ARM Macs running some form of OS X. The ultimate endgame will be to have a universal App Store that combines iOS and OS X in one place, with Swift leading the charge."
I agree. With Windows 10 having a Universal Apps for phone, tablet, PC, it can't come soon enough. A universal app store would be great for consumers.
Just think of the halo effect of having the mammoth App Store incorporating the Mac! It'll be so cool when you can buy a universal app that works on your iPhone, iPad and iMac.
It's coming sooner rather than later:
Lister: A Productivity App (Obj-C and Swift)
Last Revision: Version 1.4, 2014-10-16Update to reflect latest changes in the Swift compiler for Xcode 6.1.
Build Requirements: Xcode 6.1 or later; iOS 8 SDK or later; OS X 10.10 SDK or later.
Runtime Requirements: iOS 8 or later; OS X 10.10 or later.
Lister is a list management app written in Swift that's built on top of iCloud and the powerful NSDocument and UIDocument architectures. In a single project you'll find both iOS and OS X targets, embedded frameworks, UI extensions, live rendering of custom views in Interface Builder, iOS and OS X Storyboards, Auto Layout, and more. To abstract the storage mechanism away from the type of storage (iCloud or local), Lister uses a {AAPL}ListController class that notifies the {AAPL}ListDocumentsViewController about new lists, lists that have been removed, and also lists that have been updated. The {AAPL}ListController has an {AAPL}ListCoordinator property which is resonsible for tracking the relevant URLs. In Lister, there are two types of {AAPL}ListCoordinator objects: the {AAPL}CloudListCoordinator object as well as the {AAPL}LocalListCoordinator object. The only place that these objects are used directly is within the {AAPL}ListController. The storage mechanism is determined by the {AAPL}AppDelegate, which asks the user what their storage preference is. Once their preference is known, the app delegate creates an {AAPL}ListCoordinator and passes it to the app delegate's {AAPL}ListController property. The app delegate passes the {AAPL}ListController object throughout the application to ensure that it's used as the single place to manage lists.
No, not "what ifs". Simply - if the Nexus tablets are good, why do they not compete, in terms of sales, with the iPad, irrespective of their "stated" purpose?
Limited distribution, largely thru Google themselves? No advertising or promotion? Additional and useful consumer features reserved for the licensee version of that same general Nexus model, ie LG G3 is the enhanced Nexus 5? That's three I can come up with off the top of my head. Again it's not about whether a Nexus model is better both feature and hardware-wise than the followup licensee models that build on/improve the reference design and OS version.
The Nexus program has a specific stated purpose and competing with licensees, a requisite and harmful side-effect if they were intended as commercial successes, is not it. I'm very surprised it's not as obvious to you.
And I'm so glad that that one Apple engineer (don't recall who) convinced Jobs that Atom would be a terrible idea for the iPad. The iPad would be so far behind today, it wouldn't even be 64-bit.
That would be Tony Fadell. He gathered support from other Engineers and confronted Jobs.
From Steve Jobs book by Walter:
"Wrong, wrong, wrong!" Fadell shouted at one meeting when Jobs insisted it was best to trust Intel....Fadell even put his Apple badge on the table, threatening to resign.
Eventually Jobs relented. "I hear you," he said."
Not many people got the guts to confront Jobs this way, glad he was there then.
Thanks Tony.
Concerning Windows' compatibility, why do people need Window's compatibility? Mainly for Word and Excel?
Microsoft is smarter than Intel. At least now Microsoft is hedging its bet with Office for ipad and being very aggressive with it. (Free edit anyone?)
The next question is when larger screen ipads paired with keyboards are common and Office on iPad works well enough for many users, then will Windows' compatibility be of that much importance?
Just think of the halo effect of having the mammoth App Store incorporating the Mac! It'll be so cool when you can buy a universal app that works on your iPhone, iPad and iMac.
Eventually they'll finalize HTML5. Web-apps will be the way forward IMHO.
Not that old canard.
Web apps have been the way forward for seven years or longer. There's a place for them on the Mac, but every single app I've used on the iPhone or iPad that makes use of HTML5 in the interests of cross-platform compatibility (ie to save money) has been a dog.
Here's an interesting option:
A savvy web site could run server-side Swift (compiled or interpreted) to generate dynamic web pages instead of PHP, Java, JavaScript, Perl, Python, ColdFusion ...
When these web pages are sent to browsers on Macs and iDevices -- the browsers could use client-side Swift (compiled or interpreted) to present the pages instead of JavaScript.
On other platforms, the client-side would present pages as they do now ...
Apple could really piss off the competition by offering superior web functionality and performance -- only on Apple hardware running Apple browsers (and Apple apps to access web content).
Limited distribution, largely thru Google themselves? No advertising or promotion? Additional and useful consumer features reserved for the licensee version of that same general Nexus model, ie LG G3 is the enhanced Nexus 5? That's three I can come up with off the top of my head. Again it's not about whether a Nexus model is better both feature and hardware-wise than the licensee models taht build on the reference design and OS version.
The Nexus program has a specific purpose and competing with licensees, a requisite and harmful side-effect if they were intended as commercial successes, is not it. I'm very surprised it's not as obvious to you.
Obfuscation (or beclouding) is the hiding of intended meaning in communication, making communication confusing, willfully ambiguous, and harder to interpret. [ citation needed] The word comes from Latin obfuscatio, from obfusc?re ("to darken"), from ob ("over") and fusc?re ("to make dark"), from fuscus ("dark").
Why worry about competing with OEM's, says Google when they bought Motorola. They surely know that we just want to keep Motorola's IP from Apple and MS.
We love the OEM's, says Google after Motorola is sold to Lenovo. We only bought Motorola for the IP, but sadly, it wasn't worth anything.
We love the OEM's, excepting Samsung who is getting too uppity, but we need them so let's cross license our IP with them, and how about that KNOX!
Why don't we tell the truth.
Google's job one is to maintain a path for advertising on mobile devices, as an insurance policy against Apple, MS, or anyone else cutting them out. The Nexus is merely a demonstration of "pure" Android, and frankly, functional hardware is sufficient for that.
Google is quite aware that there's no money in Android hardware.
Google's job one is to maintain a path for advertising on mobile devices, as an insurance policy against Apple, MS, or anyone else cutting them out.
</ol>
And you could have stopped with that alone, the one gem in the gravel pit.
the fact that they don't publish their sales numbers is all you need to know.
Plenty of successful companies don't post sales figures. Do you think that Rolex's watches have been a commercial flop because they don't post sales figures?
What was Intel thinking? Did they really think there would be any substantial payoff down the road? Maybe they all had to get together and make cheap tablets just to remain relevant, or become completely irrelevant if they did nothing.
Apple had that back in the pre Power-PC days, and it really went nowhere. The card was almost as expensive as a bare-bones PC of the day, primarily because it had to duplicate most of the PC on that card. While that is a lot cheaper today, it is still a significant extra expense for a few select users.
If they can get the ARM chip to process the x86 and/or x86_64 instruction set, that will be a much better option. While it will be an extra expense, it will be spread out across a much larger set of users, and there will not be a need to duplicate the rest of the PC architecture.
There are those who think that Apple should just switch to ARM. If WindowsRT had been a success, I would say maybe, but there are too many people who list Windows compatibility as a must have feature (wether they really need it or not). While they should have an ARM version in the labs (similar to how they had the x86 version in the labs for years before it was released), as long as Windows is still a significant force in the PC Operating system market, ditching that compatibility will be a major step backwards.
I repeat my point, I think we are fast approaching the times when a Mac doesn't need compatibility for the vast majority of Mac users even if I do. If Apple were to come out with a powerful CPU/ GPU set up that was totally optimized for a version of OS X I think it would be a huge step forward, not back. The performance gains could be significant. It would also be a bonus to me that the Wintel world would cease being able to piggy-back Apple's improvements and designs not to mention cash input, helping their flakey survival.
I know all about add on PC boards, I sold them from Apple ][e through to PPC. These days a TB based dongle could be relatively inexpensive and with VMWare be fine for the person who need Windows occasionally. In many cases a Windows VM is far superior to an actual PC in terms of it being easily replicated and backed up with plug and play functionality absent in many wintel boxes. I run every Windows OS on my Mac Pro going back to XP as well as many versions of OS X. But I'd agree that an Intel PC on a board is additional cost so if one really needed a dedicated PC that badly I'd say a cheap PC as well as a Mac might cost less ... but I'd rather have a VM.
Intel will close the gap eventually. Chip design is not magic and they are not stupid. And while they're playing catch up, they have to keep an x86 ecosystem alive in that space, hence the subsidies.
closing the gap is a race term...
"Intel can close the gap in the race, even with the x86 boat anchor firmly attached to one ankle, and it's relay partners using a generic OS and "race to the bottom" chipsets running on top of their HW.... competing against teams that can cut a runner and put in a new one every year, because they control the selection of the team, heck they control the breeding of the team.
I don't think so.
My guess is Intel doesn't feel it's playing catchup, it's that PC/tablet makers are failing to differentiate "intel inside" I think they are confused as to why the 'integrators' can't sell Intel.
Intel's problem is that they haven't done an Apple and forked their product line, one with no 8/16/32 bit x86 emulation, one optimized for mobile activity, not one for generic computing. It's like saying you want a 3 cylinder LPG engine that gets 120miles per gallon, yet, make sure if the driver puts diesel into it, it can switch ignition plans and 'just' run, oh, and it can do it every gear shift (context switch).
And, if Atom is the engine and Android the transmission, and the transmission is designed to run generically on any engine, you've now added yet another layer of cruft. That takes time and effort (money and more money) to tune and remove before pushing it out the door.
Finally, you've built a power train (HW and OS) and then you integrate it into parts cobbled together for brakes, dashboard, wheels, headlights, gas tank, because designing uniquely for the product is time and effort (see above), you will get a poor handling vehicle.
In all, Intel is suffering in that whereas Windows was optimized for x86, Android is not, and the mobile market doesn't tolerate failures in user experience.
And in the end, there will be 2 types of computers in the world: DataCenter/Cloud based (intel can make money there), and mobile (Intel failing misrably). The desktop (where a 200W power supply and lots of slots can paper over any integrations issues), won't be a cash cow to Intel. Because in 5 years, the desktop will be a mobile device docking station (a PoE GbEthernet to Lightning connector).
Microsoft got it half right... You need to control the integration and construction of the end device, the user experience, to sell in mobile.
Where MS got it wrong... they haven't learned how to say 'no' (No mouse, no stylus, no x86, no ARM), even in their own product.
Apple, now going on 8 years ago, made the decision that they would control the entire vehicle. Built a new OS based on the best parts of their OS, and removed everything that didn't have to do with driving a motorcycle. Used a generic ARM chip at first, but had plans to marry a new ARM design that was optimized to the chassis and transmission and from what they learned how people 'rode' their machine.
In 3 iterations (iPhone 2G iPhone 3G iPhone 3Gs - what 3 years), the iPhone 4 came out and set this path down in stone. The HW would be optimized for the perceived use of this device, not for backwards compatibility of infinite generations.
At this point, the chipset is not the end-user purchasing decision point. The OS barely is. It's the interaction experience, the apps and the ecosystem, and Intel is so far removed from that, they must learn to refactor and build a better mobile chip and get used to competing with Qualcomm and Nvidia at mobile margins.
Or get out of the market and focus on the challenges in the Data Center, where the HW is even more virtualized away from the user.
Comments
Not that old canard.
Web apps have been the way forward for seven years or longer. There's a place for them on the Mac, but every single app I've used on the iPhone or iPad that makes use of HTML5 in the interests of cross-platform compatibility (ie to save money) has been a dog.
All those Apple/IBM iOS apps are being written/maintained/supported on Macs running OS X ... I suspect that there will be a comparable flesh out of Macs in IT departments who will write their own apps to interface the Apple/IBM offerings.
And, I just gotta' keep emphasizing this -- the lingua franca is Swift which currently runs on iOS and OS X ...
But, Swift is robust and scalable -- it can easily easily be used to serve up web pages, or feed database/Big Data to mobile/enterprise ...
It's true that sales of the iPad have declined in the past year, though I think they will return to year-on-year growth.
However, they have declined from an incredible sales rate. Remember the initial estimates for iPad sales? I recall that every analyst had them marked for the low single-digit millions. They were all made to look foolish by the extraordinary growth of the iPad. It grew faster than the iPhone or iPod. As such, the relatively modest decline in the past year is from a huge base.
If one compares iPad to Mac in sales, there is no contest. It seems that the refresh rate for iPad will probably end up being somewhere between the iPhone and the Mac. I used to buy an iPhone every two years, but have changed to every three. That is the same as the iPad. I will probably hold onto my Air 2 for even longer, as it's likely to keep speedy for a long time. A big iPad would entice me to upgrade, however.
Thanks for another great DEDitorial. It's going to be very interesting to see how Intel's fortunes play out. My feeling is that they deserve to be punished by Apple for being so slow with their chips and for not working more closely with Apple to produce optimisation.
I think that it won't be too long before we see ARM Macs running some form of OS X. The ultimate endgame will be to have a universal App Store that combines iOS and OS X in one place, with Swift leading the charge.
Seems like i see this over and over. Unless Apple knocks it out of the park every quarter and every year, that they're slipping! Being hugely profitable isn't enough, you have to be insanely profitable!
That explanation may be correct, but does it fundamentally change anything except, perhaps in the most generous sense, to say that it should not even be included in the comparison. If they never intended it to be a commercial success and it still failed to compete significantly with the iPad, what should we conclude?
That Nexus devices fulfill their stated purpose pretty well.
That is certainly the message that Google would like to project, but I'm not sure that it holds up to scrutiny. So we have a pure Android, testbed device intended to showcase the platform and its capabilities. Presumably, to that extent, they fulfilled their purpose. But, sales were not production limited, and the resulting sales were much lower than the iPad. So - again - what should we conclude?
It's only a matter of time before Intel is on the scrap heap.
I was under the impression that without Intel x86 architecture there is no Thunderbolt. It appears that Apple has pinned the future of Mac and OS X on Thunderbolt. For that reason alone it doesn't seem very likely that they would bail on Intel, at least not anytime soon.
No, not "what ifs". Simply - if the Nexus tablets are good, why do they not compete, in terms of sales, with the iPad, irrespective of their "stated" purpose?
It's coming sooner rather than later:
The Nexus program has a specific stated purpose and competing with licensees, a requisite and harmful side-effect if they were intended as commercial successes, is not it. I'm very surprised it's not as obvious to you.
And I'm so glad that that one Apple engineer (don't recall who) convinced Jobs that Atom would be a terrible idea for the iPad. The iPad would be so far behind today, it wouldn't even be 64-bit.
That would be Tony Fadell. He gathered support from other Engineers and confronted Jobs.
From Steve Jobs book by Walter:
"Wrong, wrong, wrong!" Fadell shouted at one meeting when Jobs insisted it was best to trust Intel....Fadell even put his Apple badge on the table, threatening to resign.
Eventually Jobs relented. "I hear you," he said."
Not many people got the guts to confront Jobs this way, glad he was there then.
Thanks Tony.
Concerning Windows' compatibility, why do people need Window's compatibility? Mainly for Word and Excel?
Microsoft is smarter than Intel. At least now Microsoft is hedging its bet with Office for ipad and being very aggressive with it. (Free edit anyone?)
The next question is when larger screen ipads paired with keyboards are common and Office on iPad works well enough for many users, then will Windows' compatibility be of that much importance?
Here's an interesting option:
A savvy web site could run server-side Swift (compiled or interpreted) to generate dynamic web pages instead of PHP, Java, JavaScript, Perl, Python, ColdFusion ...
When these web pages are sent to browsers on Macs and iDevices -- the browsers could use client-side Swift (compiled or interpreted) to present the pages instead of JavaScript.
On other platforms, the client-side would present pages as they do now ...
Apple could really piss off the competition by offering superior web functionality and performance -- only on Apple hardware running Apple browsers (and Apple apps to access web content).
Limited distribution, largely thru Google themselves? No advertising or promotion? Additional and useful consumer features reserved for the licensee version of that same general Nexus model, ie LG G3 is the enhanced Nexus 5? That's three I can come up with off the top of my head. Again it's not about whether a Nexus model is better both feature and hardware-wise than the licensee models taht build on the reference design and OS version.
The Nexus program has a specific purpose and competing with licensees, a requisite and harmful side-effect if they were intended as commercial successes, is not it. I'm very surprised it's not as obvious to you.
And you could have stopped with that alone, the one gem in the gravel pit.
the fact that they don't publish their sales numbers is all you need to know.
Plenty of successful companies don't post sales figures. Do you think that Rolex's watches have been a commercial flop because they don't post sales figures?
What was Intel thinking? Did they really think there would be any substantial payoff down the road? Maybe they all had to get together and make cheap tablets just to remain relevant, or become completely irrelevant if they did nothing.
Eventually Jobs relented. "I hear you," he said."
This is a testament to Steve's genius just as much. Most bosses would have shown Tony the door.
I repeat my point, I think we are fast approaching the times when a Mac doesn't need compatibility for the vast majority of Mac users even if I do. If Apple were to come out with a powerful CPU/ GPU set up that was totally optimized for a version of OS X I think it would be a huge step forward, not back. The performance gains could be significant. It would also be a bonus to me that the Wintel world would cease being able to piggy-back Apple's improvements and designs not to mention cash input, helping their flakey survival.
I know all about add on PC boards, I sold them from Apple ][e through to PPC. These days a TB based dongle could be relatively inexpensive and with VMWare be fine for the person who need Windows occasionally. In many cases a Windows VM is far superior to an actual PC in terms of it being easily replicated and backed up with plug and play functionality absent in many wintel boxes. I run every Windows OS on my Mac Pro going back to XP as well as many versions of OS X. But I'd agree that an Intel PC on a board is additional cost so if one really needed a dedicated PC that badly I'd say a cheap PC as well as a Mac might cost less ... but I'd rather have a VM.
Steve Jobs was the Chef Ramsey of his day
Intel will close the gap eventually. Chip design is not magic and they are not stupid. And while they're playing catch up, they have to keep an x86 ecosystem alive in that space, hence the subsidies.
closing the gap is a race term...
"Intel can close the gap in the race, even with the x86 boat anchor firmly attached to one ankle, and it's relay partners using a generic OS and "race to the bottom" chipsets running on top of their HW.... competing against teams that can cut a runner and put in a new one every year, because they control the selection of the team, heck they control the breeding of the team.
I don't think so.
My guess is Intel doesn't feel it's playing catchup, it's that PC/tablet makers are failing to differentiate "intel inside" I think they are confused as to why the 'integrators' can't sell Intel.
Intel's problem is that they haven't done an Apple and forked their product line, one with no 8/16/32 bit x86 emulation, one optimized for mobile activity, not one for generic computing. It's like saying you want a 3 cylinder LPG engine that gets 120miles per gallon, yet, make sure if the driver puts diesel into it, it can switch ignition plans and 'just' run, oh, and it can do it every gear shift (context switch).
And, if Atom is the engine and Android the transmission, and the transmission is designed to run generically on any engine, you've now added yet another layer of cruft. That takes time and effort (money and more money) to tune and remove before pushing it out the door.
Finally, you've built a power train (HW and OS) and then you integrate it into parts cobbled together for brakes, dashboard, wheels, headlights, gas tank, because designing uniquely for the product is time and effort (see above), you will get a poor handling vehicle.
In all, Intel is suffering in that whereas Windows was optimized for x86, Android is not, and the mobile market doesn't tolerate failures in user experience.
And in the end, there will be 2 types of computers in the world: DataCenter/Cloud based (intel can make money there), and mobile (Intel failing misrably). The desktop (where a 200W power supply and lots of slots can paper over any integrations issues), won't be a cash cow to Intel. Because in 5 years, the desktop will be a mobile device docking station (a PoE GbEthernet to Lightning connector).
Microsoft got it half right... You need to control the integration and construction of the end device, the user experience, to sell in mobile.
Where MS got it wrong... they haven't learned how to say 'no' (No mouse, no stylus, no x86, no ARM), even in their own product.
Apple, now going on 8 years ago, made the decision that they would control the entire vehicle. Built a new OS based on the best parts of their OS, and removed everything that didn't have to do with driving a motorcycle. Used a generic ARM chip at first, but had plans to marry a new ARM design that was optimized to the chassis and transmission and from what they learned how people 'rode' their machine.
In 3 iterations (iPhone 2G iPhone 3G iPhone 3Gs - what 3 years), the iPhone 4 came out and set this path down in stone. The HW would be optimized for the perceived use of this device, not for backwards compatibility of infinite generations.
At this point, the chipset is not the end-user purchasing decision point. The OS barely is. It's the interaction experience, the apps and the ecosystem, and Intel is so far removed from that, they must learn to refactor and build a better mobile chip and get used to competing with Qualcomm and Nvidia at mobile margins.
Or get out of the market and focus on the challenges in the Data Center, where the HW is even more virtualized away from the user.
This is a testament to Steve's genius just as much. Most bosses would have shown Tony the door.
Steve Jobs was the Chef Ramsey of his day
Well said!