Hacker News Reader

Comment by enko | Hacker News Reader

Dear Microsoft: Out of ARM and x86, choose one. Just choose one. For the love of god, choose ONE.

Can you imagine Apple, or anyone else, deliberately releasing two different tablets, with the same name, which can not run each others' software? It is inconceivable, and for good reason.

MS cannot make decisions. That's what will kill them, long term. Just choose one!!

View on HN

You've misread the problem.

Apple DOES have an X86 ecosystem AND an ARM ecosystem. They do not cross streams and customers are not confused, and Apple can successfully grow application ecosystems across two very different architectures.

So what is Microsoft doing wrong?

It's very simple, and I cannot believe they haven't already done this. I'm thinking that the naming scheme for their ARM platform is why they're balking, since they've stupidly named their ARM ecosystem "Windows Phone".

Apple succeeds by maintaining separation between the ecosystems, and that's where MS is failing.

MS needs to realign:

Surface means x86 tablet. Period. End of story. It's a high end tablet for those who need more than an iDevice, who want a full laptop in a tablet shell. Surface means quality. Surface means x86, and it means desktop compatible.

Windows Phone, including Windows Phone Tablet (Surface RT) means ARM. Period. The WP8 ecosystem and it's tablet arm should be the branding covering the ARM ecosystem.

I'm thinking that the name "Windows Phone" for the ecosystem is what dissuaded them from trying to attach the tablet to that ecosystem.

You're actually agreeing with the parent, who stated the problem as 'two different tablets, with the same name, which can not run each others' software'.

Yes, but the parent also said "MS cannot make decisions. That's what will kill them, long term. Just choose one!!"

The reply just said that two tablets as fine if they would just kill the confusing naming scheme.

I would rather them invest more into ARM, and use the Surface name (a great name) for the direct iPad competitor.

The Surface Pro is just a Windows laptop with decent build quality. It's the one in a boring (and shrinking!) market.

> The Surface Pro is just a Windows laptop with decent build quality

Well its a "Windows laptop", except for the fact that its a tablet that can (like most tablets) also function as a laptop and (unlike most tablets) comes bundled with the hardware for that rather than it being a separate purchase.

But its a full Windows tablet, not a WinRT tablet.

I guess I should have said "Windows tablet/laptop" then, or better yet, "Windows 8 tablet." It's very similar to other Windows 8 tablets, like the ASUS VivoTab or the Dell Latitude 10.

http://incredibletogether.asus.com/vivotab.html

http://www.dell.com/us/business/p/latitude-10-tablet/pd

Why does Apple get a pass on this? They have just taken the alternative approach of fracturing their OS and hardware. It's natural because of where they started (iPod -> iPhone -> iPad), so this isn't a criticism.

As a consequence, this is one area where MS is ahead of Apple (even if MS is having difficulties reconciling the tablet vs. desktop experience). MS has touch on the desktop. When is Apple going to have touch on Mac Books?

I know SJ said vertical touch screens don't work. They will continue to say so until they add touch. Undoubtedly, they'll do it well, and it will be well integrated with the OS. But they'll continue to say it doesn't work...until it does. Similar to the movement to flat with nary a mention of Metro.

So Apple has the challenge of bringing their user-level of their OS offerings closer together (along with the hardware), whereas MS has the challenge of dealing with the Metro/Desktop schism and the x86/ARM dichotomy. Same coin, different sides.

>Why does Apple get a pass on this?

Here's why: Last November when the iPad Mini and the iPad 4 launched, Apple sold 3 million iPads over the weekend. Surface has sold about 1.7 million devices total since it launched. http://bgr.com/2013/07/31/microsoft-surface-sales-2013/

It's hard to argue with success.

>As a consequence, this is one area where MS is ahead of Apple

The numbers seem to disagree with you. The reason Apple doesn't put touch on their Macbooks is cause no one wants it.

That's a false argument. These are separate points of discussion. You're arguing that they're right because they were first to market in a new space and have achieved significant momentum. Momentum is just that. They've been riding on it for a while. The last significant innovation was iPad, which was just an evolution of the one real innovation: iPhone. That's only going to carry them so far. Android is eating their market share while Apple is extracting what they can from this momentum and iterating on a theme.

I think Apple's products are great. Their ID is fantastic. Their developer ecosystem leaves much to be desired. The fact that I no longer buy their products due to their closed nature (and, by extension, their philosophy for the future: closed, controlled, and owned) does not take away from my admiration of the company from a product and engineering perspective.

Actually. Apple was really late to the market with the iPad. Microsoft had been doing tablets for 8 years before Apple came along (https://en.wikipedia.org/wiki/Microsoft_Tablet_PC).

Apple was the first to bring down the price, make it holdable (in your hand), give it a battery that you don't need to worry about. Give it apps that were only designed for touch.

Android is iterating as much as Apple is. They will probably keep iterating until some big hardware breakthrough, maybe when someone releases a ARM86 (compatible ARM and x86 processor)? One major breakthrough of the original iphone was smooth finger tracking made available by the new capitative touch screen.

I'll just leave this little nugget here. https://en.wikipedia.org/wiki/Apple_Newton

Before, they were called PDAs and they've existed since the late 80s. Apple wasn't late, they were simply waiting for the technology arrive to do it right.

"The reason apple doesn't put touch on their Macbooks is cause no one wants it"...

The reason they have not added touch input to OS X, is because its not design for touch. I am sure, Apple will find another input method for their non-mobile hardware. Perhaps something similar to MYO or Leap Motion (or a touchscreen).

There's a better reason. Touch doesn't work for a vertical screen at all, it makes your arms hurt.

Agree. I don't want touch on a laptop. I see no need to mar the screen of my computer with debris and fingerprints.

  >When is Apple going to have touch on Mac Books?

I find using a Magic Trackpad with a desktop Mac to be a wonderful experience. After working in a Linux/Windows world for a while, I was surprised at how quickly I stopped relying on hotkeys in favor of gestures. I'd say any Apple computer running 10.7 or later is actually built for 'touch'.

I always thought features like 'Launchpad' were useless for those able to type, but they feel more intuitive once you've practiced the gestures. Apple trackpads allow me to directly manipulate content while retaining the cursor accuracy needed for complex interfaces. I'll be surprised if Apple adds capacitive touch to their laptop screens.

If you have a chance you should try out a touchscreen laptop some time. It is a much more compelling experience then gestures on a trackpad.

> So Apple has the challenge of bringing their user-level of their OS offerings closer together (along with the hardware)

This is the assumption I don't agree with. Why do we need one OS to run all our devices? Ubuntu Unity is an interesting idea, but is that really where computing should head.

I don't see this as Apple "getting a pass", I think they mad e the right decision. It was certainly correct in '07 (for hardware reasons), and I think it still holds up as correct today.

Given the sales of Surface and touchscreen PCs I have to reject the whole premise of your question. What is Apple getting a 'pass' on here? They seem to be giving consumers what they want/expect out of these devices. Apple could snap their fingers and have a touch enabled version of OSX that's better than Windows 8. They could easily add an x86 target for iOS apps and have a massive library of native touch enabled apps for the Mac. They could do a legacy touch mode as good (low standard) as what Windows 8 has to offer. They obviously haven't done it because they think it sucks. Again given that the market has said the same thing so far the premise of the question is really backwards to me. Apple is getting a pass on not picking a failing strategy!

> Can you imagine Apple, or anyone else, deliberately releasing two different tablets, with the same name, which can not run each others' software?

Apple actually managed two very smooth transition/coexistence periods, from 68K to PPC and from PPC to x86, when it was easy to build binaries that ran on both platforms.

It's very easy to build binaries for Windows that run on both platforms if you're using their approved tools.

It's all the legacy apps that are the problem.

Apple actually just made each executable have the binaries for both platforms. It's a good trick.

Regardless of engineering cleverness it's a very bad decision marketing wise.

Apple didn't release both simultaneously.

Even if you do that, you still can't run your desktop apps on the ARM model. I really don't want to explain to luddite relatives why they can't run a game unless they buy the more expensive model.

Either ban desktop apps on both models, or neither.

Rosetta worked pretty well.

I agree with you that this is technically a solved problem. It's not difficult to develop a system that lets apps run on both architectures without users having to think about this.

I'm talking about the management decision that only Metro apps can run on RT, which is such a stupid limitation that even Microsoft sees the need to bypass it to get Office on an RT tablet.

Emulating an x86 on an ARM like the Tegra 4 would probably result in a pretty unsatisfying experience. Apple was doing the opposite: Emulating a processor with worse single threaded performance and (arguably) a simpler ISA.

The alternative is a full port of Win32, which I imagine Microsoft considered and rejected. Even after they incurred all the time and development cost, ISVs would still have to at least recompile.

I think it has a nearly-full port of Win32, it's apparently not hard to disable the signing requirement and run cross-compiled Win32 ARM binaries and turn the Surface RT into a useful computer.

Not much commercial software but the regular open-source windows stack works.

it's apparently not hard to disable the signing requirement and run cross-compiled Win32 ARM binaries

Link please?

This forum http://forum.xda-developers.com/forumdisplay.php?f=2130 appears to be ground zero for that sort of thing. I don't have an RT handy to test it on.

I don't think it's a feasibility issue, because a fair bit of Win32 had already been ported to ARM from the Windows CE days. I had .NET apps that ran on Windows XP and Windows CE from the same binary. As long as you stuck to the .NET micro framework subset, it worked great and was actually quite performant.

While I agree with your overall sentiment, in fairness it's not hard to explain.

"Well the cheap one is shite, it doesn't really work. The expensive one does work so you want to buy that if you're going to buy one at all."

> which can not run each others' software?

Windows Store apps run on both of them.

I don't believe this is true. Most windows store apps run on both of them, but I think there are some, e.g. graphics, API's that your app could access that would limit it to x86 only

Not that I know of. It's just that if you're going native code instead of JS or .NET you have to make sure yourself that you run on ARM. And apparently C++ is the only option that gets access to DirectX which may mean that 3D apps never get ported to ARM.

Or better yet, they could stay the course and realize that architectures are going to have to be abstracted over in the decades to come.

That may be true but right now the user shouldn't have to care.

While I agree, and wish they just released a budget x86 version instead of the RT, you are exaggerating the problem somewhat.

Not really. Since both run different apps then they should be distinctly separate products, and tacking "RT" or "Pro" or any other moniker does a disservice to everybody involved -- both developers and consumers.

You can easily get an x86-based tablet and run a more or less standard Linux environment on top of the Android kernel. If your performance requirements fit, you can do it on an ARM processor. Chances are you can add a bluetooth keyboard/cover for much less than a Surface 2.

> which can not run each others' software?

Actually Windows 8 can run anything Windows RT can run. The opposite isn't true however, since there's no way to write a desktop (x86/Win32) application for Windows RT.

Relax.

They have clearly made some sort of a bet on ARM, if it doesn't pan out they will drop it soon enough.

In the meantime, it's terribly confusing to consumers and could well result in a lot of ill-will.

It's their way of telling Intel it's giving too much support to Linux and Android...

"We'll support your competitor if you support ours"

I used to agree with you, however their strategy makes a lot of sense when you realize that the desktop is just going to disappear from Windows all together. I think that in the future it'll literally just be treated like any other app, and you'll be able to download 'nix desktop apps alongside the Windows desktop app. Honestly with what they're doing with Hyper-V, this is becoming more and more plausible.

Reply on news.ycombinator.com