Slashdot Mirror


Future Microsoft Devices Will Take Cues From the Surface Tablet

An anonymous reader writes: Microsoft CEO Satya Nadella says the company is committed to bringing Windows to as many computer form factors as possible — even if they have to do it themselves. He says their plan is to build out new devices with the same mindset that created the Surface line. The Surface Pro tablets (and the regular Surface tablets, now that Windows RT has been retired) have been a rare bright spot among Microsoft's mobile stumbles. Nadella seems to want Windows to become almost hardware agnostic, and he thinks the universal apps plan for Windows 10 is the way to do it.

He says, "Universal Windows apps are going to be written because you want to have those apps used on the desktop. The reason why anybody would want to write universal apps is not because of our three percent share in phones. It's because a billion consumers are going to have a Start Menu, which is going to have your app. You start the journey there and take them to multiple places. Their app can go to the phone. They can go to HoloLens. They can go to Xbox. ... And by the way, when we hook them on that, we have a phone app. This strategy is path dependent, which is a term I use that means where you start is not where you end up. And therein lies a lot of the nuance. The fundamental truth for developers is they will build if there are users. And in our case the truth is we have users on desktop."

4 of 119 comments (clear)

  1. Enterprise by tom229 · · Score: 4, Insightful

    He's still missing a big opportunity: the enterprise. Why everyone is clamoring for the crumbs of the consumer pie, I don't understand. Enterprise functionality is being ignored forcing us to adopt strange concepts like BYOD which is a logistical nightmare and security concern.

    Dominate the enterprise and the consumer market will follow. Gates knew this. Balmer seemed more interested in chasing the heels of the current trend as most sales guys do. And now I'm not sure what to think about this new guy... But he seems to be still missing the point.

    --
    If it ain't broke, don't fix it.
  2. Who wants what now? by Somebody+Is+Using+My · · Score: 4, Insightful

    "Universal Windows apps are going to be written because you want to have those apps used on the desktop."

    Wait, who wants universal Windows apps?

    Certainly it is not the desktop users. Because they must cater to the "lowest-common-denominator" of hardware, universal apps tend to be underpowered and have interfaces poorly optimized for mouse/keyboard.

    The developers have little care for Universal apps. There is no demand for the things, and requires an investment in learning new development methods. It is an added expense and complication that brings little reward for the extra effort.

    I suppose there might be some demand from Windows Winphone users - all six of them - but even they might prefer a more functional app tailored to their desktops capabilities rather than a cut-rate smartphone app. I don't hear an overwhelming clamor crying out, "oh if only the mail app on my desktop worked just like it did on my winphone!"

    No, there is only one party that is really interested in Universal apps, and that's Microsoft themselves because universal apps are sold through the Microsoft app store and they get a cut of the proceeds. It also gives them great control over what sort of programs users have access to (what are the odds they would allow a stand-alone Linux installer to be added to their store?).

    So, other than some great desire to increase Microsoft's profits, what reason is there to develop or use Universal apps?

  3. Re:bullshit translator go: by Anonymous Coward · · Score: 3, Insightful

    There is also the problem, independent of Microsoft's particular business strategy or platform management/mismanagement, that 'universal' applications just don't really work.

    At a basic level they can be done; assuming that you stick to the aspects of .net that stay fully in the CLR and don't drag in some useful-but-legacy entanglements, you should indeed be able to take the same binary and run it without errors on anything from a fairly punchy microcontroller running .net Micro on up through phones, consoles, and PCs. So far, so good; or at least adequate.

    The trouble is that different situations require different sorts of application designs. Mouse/keyboard and touchscreen are not even remotely equivalent; different things work better or worse on each. Small screens require very different tradeoffs from larger screens or multiple screens. 'Best experience' on a heavily battery constrained device likely involves a variety of very careful restrictions on any resource usage that isn't strictly necessary. On a mains powered device, preemptively doing everything even remotely likely so it will be available in RAM instantly if the user chooses it is probably what they want.

    It is architecturally possible to build something that will run everywhere; but unless you manually design and glob together all the UI/UX cases into a 'single application', that doesn't really address the problem. It's more elegant than having to compile the same code for 8 different target architectures or whatever; but that only saves you CPU time, not the time you need to adapt the chunk of your program that can be reused to the particular needs of each environment. Some of that may be modestly automatable(resizing UI elements based on reported physical screen size and DPI to preserve the same perceived size, perhaps); but doing it well is going to require a great deal of care and attention and time.

    People don't really want a 'universal app', they want to be able to interact with a given service, set of data, gameworld, tool, etc. when and where they wish; if the constraints of the platform allow it(data entry on a phone just isn't ever going to be fun, doing confidential stuff in public isn't going to work, console controller and mouse/keyboard gamers in the same multiplayer match may be a problem, and so on). This obviously benefits from some amount of code re-use; but it can't be addressed by a 'universal app', except in the weak sense that you can glob all the apps into one bundle if you go to the trouble of designing them all.

    Microsoft is at least theoretically capable of making 'universal' technically possible; but nothing short of strong AI is going to make 'universal' automatic or useful.

  4. Re:bullshit translator go: by tompaulco · · Score: 4, Insightful

    The trouble is that different situations require different sorts of application designs. Mouse/keyboard and touchscreen are not even remotely equivalent; different things work better or worse on each. Small screens require very different tradeoffs from larger screens or multiple screens. 'Best experience' on a heavily battery constrained device likely involves a variety of very careful restrictions on any resource usage that isn't strictly necessary.

    Exactly this.

    People don't really want a 'universal app',

    While this is true, they unfortunately THINK they do, or at least Management thinks that people want a universal app.
    As a developer, I am often asked to make my desktop application work on Mobile. There are many things that just suck on Mobile. You can't display vast amounts of data on a 3 inch screen like you can on a 24 inch monitor. You can't type in data very efficiently on a mobile. Honestly, there is nothing on a mobile that works better than it does on a desktop. A mobile is what you are forced to use if you don't have a desktop handy. But everybody wants your wizbang application that looks and works great on a desktop to be available on a Mobile.
    I even play mobile games on my desktop using BlueStacks App Player. I hate the UI on the phone, but it plays great and looks awesome on the desktop.

    --
    If you are not allowed to question your government then the government has answered your question.