Former Micrsoftie (left in August) and founder of a startup that exclusively targets Win8 developers here. There's some FUD here that I need to clear up.
1. All Metro / Windows Store apps have to be distributed through the Windows Store for mass distribution - that much is true. There are good reasons for this - namely it saves you the trouble of having to write installers for three separate hardware architectures (ARM, X86, X64) and makes it comparatively low-friction to distribute your app compared to traditional desktop applications.
2. It's pretty easy to side-load applications on a one-to-one basis for developers who are running non-ARM devices; you can package an app automatically from Visual Studio with a self-install PowerShell script and installing an app is as easy as emailing someone a zip file with the app package + install script. This was how we tested apps among distributed teams internally at Microsoft and how all of the early third-party developers did it. I only did this on X64 devices so I'm not sure if there's support for X86 and I'm pretty sure ARM doesn't support this.
3. Enterprises have the ability to create their own bolt-on to the app store using Active Directory to ensure that all of their employees can get access to private, line-of-business Metro apps that they don't want exposed in the public app store.
4. Traditional desktop applications, non-Metro ones, can still be distributed the old fashioned way. My previous employer before Microsoft has no plans to put its desktop software in the Windows Store and frankly I don't think it will affect them much.
For the desktop, look at Metro as a way to have some content-focused apps and games that come to life directly on your desktop through live tiles. Don't look at them as replacements for heavy-weight, traditional desktop applications. That's not their point.
Windows isn't going to be a closed platform - if anything Metro makes Windows more accessible to a larger number of developers. Shipping a traditional desktop application is an epic pain in the ass compared to shipping a Metro-style app. Metro makes it a lot easier for most developers to sell native apps on Windows.
But I'd like to know, was the downvote (or votes) related to:
- Pointing out that Microsoft has gateway approval authority is old and tired.
- Using the term politically incorrect is old and tired.
- Boobs (including porn, fine art, vacation pictures etc.) should be hidden.
- Fart apps are too low brow.
- No legitimate user needs surveillance evasion tools.
- Microsoft has a social or corporate responsibility to prevent the spread of surveillance evasion tools.
- Only unsavory users need to know how to make bombs, or learn about how people make bombs to understand those people.
- Microsoft has a social or corporate responsibility to prevent the spread of information on bomb making.
- Microsoft would never disapprove an app based on anything like the above.
- Other.
Yes, I am free to avoid the ecosystem as a developer and consumer, and honestly I probably will. But I think discussion of corporate throttling of expression is necessary.
> For the desktop, look at Metro as a way to have some content-focused apps and games that come to life directly on your desktop through live tiles. Don't look at them as replacements for heavy-weight, traditional desktop applications. That's not their point.
I don't think Microsoft has communicated this at all. I don't think there is any definitive statement that the desktop is not being deprecated, and will continue to exist and be developed for.
They don't need to issue a definitive statement on it, because that's not a possibility.
Why would Microsoft set a 250+ billion dollar ecosystem on fire (third party Windows desktop software) when it's responsible for locking in all of Microsoft's enterprise customers?
It's probably fair to assume that the desktop won't be going away soon by the fact that they've improved it on Win8 (there are a lot of non-metro improvements in Win8 - surprised?). If the desktop were there right now purely temporarily then I doubt MS would have spent time improving it.
Maybe one day they'll nuke it, but they have a long way to go.
> There are good reasons for this - namely it saves you the trouble of having to write installers for three separate hardware architectures (ARM, X86, X64) and makes it comparatively low-friction to distribute your app compared to traditional desktop applications.
That is a convenience of using the app store. That is not a good reason to lock out other channels. Nothing intrinsic to either option necessitates making them mutually exclusive.
"Shipping a traditional desktop application is an epic pain in the ass compared to shipping a Metro-style app."
No, it's not. Desktop software installation and update packaging is a solved problem. On top of it... Multitude hosting and payment processing providers to choose from, nobody puts any restrictions on whether your app aligns with store owner's overall strategy, how/whether you can process payments outside the store, better margins, etc. "Ease of discovery" is overrated: maybe store makes it easier for big partners to be more prominent, but the more it grows the more it becomes just another silo, and you still must do your own legwork to market your app, making 25-30% cut look even worse. So, yeah, Windows does take a route to become more closed platform.
Uh huh. So, I take it then you've actually built a Windows desktop application that was popular / important enough that you had to do all of the following on a startup / small-business budget:
1. Installers that have to cover three versions of Windows, two hardware architectures (x86, x64), and thousands of different localization permutations.
2. Deal with piracy, cracking, and malware-ified version of your own company's desktop applications.
3. UI design in C/C++ that didn't look like absolute shit. Oh, and let's go ahead and throw foreign language localization into the mix too (German words are looooooong)
4. Company-defined security policies / UAC / third party virus scanning technology that can interfere with the install process in ways you couldn't even imagine.
5. Bad images and weird registry issues caused by things that are largely out of your control, but you have to account for anyway.
The only item on that list Metro forces me to deal with (if I want to) is localization, which is 10000x easier to do in HTML5/JS or C#/XAML than it ever was in C.
If all of those things seem simple to you, I'm calling bullshit.
3: Desktop application != C++ application. C#/WinForms, C#/WPF are desktop applications as well. But yeah, developing decent UI in MFC is quite a bit of pain. Although, you could take some shortcuts with implementing UI with HTML/js, loading them from resources. I did that once, it is possible. Localization is a solved problem for both native and .NET.
4: So? Company policies and UAC doesn't go anywhere with Store. Antiviruses - yeah, those are more painful for regular software.
5: Applies to Store too. If something is wrong with customer's PC, it will be wrong regardless of your path of distribution.
Look, software is never simple (unless we're talking about fart apps), but the Store is not some kind of unicorn that alleviates all problems. Yes, certain things will be simpler, but usually the bigger problem is marketing and promotion and for its costs and risks of rejections/removals Store just not worth it.
I guess the point I'm trying to make is that technical difficulties during "traditional"desktop development are, while real, more of a red herring with app stores: the main problem is sales and Store doesn't really solve it.
Bad phrasing on my part - I mean to say that I don't know of any similar way to configure a private Mac app store. I feel like the concept could be a great way to manage distribution.
> Windows isn't going to be a closed platform - if anything Metro makes Windows more accessible to a larger number of developers.
If I have to go through Microsoft to get a Metro app onto Windows, then at the very least...the platform is partially closed.
Besides that, I think that many, many people already consider Windows a closed platform (I certainly do), so I guess it depends on your definition of "closed".
I don't really understand why folks are hung up on this; this is par for the course when you're using a centralized app store.
Having said that, let me do my best to try to address some concerns you might have.
1. The Windows Store is really transparent about why your app failed certification - first time our app failed it was because we forgot to attach some art assets (icons) to our app. This is the way it was with the Windows Phone marketplace too - when some of the startups I worked with had apps fail for more content-related reasons, it was always followed with an explanation as to what (1) region-specific content restriction was the source of the issue [don't submit apps with maps of disputed territories to China or fashion apps to countries with Sharia law] and (2) were reponsive when asked follow-up questions. This isn't the Apple store - Microsoft has a much higher investment in its relationships with developers.
2. The Windows Store terms of service are reasonable / understandable by human beings, and they give you the automated tool they use for checking for bugs / permissions issues (i.e. if you make calls to the location service, you have to make that explicit in your app manifest so the end-user is aware of it) so you can run it yourself before submitting.
3. Turnaround time is pretty fast now - our app got approved after we re-submitted within about 8 hours. It'll slow down more once the volume of submissions goes up.
Overall, I think it's a lot easier and cheaper to both develop and distribute a Metro app that looks good, works well, and is easily accessible to a large market. You don't have to invent your own distribution channel, and the development experience is much easier and gives you better-looking results right out of the box.
I'm not sure what you're responding to. What is par for the course when using a centralized app store?
I simply stated that if you must go through a third party to get your apps to the consumer, then that platform (or at least that part of it) is "closed". You responded with the value proposition, but I'll be glad to address that below.
I understand the value of Microsoft's offering and I'm simply not interested on principle...and I'm speaking as a hard-core Windows user. I love Windows, I think it's far better than the alternatives, but the reason I like it is because I get the right mix of quality and flexibility (read: freedom).
I'll be damned if Microsoft is going to force me into using only their store to buy and sell Metro apps. The only way I'll ever use Metro or develop for it is if Microsoft opens up the channel for alternative stores and side-loading. Barring that, I may use it begrudgingly if someone jail-breaks it and develops a store like Cydia.
Honestly, I'm sure the Windows Store is going to be great and all, but that's not my concern. I will not participate in a captive market.
1. All Metro / Windows Store apps have to be distributed through the Windows Store for mass distribution - that much is true. There are good reasons for this - namely it saves you the trouble of having to write installers for three separate hardware architectures (ARM, X86, X64) and makes it comparatively low-friction to distribute your app compared to traditional desktop applications.
2. It's pretty easy to side-load applications on a one-to-one basis for developers who are running non-ARM devices; you can package an app automatically from Visual Studio with a self-install PowerShell script and installing an app is as easy as emailing someone a zip file with the app package + install script. This was how we tested apps among distributed teams internally at Microsoft and how all of the early third-party developers did it. I only did this on X64 devices so I'm not sure if there's support for X86 and I'm pretty sure ARM doesn't support this.
3. Enterprises have the ability to create their own bolt-on to the app store using Active Directory to ensure that all of their employees can get access to private, line-of-business Metro apps that they don't want exposed in the public app store.
4. Traditional desktop applications, non-Metro ones, can still be distributed the old fashioned way. My previous employer before Microsoft has no plans to put its desktop software in the Windows Store and frankly I don't think it will affect them much.
For the desktop, look at Metro as a way to have some content-focused apps and games that come to life directly on your desktop through live tiles. Don't look at them as replacements for heavy-weight, traditional desktop applications. That's not their point.
Windows isn't going to be a closed platform - if anything Metro makes Windows more accessible to a larger number of developers. Shipping a traditional desktop application is an epic pain in the ass compared to shipping a Metro-style app. Metro makes it a lot easier for most developers to sell native apps on Windows.