A comment Matt Rosoff over on CITEworld’s re-publication of my post on Microsoft’s current approach to developer (and other customer) engagement inspired me to do a thought experiment. What if Microsoft had approached “Windows 8” differently? There are a number of scenarios that they could have followed, so let’s explore a few of the more likely ones.
Most of the criticism about Windows 8 is around how Microsoft attempted to bridge the gap between a modern touch-based, tablet-centric (if you will), UI and the traditional desktop. Now this is actually a complex situation, because we aren’t just talking about UI but also about a new app model. In theory you could have one without the other. In previous posts I’ve talked about why the overall re-imagination of Windows, including all these elements, was important. So as one thinks about various alternative scenarios you have to consider that not all of the modernization would have necessarily occurred.
What if Microsoft had kept tablet support separate from desktop Windows? That is, what if Microsoft had three different versions of Windows (Windows Phone, Windows Tablet, and Windows Desktop) with a common underpinning (NT) but separate user experiences. Windows Tablet would have been the full-on “Metro” experience, including finishing the job of completely eliminating the Desktop. Windows Desktop would have been more of a 7.5 product, combining the underlying architectural improvements in Windows 8 (faster boot, Secure Boot, better multi-monitor support, SmartScreen, etc.) with minor tweaks to the Windows 7 desktop UI experience and Win32. What Windows Desktop would NOT include is the Start Screen, Contracts, or the ability to run Windows Store apps. Why not? Well, the claim is that desktop users hate those so why not keep the environment clean? (And if did include them, then it would just be Windows 8 as we know it!)
What would have happened in this scenario? Well we know what would have happened to Windows Tablet because Windows RT is Windows Tablet V1.0. Windows RT is struggling due to the relative paucity of Windows Store apps. The main criticism of Windows RT is that it can’t run desktop apps. But the ability to run desktop apps is exactly what gives you the “jarring” and unnatural experience that people complain about. So analyzing the first few months of Windows RT suggests that a pure tablet OS from Microsoft would have failed.
As I mentioned a Windows Desktop 8 would have seemed more like a “.5” release than a real next generation. One reason for that would be that a lot of resources were tied up on the Tablet effort. But the bigger one is quite simple, we’re exploring this scenario because many users claim they don’t want the paradigm to change much. And as long as that is true then most of what you do is pretty minor. So what would Windows Desktop have done in terms of the market? Nothing. Absolutely Nothing. The decline in traditional PC form factor sales would have continued, at a rate no higher or lower than we are already seeing. People who need a new PC would buy a new PC independent of if it ran Windows 7 or Windows Desktop 8. No one would rush to buy a new PC just because of Windows Desktop. Corporate adoption of Windows Desktop would be no quicker because they would stick to their existing schedules. It might actually slow down because they were busy deploying iPads or other tablets and, with Windows Desktop 8 not offering a Tablet alternative, they could afford to skip it. Consumer upgrades of Windows 7 systems to Windows Desktop 8 would probably be a bit more robust, but that was already a fairly insignificant business for Microsoft. Basically, it wouldn’t have moved the needle on the health of the overall PC business.
So the bottom line for the above scenario is that Microsoft would be at best in the same position it is in today, and more likely in much worse condition as it would have completely bombed out in the tablet space. They would have been tagged with “they don’t get it” and relegated to the dustbin of technology leadership.
For our next scenario let’s just take a minor variation on the above and suggest that Windows Tablet was not the Metro experience of Windows 8 as we now know it, but rather an evolution of Windows Phone 8. Now this is interesting in that Windows Phone already had some momentum amongst developers and a well-regarded user experience amongst end-users. But it had no traction in the market. And its Achilles Heal is the same application library problem as facing Windows RT. So I don’t see how things would be different in this scenario. Windows Desktop would be a highly regarded addition to a shrinking market segment. Windows Tablet would be Android 3.0 from a market perspective (e.g., Phone on steroid experience rather than being evolved for the tablet, OK Phone app library but almost no apps designed specifically for a tablet, etc.) and critics would be quick to hammer home the comparison.
If Windows Phone market share was exploding then a WP-tablet strategy might have been more successful short-term for Microsoft in the tablet space. But with Windows Phone struggling, this strategy would not have yielded short-term (nor probably long-term) success in tablets.
Next up would be the strategy the most frantic Windows 8 critics believe Microsoft should have followed. Retain the Classic Windows user experience and evolve it to (optionally) be more touch friendly. Add a mode that makes the Start Menu more touchable. Add spacing and size to common controls etc. as Office 2013 does and as Windows Mobile 6.5 did. Improve on the existing app model, perhaps by basing it on .NET or by a somewhat evolved Win32. Etc. It all sounds good except when you consider two things. (A) Been there, done that, and all I have to show for it is a T-shirt and (B) it makes the assumption that the technical mishmash that would be created would be cleaner, less jarring, and any better received than today’s Metro/Desktop duality.
On the first point, Microsoft has a long history of trying to evolve the Win95 desktop UI model to address mobile computing. When Windows CE was created the UI was modeled on Windows 95, Start button and all. The Handheld PC made no headway against the Palm Pilot, and much of the criticism of the HPC was specifically aimed at having used the desktop paradigm. So for the Pocket PC Microsoft started to move away from trying to look like Windows 95 and the more it did that the more successful it was! This culminated in Windows Mobile 6.5 (which itself borrowed from the abandoned “Book of 7” Windows Mobile 7 plan), that went the next step in making that UI family truly finger friendly. It was too little, too late, as Apple had already redefined the entire user experience and app model expectations. Windows Phone 7 dispensed with WM compatibility in order to leapfrog the iPhone.
Of course Microsoft also tried to evolve mainstream Windows to be more touch friendly, from the Tablet PC work itself to Origami to having Windows 7 support modern capabilities at its core (capacitive multi-touch). What they didn’t do was try to alter the basic Windows desktop UI model nor fix the app model (so, for example, uninstall actually worked or apps were sufficiently isolated to keep them from interfering with one another). Lots of Windows 7 PCs and convertibles were available with touch capability. None sold in great quantity. Few that were sold actually saw fingers hitting screens. In fact once the iPad was out the notion that Windows 7 was touchable was considered laughable. And that makes the notion that one could have sufficiently evolved the existing desktop UI model to be competitive, highly questionable.
The second part of this scenario is that leap of faith that when you were done evolving the existing user experience etc. the result would be any better than what Microsoft did by moving to the Metro experience. I happen to believe it would be the opposite. The closer you got it to something that would work in the tablet market the more bizarre an evolution it would seem to desktop users. The more you catered to desktop users the more likely it would seem like a kludged force feed of the desktop PC on to a tablet.
My actual bottom line on this strategy? It would have accelerated the decline of the PC business while failing to gain any traction in tablets. All the criticisms leveled at Windows 8 would have been repeated, with only modest changes in wording.
So if the alternate scenarios are worse than what Microsoft actually did, yet Windows 8 has not yet captured the hearts of the masses, what “scenario” would work?
Here is the problem plain and simple. Windows 8 is a V1 product and it needs to be a V2 product. Take the Start Screen vs. Start Menu debate. It isn’t that Microsoft needed to retain the cascading start menu, it is that it needed to provide a reasonable alternative for desktop workstation users. In previous blogs I’ve thrown out an example. Why isn’t there a snapped view for the Start Screen? Then when in the desktop you could have Start bring up the snapped view (assuming a monitor that supports it, which is overwhelmingly the case) instead of losing the desktop to a full screen Start Screen. Why on high-resolution monitors can’t you have multiple snapped views, or even a couple of “full screen” views? That would mitigate the desktop user complaint of Metro not being suitable for large monitors. I don’t think Microsoft completely ignored these questions (and as I’ve mentioned before, I once saw a Windows 8 build demoed with a snapped view on the left and one on the right with full screen view in the middle), I think they just didn’t make the cut for “V1”.
Overall I think the direction Microsoft chose, and the decision to force accelerate the move to the new app and user model, was the correct one. I would more question some of the individual tradeoffs that were made to make sure the release was in the market for the holiday 2012 shopping season. And, as my previous post discussed, is Microsoft’s current penchant for secrecy making this situation worse? Imagine if Microsoft was talking about, and perhaps even demoing, my idea for a Snapped Start Screen (or some other alternative) already. And publicly promising a free upgrade to the release containing it for Windows 8 purchasers. And even talking about that release being later this year. I doubt Microsoft would have lost a single Windows 8 sale. In fact, I think the change in criticism (from “Windows 8 is a disaster for desktop workstation users” to a softer “some users would be better off waiting a few months for the update” would actually boost overall Windows 8 sales. When it comes to computing devices, even consumers often buy-in to where you are going more than buying the specific product.