Three Screens and a Cloud


This is a phrase I think we’re going to hear more of from Microsoft over the coming months and years. In many ways it nicely captures the the end user vision for Microsoft.

Ray Ozzie has used the phrase twice in as many weeks, first at the Technology Alliance luncheon in Seattle and then again yesterday in a fireside chat with JP Morgan. Yesterday he said

“So, moving forward, again I believe that the world some number of years from now in terms of how we consume IT is really shifting from a machine-centric viewpoint to what we refer to as three screens and a cloud:  the phone, the PC, and the TV ultimately, and how we deliver value to them.”

There were a tonne of other interesting things during the JP Morgan session including Microsoft's approach to cloud and five areas where Ray believes we have significant advantage in this transition relative to these other competitors. I’ll post on that some other time as I’d like to dwell on three screens and a cloud for a while.

I think three screens and a cloud is really an evolution of what Ray said back in his last memo where he positioned the web (aka cloud) as the hub and a world of devices. In terms of the consumption of information we typically consume it via one of three devices – the phone, the PC and the TV (not sure what the radio folks would say). On the PC and mobile we’re making progress in connecting them to the cloud through services such as Live Mesh and MyPhone (which went in to public beta this week). Netbooks have sped up this process in many ways as a companion device to a larger PC you can get a great experience through sharing or synchronising your devices through the cloud. Clearly we have some very strong competitors in this space in the shape of Apple with their laptop range and MobileMe and of course the iPhone which is a remarkably well connected device with lots of great cloud applications from the App Store. Google is entering the game with their Android OS on both phone and small form factor PC’s.

On the PC and mobile side there is very strong and admirable competition which should be good for the consumer as each of the companies forges ahead to try to leapfrog the other. The TV side is a much more cloudy picture if you’ll forgive the pun. It’s actually not the TV that is the battleground here – more so devices that use that large screen or screens in your home to make it another point of delivery/sync with the cloud. As I thought about this earlier today, it struck me that the expensive LCD panel I have in my own home is a pretty dumb device in reality – all the smart stuff comes from the boxes that are connected to it. In my case those boxes are a Windows Media Centre, an XBOX360 and a Sky HD box – all driven by software to light up the relatively dumb TV.

Apple has Apple TV in this arena and as far as I know, it’s not somewhere Google plays at this point. As BusinessInsider noted today, this is an areas where Microsoft is strong and with announcements like the integration of Netflix streaming in to Media Centre we’re getting stronger. I’d personally like to see more of these content deals outside of the US but the reality is I actually want much more from my TV. If it’s to replace the PC as the screen I look at most there is going to have to be much smarter integration of web content than there is today.

Coming back to the beginning then…I generally like this new terminology as a way to explain to consumers what this new era will look like and what part Microsoft will play. I’m not 100% sure that “cloud” will ever be a term that is instantly understandable for the average consumer but the philosophy is sound I think.

Where I think Microsoft has real potential is to connect these worlds. There are some great examples of this now on the iPhone – the Sky+ app that allows me to quickly and easily remote record any program is a terrific example – but there are many more ways I want to see not only my information but also my preferences flow between these three screens based on where I am and what I’m doing. For example I want an application that knows the moment I put a calendar entry in to Outlook that clashes with my favourite TV program to know it should send a request to by Media Centre or Sky+ box to record. Similarly I want my TV to give me the option of seeing the Twitter feed from my friends to see what they’re thinking as we’re watching The Apprentice.

There are many more scenarios like this so the three screens and a cloud mantra is as much a call to arms for our product teams as it is for setting a vision that consumers can understand. The era of the cloud is now upon us and the race is on to connect these 3 screens to it.

Comments (5)

  1. jon honeyball says:

    absolutely the race is on

    but the protagonists need to remember that the race is not one about baking in and locking in customers to your platform. The race is to provide best of breed tech on the platforms and an open, licensable etc back-end process which glues it together.

    If ms thinks that this is a good idea to (say) tie users into ms phones and xbox360, then it will have a very rude awakening.

  2. "As I thought about this earlier today, it struck me that the expensive LCD panel I have in my own home is a pretty dumb device in reality – all the smart stuff comes from the boxes that are connected to it. In my case those boxes are a Windows Media Centre, an XBOX360 and a Sky HD box – all driven by software to light up the relatively dumb TV. "

    This is a key statement. These days I typically call TVs "panels" as the main difference to focuse on between models is size/resolution/contrast/inputs; everything else is secondary.

  3. Tom Basham says:

    Hi Steve,

    That’s a really interesting post, and echoes some the thinking we’ve been doing here over the past few months.  As phones and online devices like sky boxes become more capable people are bound to start expecting them to interact.  Your sky+ example is a good one, and I think there’s some interesting stuff in there about where work info and apps fit into our three screens.  HP have some pretty good demos showing work info like calendar alerts and calls being surfaced on home TV screens during the breakfast news – not sure the misses would approve but potentially pretty useful stuff.

    I guess ideally you’d want the apps that do all these cool things to work on any device you chose, so runtime environments like Air or Silverlight could be the key.  Combine those with a nice App Store so it’s simple to access and buy them and you’re away – as long as you’re first and best to market 🙂

    A few months ago I wrote a post ( about the Live Mesh Desktop and whether it had a wider future in offering a more complete desktop for thin devices.  It was a little tongue in cheek, but I think the point is valid.   As you say TV’s are pretty dumb, but adding just enough technology to render a basic virtual desktop wouldn’t be too hard or expensive.  It would offer those devices any almost any service that you cared to hang off the backend services (given some smart protocols etc).  It would also widen the audience beyond those who typically buy Apple TV’s, Xbox’s or PS3’s.  If you could run those thin desktops as ‘channels’ to add some healthy competition even better.


  4. Sound of reason says:

    Great post. No, really.

    Jon makes a good point about cross-platform inter-operation. The only open option at the moment is predictably open source, with the good and bad that brings. DLNA is growing, but it’s an ugly mish-mash that needs to focus on standards and implementation.

    The proprietary systems at the moment such as Apple TV and XB360 are good, but require tweaks to get enhanced experience. Get yourself a media PC / Mac mini and the options increase greatly, but so has your budget.

    I think your thin-client thinking has been the Great White Cloud hope for some time, but we are at least making steps. Slingbox is one proprietary example and it’s sad that Apple and AT&T have colluded to hobble the app on iPhones when it works well on other mediaphones.

    Ideally your phone should be another interface that works both with the home panel as a controller, then independently as a remote interface. We’re pretty much there in some ways, but DRM and wider copyright is creating some hindrance.

    This also seems to relegate traditional broadcast and cable tv to a poor second, or even the pages of history. This thought is possibly going off-topic though.

  5. XBOX’es play nice with iPod’s Jon and I’d argue that Windows Mobile is more open than some other platforms out there 🙂 …but these are two simple examples and I agree with your point, openness is key to success here.

    From a cloud perspective I was pleased to see a demo at MIX09 of our Identity Service and Service Bus with an app written in Python and deployed into Google App Engine.

Skip to main content