I've Seen the Future of Computing: It's a Screen.

I've Seen the Future of Computing: It's a Screen.

Summary: In the not so far off future, computing for most of us will be reduced to remotely delivered subscriber services, running on cheap, commodity high-definition display units.The last few weeks have been a rush of virtualization and cloud-based announcements.


In the not so far off future, computing for most of us will be reduced to remotely delivered subscriber services, running on cheap, commodity high-definition display units.

The last few weeks have been a rush of virtualization and cloud-based announcements. In December, IBM announced a partnership with Canonical to deliver virtualized Ubuntu Linux desktops for enterprises.

Earlier this month VMWare followed with the availability of open source code for a virtual remote desktop viewer client, and HP countered IBM's virtual desktops with a virtual desktop blade solution with Citrix.

In the last few days, we've heard about IBM and Amazon cloud partnerships and Red Hat and Microsoft agreeing to certify each other's operating systems on their respective hypervisor platforms, Hyper-V and oVirt/KVM.

This is just the beginning of a huge rush of virtualization and clouded services announcements to come this year. On the heels of VMWare's VMWorld Europe event next week, Citrix is expected to announce that their upcoming XenServer 5 Enterprise bare-metal hypervisor will be released for free, which was previously a commercial product that used to sell for $900 per copy, and includes resource pooling/clustering and live migration capabilities which is competitive with VMWare's ESX 3.5 and VI3 stack that sells for several thousand dollars per license.

Click on the "Read the rest of this entry" link below for more.

What does all this virtual and cloud stuff mean? Why is this all happening in a rush now? It could be that a economic scarcity is acting like a catalyst -- tough times are forcing the movement to virtualization and virtual desktops in the enterprise, and it's being pushed out into the cloud, and eventually to consumers and end-users.

If we project this sort of technological change several years into the future, it might very well look something like what I wrote about during last summer.

Josef Konsumer's 2016 HD-connected "ThinTerm" may sound like science fiction, but it's really not that far-fetched. If we can look and see where we are heading with embedding Internet-enabled chipsets into Digital TV's for multimedia content streaming (such as with devices like the Roku) as well as other projects such as Ndiyo to embed thin client chips into commodity monitors, we start to see an emerging trend.

What we are seeing is the advancement of low-power high performance embedded systems for sophisticated next generation Linux based set-top devices as well as an upsurge in popularity of MIDs and Netbooks where the balance of the user experience is Internet-delivered. The future is not the "Desktop Computer" or even the "Notebook", or even today's "Netbook" or "Smartphone". The future is The Screen.

The Screen is what I believe the future end-state for personal computing will eventually become, likely within the next 10 years, or possibly even less, due to the commoditization of hypervisor virtualization technology combined with a renaissance of the centralized computing paradigm (the mainframe never died folks, he was just sleeping and waiting for this moment to arrive) and a maturation in virtual infrastructure management platforms.

These otherwise enterprise-targeted technologies will will trickle down to the end-user via the standardization of rich desktop delivery thin-client protocols (such as Novell's Compiz-enabled NOMAD for Linux RDP or Red Hat's SPICE), low-power device chipsets, broadband and high-speed wireless deployments, and adoption of High Definition/Digital TV.

Okay, forget the buzzword bingo for a minute. What is The Screen? I don't think it has been well defined what the interface or the experience really is going to look like, but I have a very good idea.

Certainly, I'm not expecting anything along the lines of Minority Report or even something like Microsoft's "Surface", although it's certainly possible that some day, people might use UIs like that for certain niche applications. Initially, early versions of The Screen will almost certainly look very much like the platforms you use now -- Windows, Mac, and definitely Linux.

The only difference is you won't own the computing hardware it runs on -- all you'll really need is a screen (an HDTV with HDMI inputs) mouse, keyboard and broadband, and you'll be buying your computing services like a utility, just like you pay your electric or Cable TV bill today. And like your Cable TV bill, you'll subscribe to computing "Channels", complete with applications and hosted data, with balls to the wall clouded backup services to match.

Someday, all these capabilities will be built into every HDTV unit, but the initial Screen will likely be deployed using some sort of carrier-provided thin client box, perhaps based on a low-power Linux device running on something like a BeagleBoard with an Android-based session manager UI with some basic local applications for cached data use or direct content streaming (a la Roku) and costing less than $100 to manufacture.

I suspect that Linux will finally make significant desktop inroads when it can be spoon-fed to consumers as "Basic Screen", much like you have your Basic Cable now. The Open Source and Free nature of Linux desktops such as Ubuntu along with Open Source desktop applications will be a compelling basic offering for Screen service providers because of the fact that they won't have to buy nearly as many software licenses to host them -- they'll develop them and tweak them in-house.

Frankly, so much of your typical desktop experience will be Web-based, so there will be little cost justification for paying extra for Windows or Mac on a subscriber channel.

Compared to the ever present Linux hardware/software compatibility issues we have today, things will "Just Work" because the provider will be able to completely tweak the environment for their subscribers. As a result of Screen deployments, today's Linux desktop migration issues will no longer exist.

However, some of you will inevitably opt to pay premium money for the Microsoft Channel or the Apple Channel, and some of you might even buy Premium Screens from Apple or another vendor that provides enhanced multimedia capabilities or some other feature that Basic Screen doesn't have.

The Playstations and XBOXen of the future will be nothing more than high-powered Screen servers for those people that want to play games -- and yet, I suspect that even those specialized functions are likely to be commoditized as integrated chip sets get more and more sophisticated. The focus will be buying on-line content from Microsoft and SONY, not the game units themselves.

And of course, in addition to our home Screens, we'll have Mobile Screens -- things like MIDs and Netbooks, as well as Smartphones. The difference between the Mobile and Netbook/MID devices of the future as opposed to what exists now is that they'll exist to connect you to your Computing Channels that you subscribe to, or to present data and services from clouded applications that you access frequently, such as what Amazon or Google is trying to create.

Local data? Sure, perhaps you'll sync up emails,  and other recent data for store and forward purposes if you get knocked of the Net (a la Google Gears) but all your important data will be in the Cloud. Ubiquitous 5G/6G/nG high speed wireless access will be everywhere, so the notion of "What if I can't get on" will be as anachronistic as we view manual typewriters or rotary phones today.

Before you say, "Stop dreaming, Perlow" I acknowledge that yes, some people will still require dedicated PCs -- like my younger brother who does 3D modeling for Hollywood movies or people who do sophisticated video editing. But most of us won't need or even want them anymore.

Why worry about maintaining systems or installing apps when we can have it spoon fed to us? Every application, game and service we could ever possibly want will be completely on-demand.

February 17, 2009 was a major step in the progression towards The Screen in that the first of the major barriers came down, which was the initial demolition of this country's analog broadcast TV infrastructure.

There will be a great many holdouts to move toward broadband-delivered digital content, particularly from the older generation, but even they won't be able to stop the march of technology towards the digital world. Eventually, when they see the benefits that The Screen brings, and how it will liberate them from the Personal Computer which has become too cumbersome and too complex to own and maintain on their own, they too will relent.

Are you waiting for The Screen? Will it happen in 5, 10, or 20 years? Talk Back and Let Me Know.

Topics: Operating Systems, CXO, Cloud, Hardware, Linux, Open Source, Software, Storage, Virtualization


Jason Perlow, Sr. Technology Editor at ZDNet, is a technologist with over two decades of experience integrating large heterogeneous multi-vendor computing environments in Fortune 500 companies. Jason is currently a Partner Technology Strategist with Microsoft Corp. His expressed views do not necessarily represent those of his employer.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • cloud?

    So does this mean that I will be able to run my Access 97 databases from the 'cloud', what about all the 100mb photos I have in PSD format, or how about the music files I'm working on on my mac using Logic Express?

    Where is all the versions of Access to be installed?, lets see, version 2, 97, 2000, XP(2002), 2003, etc
  • Obsessed by the cloud, aren't you ?

    Are you waiting for The Screen? Will it happen in 5, 10, or 20 years?

    What about not in a foreseeable future ?
    I will never understand why some are so egearly waiting that all the computing come from the cloud and that all ,if not most of, computers become glorified thin clients.
    That smartphones, netbooks and tablets become mostly thin clients could be interesting.However assuming that smartphones are also becoming powerful handheld multimedia powerhouse, it doesn't seem that they will only be glorified thin clients.
    That desktops,laptops or even game consoles become just glorified thin clients would be a fantastic regression and make people too dependent of internet and of networks. In a country such as mine this is not even thinkable and would not be even in the context of an enterprise for a very long time.
    • Let me try to explain...

      I live outside of town and there is no bus/train service to my home.

      I have an old Porsche 911 which I love and will keep until it breaks in half. I enjoy this car because it has no ABS, no airbags, really not much except that it goes from zero to what seems to be the problem officer in a few seconds and it makes me smile like little else. It is loud, and scary fast, and if you disrespect it's abilities it will make you suffer for it. I enjoy doing things like changing the oil and regular maintenance myself. I will never get rid of this car short of a major catastrophe. Think of this car as my PC.

      I also drive a brand new Jetta 4 door. It's peppy, with A/C for the summer and heated seats for the winter it's comfortable, it's well behaved on the road and, well, it's utilitarian. I use this car to go to work, to carry clients, in the snow or on rainy days. I would never consider changing the oil on this car myself, however. It's a car; it's transportation, and gets me where I need to go in comfort. I still have to add fuel and wiper fluid however, so there's still maintenance. Think of this as a thin client like a Netbook.

      Then there's the bus. It's not as much fun as the Porsche nor as flexible as the Jetta, but it's a whole lot cheaper than either and for someone who lives and works downtown it's probably a better solution. And just because you own a car doesn't mean you won't ever take the bus.

      The point is that most people's "need" for their own PC is an emotional thing. People that aren't passionate about their PC's most likely won't miss them. You'd be surprised how many people, if offered not to have to worry about their antivirus etc would be willing to give up their PC's. Imagine never having to change your oil or add gas ever again.

      One last thing. Cloud Computing does not have to be public. There is nothing preventing a corporation from creating it's own secure cloud and giving access to it's employees through the internet. Google and Amazon are not the only providers...
    • not just your country

      Let's just bring this home to everyone. I was one of those lucky ones who lost
      power for more than four days, in an ice storm this winter (not-so-upstate NY).
      Four days to replace downed cabling for half a million customers.

      When my user-base has a "large" outage of service, as many as 3% will be
      completely out service until I resolve the issue. Maybe it's social engineering
      malware or a power outage for their switch. Whatever. We lose some

      Now image a thin client outage. Network problems. Uh-oh. 100% productivity
      loss. Nice. Now let's expand that to the screen. Maybe that level of loss
      reaches whole regions.

      What company or goverment can afford that? Our own government can't
      evacuate one small city from hurricane flooding and we're going to one day trust
      all of our personaland business info to screens and clouds? Get real.
      • A good point, but...

        ...the issue with the power grid is that it is not redundant, therefor when it goes down everything on the grid goes down. Some power companies have redundancy in some of their larger segments but few have two circuits into their homes or businesses. Some businesses will have redundant power circuits, and diesel generators, etc. I was without power for 7 days in Southern Quebec during the Great Ice Storm ten years ago. I used my cell phone but never my laptop computer during the outage, and the house phone was out for about three days as well.
        The point is that there is a lot more redundancy built into the Internet than there is in the power grid. If your land line dies, you can use a cell. If your internet provider goes down, you can use WiMax (802.16n) or some other method. The fact is that a properly developed cloud can be much more reliable than a wired office with PCs because the Cloud can be hosted anywhere, be it New York or San Francisco, and the method to reach it can be one of many as well. If your office is hit by a bad virus and you don't have a back up datacenter you are toast.
        No, I'm afraid that that old argument is quickly losing it's relevance my friend.
        • Agreed, in part.

          Still, the immediate assumption made is that in your argument is that broadband internet service fail-over (to smart phones or vice versa) would become ubiquitous. You must agree that not even dependable broadband has reached that ubiquity, much less the capability of fail-over mechanics.

          I understand and hope for your argument to win out, but truthfully? I don't see it happening any time in the next century.

          Also, I just don't see people becoming productive with their smartphones where replacement of the "Screen" is concerned. Is it really possible, even on an ergonomic level? MMmmm. I think not.
          • All I can say is that Cisco stock is looking more attractive...

            But all joking aside, it's really just a matter of time. My first modem was a 300 baud accoustic, then I had a 1200, a 2400, a 9600, 14400, 28800, and then a 56k baud oh joy of joys! Then I discovered DSL, cable, WiMax (802.16n), and each time I thought "wow, this is pretty fast". I think it's just a matter of time. As to the smartphone, there's no need for the display to be limited to a 2x3" LCD. Back in the early 90ies I was on a concall with a bunch of HP engineers, and one of them said that in the future we would wear glasses with a 3D display built in and type on a flat piece of plastic. While not mainstream, this technology exists today. Who knows what will happen even 5 years from now. It was only 10 years ago that it was still claimed impossible to get a computer virus through email unless you opened the attachment. Remember that? As to ergonomics, do you really [i]need[/i] to type? Voice Recognition may have a new lease on life.
      • Cable?

        Do you have cable? How often does it go down? What if your computer usage were delivered over cable by the same company? Would you really have a problem?

        If there's a hurricane or major disaster, your computer isn't going to be usable, and your data is going to be a lot safer with an enterprise, off-site backup system (the company's) than using DVDs in your house (if you actually DO that).

        I'll paraphrase Google in their Apps, Education Edition promo: "Not a hundred years ago, power was considered an on-site service and no university would have considered allowing an outside company to supply something that important. Who will take that same stance now?"
      • Your suffering from brain burn.


        Internet redundancy is a long way from being nearly as wide spread as you seem to imply.

        Your just talking plain foolish right now. For starters, there are numerous internet breakdowns resulting right on sight at location that have nothing to do with the provider and whatever redundancy the provider has built in. You have to grow a brain because there are endless permutations where internet can breakdown leaving any fool who decided to rely completely on all computer related tasks to being supplied by a provider swinging in the breeze as far as productivity goes.

        Its one bad thing to loose your internet connection and not be able to connect with customers, email etc. It simply adds insult to injury when you cannot even produce a simple document.

        We are not there yet and will not for many many years.

        I just don't get what the huge interest in cloud computing is if your not a content/software creator that is terrified of piracy. Who cares about cloud computing? Seriously, it has to be one of the worlds biggest solutions looking for a problem to solve.
    • you mean cloud9

      thout that was illegal
      • RE: I

        You appreciate, i affect <a href="">check</a> sounds great, i <a href="">here</a> would exceedingly attempt appear in approval of <a href="">here</a> otherwise <a href="">fundas para iphone 4</a> condition i was certain the accidental <a href="">here</a>
        http://www.beebond.com/apple/macbook/macbook-bags-cases | http://www.beebond.com/apple/iphone | http://www.beebond.com/led-lights/led-watch | http://www.beebond.com/es/apple/iphone/iphone-4-cases | http://www.beebond.com/es/mobile-phone-accessories
    • I agree. And no it will not.

      Do you ever wonder how many beers it takes before someone can write an article like this?

  • Time for the LED Disco Ball

    This is SO back to the '70s. And not surprising who is one of the driving forces - IBM. There are probably former employees of their VDT unit with big smiles on their faces thinking that darned PC that took away their job is finally done. And old mainframe guys saying 'told you so'.

    But you know what, someone will want to have personal control and possession of their data again and freedom to work independently and guess what will be (re)invented...

    The PC. We are good at repeating history. Nos for some Jonas (Osmond) Brothers and a dance music comeback.
    • disco ball

      ya how about bring back logans run or dynomyte or good times
    • Ya, there is the bottom line in a nut shell.


      This whole cloud computing "for the world" is a joke. We already see the endless privacy complaints about a relatively simple website like Facebook, but then we have some kooks on here suggesting that the very thing thats in our future is to have every single solitary keystroke we make on our computer traveling right through the entrails of some service provider.

      What a ridiculous joke. No thanks and no thanks from pretty much every single person who I have ever had a conversation about this with.
  • Tomorrow is good for me

    I share your vision insofar as the technology goes.

    There are several global corporations who do not:

    - media rights holders for all content
    - ISP's
    - M$$$$$$$$, A????

    Once they share the vision, and realise that the Internet means they are going to be dinosaurs unless they change their cost model and revenue expectations, then we might get somewhere.

    Until then ... C'est la guerre!
  • Great article!

    A provocative, challenging and well documented view into the future. And you may be right.

    However, if all this would come to pass, I think 20 years is the better bet. Don't underestimate the power of traditions.

    My personal hope is that, one day, *all* hardware drivers will be present in the BIOS of a motherboard. So that operating systems won't have to bother with them anymore. Of course, this would require a higher frequency of BIOS flashing by the users and of support by the motherboard makers.

    Virtualization on a much smaller and much more limited scale..... :-)
    • Not so far away...

      Sun's CMT series servers already have a hypervisor built into the firmware. These are based on multicore CPUs that allow individual cores to be isolated, so it only makes sense. Imagine your own copy of Vista available only to you, running on a server in a secure room somewhere and accessible to you anywhere there is "a Screen". You can do this today, except the screen will be called a SunRay.
      • Sounds good

        Except for the Vista bit... Yikes. :-)

        Maybe the current economical crisis will speed up this development further, because it allows for cost savings on the hardware. We live in interesting times.
  • Everything old is new again?

    This whole thin-client and cloud business is just a return to the massive centralized computing model ... which, interestingly, failed to meet the general user's needs.

    This type of centralized system is ideal for the administrator, and it has a pace in certain types of corporate environment where draconian controls are necessary, but it doesn't fit into the real world where use is driven by personal micro-projects and unstructured data.

    Why would I trust a central authority with *my* files: family pictures, music collection, medical information, or family financial data?

    Look at the security issues around financial at organizations like TJ Max and the failing grades assessed against the government's critical (and supposedly secure) systems by *their own agencies*.

    No, neither a government "Omnivac" nor an ASP's cloud is safe ... and won't be for the foreseeable future.

    Utility computing has it's place, within corps and govts perhaps, but it won't be acceptable for the general user who has (at long last) escaped the tyrannical stranglehold of the glass-room sys-admin dictators.

    Seriously, why would *anyone* want to go back to that horror-show?

    As always, this is just my $0.02 USD based on 30+ years in the business.