OnLive isn’t. As already mentioned, the cloud gaming provider and desktop service provider has ceased to be. Poor budgeting; ridiculous hardware-to-user ratios; low take-up. Quite simply – more money out than in. Ergo, failure  a simple question of finance and poor management.

Nothing to learn here, move on?

Or, can OnLive’s demise give a wider lesson to enterprises? Sure, OnLive were predominantly a games focused company. Yet, the delivery and development of games has driven a lot of technology advances that enterprises use in desktop delivery today: Microsoft’s App-V is software at the heart of desktop virtualisation and was a gaming technology back in the day. Moreover, the concept of any-device access is inherent in range of marketing material from virtual desktop vendors and service providers and also key to pushing game titles to consumers.

But for the better financial planning and an understanding of Microsoft licensing, would OnLive have succeeded? Were they doomed to failure to failure before the off?

What are the key questions you should be looking to have answered from your DaaS service provider?


It’s not about winning, it’s about taking part?

The art of gaming has led to advancements in desktop technologies and mobile devices. The wider consumer market is huge. It has a fast buy cycle.  Software innovation in gaming was arguably the foundation of application virtualisation, which in turn is arguably the key to desktop virtualisation. Arguments aside, gaming drives technology faster and higher than the enterprise ever will.

Why doesn’t VDI get a foothold in the consumer market? Because any modern game-play in it is awful.

Existing “cloud based” game providers such as Origin and Steam have successful models: so successful that they are reducing the revenues from “traditional” games retailers – even those who embrace an on-line selling presence. Why successful? For this, dear reader I went to a source. I tempted my teenage son away from his Battlefront Elder Scrolls Zombie Space Pirates IV by suggestions of unexpected household power failure if chores weren’t done. Over cleaning fluids and marigolds, I asked what were the advantages that successful games companies had? Interestingly, user expectation of games delivery has synergy with enterprise application delivery models being considered by many CTOs today:

  • offer fast and reliable delivery mechanism from a range of titles
  • delivery to variety of devices (you can install multiple PCs) with a hybrid delivery model   – in this case, services running locally (game installed local device), but with data optionally stored in the cloud (game saves, high scores) to be available between devices.
  • on-line services for collaboration (forums, multi-player game hosting)
  • expectation that all this is delivered for free, with 24-hour availability.
  • some form of  raise in the current current pay scheme would be appreciated

Final two points aside, when sitting down with him cleaning sinks and considering the service offered he had – helpdesk functions automated, signup automated, delivery automated, chat facilities to helpdesk staff within operating hours, moderation of boards, option on feedback on service delivery.. its a wish list for many organisations.

Consumerisation of IT is far more than enabling tablet use in the workspace or not having two phones: it is about delivering effective Enterprise IT as if it were a paid for service that the business is built on, relies on, operates on.

IT shouldn’t be in competition with the business – it should be with the business. The key is in customer satisfaction. What customers, your users, want is a seamless experience and for the technology to be immaterial.

Technology and process should not be  “in the way” of getting  what it is they want done.. done.

If you’re not keeping score, what is the point in playing?

Besides the finances; besides the low-take up; besides the illegal Microsoft desktop OS licensing practices – where did OnLive go wrong?

  1. Poor experience: OnLive ‘s remoting model relied on hefty bandwidth capability. This isn’t prevalent today: especially to home users. Gamers don’t like delay. Latency, lag, jitter – call it what you will – but for a gamer, an unresponsive application  really is the difference between life and death.
  2. Lack of application support: OnLive’s games failed to deliver the big titles. More importantly, their desktop service was resolutely fixed and unchangeable. “Fixed” and “Unchangeable” are the antithesis of what a virtualised desktop should be.
  3. Persistence of user data and personalisation.  My personal experience is – I moved from an iPhone to Andriod: on the iPhone I’d completed Angry Birds – on the Android, I had to start again. Not happy. Presenting a new fresh interface may well be easier and cheaper – but it alienates customers and hampers productivity.

Is the DaaS provider you are considering, the VDI service provider you are considering, offering services that are essentially having the same issues as OnLive?

Delivering desktop as a service must not be about simply delivering a VM with an OS that your users can connect to.

Some service providers hold MS licensing as issue to wider adoption of VDI and DaaS. Arguably this gives a greater incentive to develop a service that doesn’t need a Microsoft platform. Deliver a non-MS OS; deliver non-MS office products; non-MS file and collaboration services, non-MS email. Entirely possible.

And yet, in an enterprise environment there are often too many windows based apps – to remove/replace them large business exercise: contemplate desktop virtualisation can also be about business change.

Key deliverables for DaaS providers is to:

  • Allow new application support:  they’ve provided a core OS. What is the cost model and time-frame for on-boarding a new application?
  • Security and authentication: new domain and userid, or your existing one? separation from your environment or integration?
  • Access to corporate resources: moving to Win7 and Office 365 is all very well, but what about the archived documents, the last 5 years of sales data? Your CRM system, your intranet, your finance system?
  • Persistence of settings  do my users send 20 minutes a day setting up their email signatures, their printers, mapping their drives, their bookmarks: or is that automated
  • Focus on experience – consider reliance on a pure remote protocol delivery- it can be viable. What do  you do when it isn’t? When it is – how do you make sure it is consistent?

OnLive had bad financial management. Not the first tech entrepreneurial start-up to do that. Won’t be the last.

But, at any funeral there is always hope of a redeeming lifetime feature: some ray of hope that existence was not just  fleeting, it wasn’t worthless. OnLive give an example of immaturity, of how-not-to-do DaaS. Microsoft licensing incompetence aside, there is a wider lesson,  an example of not considering enterprise needs and the requirements for a successful DaaS service.

Those who cannot learn from history are doomed to repeat it: quoting  George Santayana, again.

Share this Article:

Share Button
Andrew Wood (144 Posts)

Andrew is a Director of Gilwood CS Ltd, based in the North East of England, which specialises in delivering and optimising server and application virtualisation solutions. With 12 years of experience in developing architectures that deliver server based computing implementations from small-medium size business to global enterprise solutions, his role involves examining emerging technology trends, vendor strategies, development and integration issues, and management best practices.

Connect with Andrew Wood:


Related Posts:

4 comments for “OnLive – bad management, or an example of DaaS immaturity?

  1. Mike Norman
    August 27, 2012 at 7:45 AM

    Microsoft Licensing isn’t that complicated. It would take them a few minutes talking to a distributor to work out that what they were planning wasn’t going to work. Also I think we all know if you are doing PV or VDI you don’t start by pumping high-quality video across a WAN.

    You can’t really blame Microsoft for this. Blame the person who wrote the business plan, and the investors who funded it, and above all blame the numpty the investors got in to do due dilligence.

  2. August 27, 2012 at 11:00 AM

    OnLive made a basic mistake. When as the money started running out rather than cut costs it tried to open a second revenue stream (OnLive Desktop) without understandign the first thing about DaaS. The service was dreadful, and was rightly hammered by the critics; it pulled resources – people/$$$ away from the main problem and accelerated the demise.

    A simple first time startup failure, that everyone involved, not least Perlman, should have recognised.

    Embarrassing.

  3. August 27, 2012 at 5:46 PM

    It looks like someone recognized Perlman’s mismanagement. According to this story in The Verge, http://www.theverge.com/2012/8/27/3272640/onlive-founder-steve-perlman-is-out-as-investor-gary-lauder-assumes

    Perlman is out. I hope he has better luck with his next venture.

  4. Andrew Wood
    August 28, 2012 at 2:10 AM

    Thanks for your input guys. “Crap management” it is.

    Mike, MS Licensing is indeed not rocket science (http://www.virtualizationpractice.com/licensing-vdi-for-microsoft-desktops-is-it-rocket-science-9389/). Still, despite the clear guidelines I think there is room for improvement to support desktop virtualisation. In that respect, Citrix changed their licensing policy, VMware changed theirs ..twice.. Microsoft started poorly and made some enhancements – but will they go further or stick?

Leave a Reply

Your email address will not be published. Required fields are marked *


9 + = thirteen