My Friday Rant -- Moving Beyond VDI?

It's Friday.  I'm back from an exceptionally long holiday.  Thanks, it's great to be back.

Today's rant was sparked by a great conversation on Quora: what is holding back business adoption of VDI?  Lots of useful perspectives -- all worth reading -- but I felt compelled to add some additional thoughts.

On one hand, enterprise adoption of desktop virtualization is probably more widespread than the poster might realize.  On the other hand, I would guess that the vast majority of work-related user experiences today are most decidedly un-virtualized.

Part of the problem, I believe, lies in the name: it can send people down a pathway that may or may not be what they're looking for.  

And, no, I don't have a better name.

Names Are Powerful Things

We, as human beings, think mostly in symbols, concepts and emotions.  Give something a name, and it frequently invokes all three.

If we deconstruct the term VDI -- virtual desktop interface -- it invokes the idea of recreating our familiar windows desktop experience on any suitable device of our choosing.

Nice -- but what if I don't *like* my familiar windows desktop experience?

Stepping Back A Bit

The first round of VDI implementations tended to focus on cost reduction.  The target was what I refer to as "task workers" -- people who used their desktop in a largely repetitive manner to assist in a fairly proscribed set of tasks.

Call center workers are the most familiar example, but there are many others.

Rather than build a web-friendly app from the ground up, many organizations took their familiar desktop applications and simply made them usable on a thinner client.  Many thousands examples are in production today -- they're not hard to find.

But what if you're more of a knowledge worker, and less of a task worker?  How do you respond to the concept of VDI as frequently articulated today by IT people?

I'm guessing you're holding out for something better :-)

The Growing Importance Of Knowledge Workers

If you're a frequent reader of this blog, you'll probably remember me ranting about how knowledge workers -- of all varieties -- are increasingly creating a disproportionate amount of value in most organizations.  

Going further, why it makes strategic sense for many IT organizations to devote a disproportionate amount of resource and attention to their growing needs.

And, for some important classes of knowledge workers, simply recreating (or virtualizing) the familiar desktop / WIMP paradigm might not be the best path going forward.

The Three C's For Knowlege Workers: Content, Consumption and Collaboration

For the last five years or so, I've been a passionate armchair observer of knowledge workers -- watching very closely how they do what they do.  The patterns are now familiar, at least to me.

Many important knowledge workers create a lot of content -- documents, software, music, video, etc.  And it's hard to argue with the inherent value of "desktops" -- big screens, full keyboards, etc. -- in the process of creating rich content.   Don't ask me to write a long document on my iPad, please.

For this class of knowledge worker activity -- creating content -- virtualizing their full desktop experience, and making it easily re-creatable on any suitable device (and in any location) makes a whole lot of sense.  For this use case, VDI wins big.

However, if you study most knowledge workers, you'll find that they often spend an enormous amount of time consuming content created by others, and -- more recently -- collaborating as well.  

Do I want or need a full, rich desktop experience to consume and collaborate?  For most people, the answer is no.

By now, a critical mass of knowledge workers have come to appreciate the sheer power of mobile devices in making information consumption and collaboration far more convenient and useful than was ever possible using traditional desktops and laptops.  

Read and comment -- anywhere, anytime.  Tap, tap, tap.

These mobile knowledge workers who do a lot of consumption and collaboration aren't thinking desktops -- they're thinking iPads and Android phones and all of that.  

Calling your IT project "virtual desktop" might miss the point of what these knowledge workers are after.  They're probably not excited about the idea of lugging around a laptop and booting everytime they want to do something.  Or struggling through a traditional windows interface on a smartphone screen.  

Don't expect this crew to be passionate endorsers of the proposed approach.

I don't know if there's a good industry term for a mobile, managed knowledge-worker experience that focuses on content consumption and social collaboration.  Whatever that ideal term might be, VDI doesn't do a good job of describing what these people are after.

Terms And Targets

I have been involved in my fair share of VDI discussions with IT groups over the year.  I now always start the conversation -- almost abrubtly -- with a very pointed question: what exactly are you trying to accomplish?  

About two-thirds lead with "we're trying to deliver desktop applications more efficiently to our users" -- lower cost, less effort, etc.  If their user population is made up of task workers, or knowledge workers that prefer a rich desktop experience -- well, then, let's call it VDI, and get on with it.

EMC is very good at both small-scale and very-large-scale VDI projects these days: technology, process, solutions, consulting, etc.  It's become very familiar ground for us and ourpartners.

But about a third of these conversations take a decidedly different direction: they describe their most important knowledge workers, and how they want to deliver newer forms of applications and information services with a clear focus on maximizing their value and productivity.

It's hard for me to use the term "VDI" in most of those conversations.  We end up making up terms like "mobile knowledge worker experience" or something similar.  

These people are after value generation, and not simply cost savings.

The Apps Are Becoming Very Different

A corresponding observation is the 180-degree turn I've noticed in application development thinking, especially when aimed at this knowledge worker crew.  Frankly, I'm blaming the iPad for this dramatic shift of thinking.

Old school: massive portals and sophisticated app environments that do everything you could ever want, but require substantial training and repeated use before proficiency is achieved.

New school: the familiar"there an app for that", smaller function-specific applications that largely emulate the iPad experience -- no training, no manual, get-in-and-get-out-quickly applets.

Reconstructing End User Computing Service Delivery

There's a bigger, structural issue here as well.

For the last few decades, "end user computing" largely involved buying people desktops and laptops, loading them up with packaged applications, and supporting them when things broke.

Now we're transitioning to a world where users are comfortable selecting (and frequently paying for!) their own devices, where powerful personal productivity apps are cheap and plentiful, where more time is likely spent outside the firewall than inside, and self-help is becoming the preferred support model.

Which begs the question -- how many IT organizations are prepared for this new world?

Back To The Question

So, why is it that classical desktop virtualization hasn't taken off to the degree as some of us might have thought just a few years ago?

I would offer the following personal opinions:

(1) Many knowledge worker populations don't see enough value in simply recreating virtual editions of their existing large desktops and rich application experiences.  Instead, they're lusting after tablets, smartphones and function-specific applets -- as am I.

(2) Many IT end user computing organizations are structured for the old world, and aren't resourced to transition to the new one.  Running lean-and-mean results in no bandwidth to try new things.

(3) Without a critical mass of newer, lighter applets that work the way these newer devices do, much of the value and appeal will be lost.  And, if you develop a lot of internal applications,  it's the uncommon IT app dev team that has fully embraced the new applet paradigm for their knowledge workers.

(4) Unfortunately, both of these transitions (device and app) have to happen in a somewhat coordinated fashion -- and require substantial resources -- otherwise the value won't be there for many.

Traditional VDI projects still have their place -- there are gazillions of task workers out there, and many knowledge workers who rely on rich desktop experiences that could benefit from virtualization.

But for the growing cadre of increasingly important mobile knowledge workers who primarily consume and collaborate -- I'm guessing we'll need to come up with a richer -- and more challenging -- conceptual framework.

Comments

[Stuart Miniman]

Chuck,
We've seen a lot of technologies that have taken a while to mature which need new names - we had the same discussion moving from "next generation" or "virtual" data centers to private cloud or the similarities between ASPs in the 90's vs the xSPs today.
Simply replacing an old Windows desktop with a virtual Windows desktop didn't seem compelling to many while the opportunity to truly mobilize someone to have flexibility between various work options is attractive to the knowledge worker. To help the end-user community understand the opportunities and challenges better, the Wikibon community is holding a Peer Incite on January 25th - details at http://wikibon.org/blog/desktop-virtualization-reality-check/
Thanks,
Stu

[Sahaynes]

I'm in agreement with Stu.

A funny thing happened on the long trek to a feasable VDI solution. Someone forgot to define the V and the D.

Those definitions are what a lot of midsized business decision makers are waiting on. We have been sitting on aging desktop hardware through a rough economy but now that it is time to make a move it seems the desktop we were planning on virtualizing has evolved to something new. For so long corporate use determined what home users installed on their PCs. That time has passed. Users like the way they are able to consume and use apps on demand with their personal laptops, tablets, and phones regardless of the OS. They want and expect the same from their corporate devices.

So before recommending what top level management will see as a bold change to our desktop strategy, I want to see how the V and D get defined. Five years ago the V just meant we were going to simulate the physical environment. Today's definition is quickly becoming "cloudy." The decision to simulate the old standard workstation environment will be driven by your definition of D. Why is the definition of D even tied to the OS anymore? Every purpose of the OS was to allow for the consumption of applications, computing power and storage in a way that was meaningful for business. At the rate at which app stores are popping up and application vendors are moving to ASP models will I even need to worry about the desktop OS layer in five years?

So there is the gist of the problem that I see with companies adopting a new VDI strategy. IT people see the benefits of a VDI solution right now. We can even work the ROI to make it look good to upper management. The problem is that we just aren't sure that the D in VDI is worth virtualizing if we can find a better way to provide for business needs and better mobilize our users.
Thanks,
Stephen

[Blaine Berger]

There are actually some basic reasons why VDI isn't the slam dunk that many would like.

1. Costs. The math needs to result in a real reduction in the cost per seat WITHOUT giving up the user experience and taking into account the cost-shift to the more expensive resources (like storage) in the datacenter. Reduce costs in an accelerated time-frame before the deployment rots like a banana.

2. Windows XP migration to what? Enterprises are still wrestling with their plan on this. If the math was favorable to deploying VDI for Windows 7 for example by keeping the desktop gear booting XP, you'd have a compelling reason to deploy VDI.

Your points about the extraordinary knowledge worker are well taken. However, these workers are by now used to routing around the IT departments and their leaders who decide what gets official support.

In the defense of IT, they are pretty much consumed with the hard, boring work. http://e-oasis.com/alerts/2011/boring/

The more likely outcome is that as these workers increasingly show up with their mobile devices they will find the applets on their own that provide the bridge in the corporate environment.

Haven't you found a few using your iPad at EMC?

That bridge won't come from VDI in our lifetime, but that doesn't mean that VDI is obsolete.

Blaine

[Brian Gracely]

Chuck,

Thanks for picking up on my final point about frustrations with VDI and looking at a new approach in 2011 and beyond, http://www.cloudsofchange.com/2010/12/musings-on-2010-and-predictions-for.html. I think the reality is that the desktop has reached a point for a growing number of users where they really don't need anything more from IT than an Internet connection (wired or wireless). They can figure the rest out themselves via apps, public cloud services, and self-service assistance (called a friend, ask a community, etc.).

Here's a simple example that I'd love to know if most IT organizations would beat in terms of overall price, level of service and productivity:

- MacBookAir 11" - $999
- iPad - $499
- DropBox Pro - $480 (2yr, 100Gb service)
- MiFi - $1250 (device + 2yr, $50/mo service)
- Windows license + VMware fusion = $150

= Awesome environment for a mobile, knowledge workers that can support every type of scenario.

= $135/mo over a 2yr period and I'd promise NEVER to call desktop support for anything. In fact, I would opt-out of being able to call them. I'd encourage them to taunt me and laugh if I did actually call them.

The only service where I think IT-Deskgroups would add value in that scenario is a simple Directory-lookup service ("app") if I want to add groups to one of the services.

Would some people argue that data is then exposed outside the firewall? Maybe. But that's both an opportunity (help SPs build more secure services, etc.) and a risk-vs-cost discussion and shouldn't just be a "but it's outside the firewall, sound the alarm!!" discussion. Sometimes disruption to old ways of thinking is good.

The desktop group within IT might need to realize that the kids are growing up and are ready to leave the house. It'll be sad at first when they are empty-nesters, maintaining the NAS boxes for those task workers, but I'm sure they'll find new hobbies and things to do with themselves eventually.

[Chuck Hollis]

Blaine

I think EMC itself is a good example of what can happen.

Our company is chock-full of mobile knowledge workers who tend to be technically savvy, there are iPads everywhere these days, which will be undoubtedly followed by a wave of Android tablets in 2011 and $deity knows how many other flavors of devices in the future.

Watching how people use these tablets in a work environment is fascinating -- they're frequently consuming, commenting and collaborating -- in near real-time during almost every waking hour -- in a way that was never seen on smartphones.

Sure, they'll occasionally use a traditional desktop when it makes sense, but -- otherwise -- why bother? If we had good tools and frameworks to measure the productivity-increase-per-mobile-knowledge-worker, I'm sure we'd find eye-popping results.

Thankfully, our IT group let this transition happen naturally and organically. People can opt for the more traditional desktop IT option, or do their own thing with any variety of devices. Self-help is becoming the norm here (thanks to our internal forums). Cool apps and integrations are passed around and shared so it's not hard to always have the best SW on your device.

It's a great model -- plays well with Brian Gracely's comments as well.

-- Chuck