Nick "IT doesn't matter" Carr's latest post on why the business PC doesn't matter (those are my words, his were "Rethinking the business PC") resonated with me on so many levels that it could easily be turned into a manager's handbook on business agility. In my RSS reader (Google Reader), his headline was ironically juxtaposed against the news from eWeek that Toshiba was launching some new enterprise notebooks.
When I first saw the headline, I wrongly assumed that this would be a write up of how, technologically speaking, attempting to distinguish between a consumer PC (a keyboard, a display, a hard drive, ports, built-in audio) and a business PC (a keyboard, a display, a hard drive, ports, built-in audio) is futile. It's sort of like distinguishing between a Chevy Suburban, a GMC Yukon, and a Cadillac Escalade. Same platform. Different appointments. In fact, one system distinction that I often laugh at is "workstation" (never mind that today's average Joe desktop can run circles around yesterday's so-called workstations).
But that's not what Carr wrote about. Wrote Carr:
...the digital divide between business devices and personal devices has always been more theoretical then real. People do personal stuff on their corporate PCs, and they do business stuff on their home PCs. In the last couple of years, the divide has begun to blur completely, thanks to the ubiquity of the internet, the proliferation of networked computing devices like smartphones, and the increasing delivery of software applications over the web. The time may have arrived when personal and business computing merge, at the device level, and companies get out of the business of buying PCs and other end-user devices altogether.
He goes on to mention the potential role of virtualization in cutting the so-called business PC down to size (worth two or three chapters in the aforementioned handbook) and refers to a story about how consulting outfit Oliver Wyman Delta has started a program whereby employees bring their own PCs to work (backed up by a reference to some Gartner research that says this idea is actually a trend).
Although I'm guessing at all the things Carr was thinking when he wrote the post, it was pretty high level and made me wonder what the conversation would be like if some CEO asked the CIO for a quick interpretation. With apologies to Carr (since he may not have meant these things at all), here's what what I would have said:
- Today, more than ever before, our constituencies are already interacting with our systems and to facilitate this, we've moved to more of a Web-delivered and services-oriented approach. We did this more out of necessity than anything else. After all we don't buy them PCs to facilitate their interaction with our systems nor do we presume to tell them that they can't access our systems unless they have one kind of device. They're coming in from all sorts of devices (desktops, notebooks, BlackBerries, Treos, etc.). So, by allowing all but our most important constituency (the staff) that sort of flexibility, we'd be sending our employees the message that they're chopped liver compared to everyone else.
- Dovetailing that last comment, one benefit of Web-delivered (and in some cases services-oriented application delivery as well) is, at the very least, client independence. We already know that when users don't feel as though their needs are being met by the IT department, they take matters into their own hands and there's often very little we can do about it. By moving to a Web architecture for our applications, very few if any of today's or tomorrow's client side devices (desktops, notebooks, smartphones, etc.) are wholly excluded from the party. Some may have an easier time accessing information than others. But the efficacy of allowing end-users more choice outweighs the compromises that they're usually savvy enough to consider anyway.
- Where security is a real concern and we still feel the need to have more control over the means of access to our infrastructure, today's virtualization architectures make it possible for us to deliver (even over the network) the entire PC (hardware and all) that we might normally hand to employees as software. In other words, their "business PC" (the one that has access to our infrastructure), runs securely in a separate window that's partitioned from everything else on their PC. This approach shouldn't impinge on end-user choice of platforms and, to the extent that we deliver this experience over the network, should allow for access from mobile platforms like smartphones when users are in a pinch (today, this is clunky, but it works). For users that don't want to provide their own technology, we can furnish them with a network terminal whose life expectancy and total cost of ownership makes a normal PC look like an Edsel.
Carr ends by saying "Computing is being freed from the computer." On this point, he either mispoke (and meant "from the PC") or I disagree. Maybe it's splitting hairs, but the the term computer has always been somewhat fungible. Somewhere out there is a thesis I read in 1992 (written by whom, I can't remember) that, speaking of rethinking the PC, made me rethink the computer. It was definitely an aha moment. The basic idea was that all computers have a bus (the conduit through which the data passes from one component to another) and that when you connect a computer to a network, the network simply becomes an extension of that bus. Taken one step further, if there's another computer on the other end of that bus and the two are collaborating on a task, can you really say where, in the context of computing one computer ends and the other begins?
Back in the 70's and 80's, users would sit down at IBM 3270 terminals with no intelligence in them and call them computers. But the actual computer was on the other end of a network (good ole' SNA!). Call it "coming full circle" or whatever you like. Until there's a USB or Firewire port next to the audio input port on the sides of our heads, computing will never be freed from the computer.