% fortune -ae paul murphy

Smart Displays and Productivity (2)

So if smart displays are so wonderfully empowering on the user side, what do they do to IT?

Do Unix right, and they enable you to simplify IT operations, reduce your IT costs, and practice leadership instead of management.

Notice, however, that displays like Sun Rays will deliver Windows server based applications, including Microsoft Office, just as well as they deliver Openoffice.org or other Unix applications on Unix. The difference ultimately comes down only to cost and reliability: no Microsoft licenses, and no Wintel servers, means reduced staffing, reduced data center infrastructure costs, reduced software cost, and the elimination of Wintel style security issues.

Number one on the IT hit list with smart displays is the help desk. No PCs, no help desk. Period. Application help moves to the user groups and the basis for the user/IT partnership becomes crystal clear: it's our job to make sure that things work when an authorised user accesses an authorised application for an authorised task. It's the user department's the job to understand that task, pick the application, understand how the application works, and get the job done.

Basically smart displays get rid of the multiple layers of ambiguity in applications problem solving. Everyone knows who's responsible for what, and the device either works or it doesn't: no ambiguity, no debug, no reboots, no wasted user time, and no finger pointing.

An odd side effect illustrates this nicely: in real businesses using Windows, the business units almost always have hidden IT support and hidden applications. Usually these originated as hedges against IT control or as patches for missing functions, but they tie up staff, are unuditable from a corporate perspective, lead to interesting dodges like hardware upgrades coming in as expense account items, and exist in opposition to corporate standards and practices. Switch the departmental grouping to smart displays, however, and those applications move to the data center where they can be seen, audited, and over time functionally incorporated into broader application frameworks -and the hidden IT people either go back to their real jobs, or get forced out.

Meanwhile, in the data center, the CIO will need far fewer staff - but they'll need to be experts, both at their technical jobs and at working with users. In other words, fewer people will work in IT, but they'll need extensive cross training, power to directly affect change, and the ability to work well with users.

A CIO, no matter how effective, with 25,000 desktop PCs to account for is going to have an organization that manages the organization that manages the desktops - and he's still going to get nailed every day by people complaining about down PCs, poor desktop support, and the company's absurdly ludite failure to retroactively adopt whatever magic bullet got lauded in the latest PC hyploid website or magazine.

With smart displays, all of that goes away. The desktop PC with all its hassles and pretence to user friendliness and controllability, goes away. In its place is a device that just works, that neither threatens some workers nor beguiles others into thinking they're IT experts. It belongs unambiguously to the company, and the applications are 100% available, 100% of the time.

Another problem goes away too: give your power users remote access to the system, and when somebody steals a laptop full of confidential data you'll read about it in the newspaper, not in a federal subpoena or other legal document landing on your desk.

A CIO, no matter how effective, with 25,000 desktop PCs to account for is going to spend time in every meeting with user and financial executives talking about PC security issues, the network threat, the phishing threat, the theft threat, and overall recoverability.

With smart displays, none of that exists. Security becomes an operational issue, not a strategic one. As a CIO you put staffing, server, data, network, and power redundancies in place, hold regular drills, and otherwise forget about it.

Sounds good doesn't it? but here's the killer issue: management is easy, leadership is hard.

Manage a smart display architecture and it will turn into a 1970s mainframe system in large part because management is about control, about reports, about staff growth and budgets. Leadership isn't, it's about getting the job done: but building a team is hard, giving up control is hard, helping users through the adjustment period is hard, dealing with suspicious departmental managers is hard.

And do you know what's worst of all? the better you get at doing the IT job the more invisible you'll become - so the next job you go after will go to a candidate who spends more for less; and your successor, when the time comes, will be somebody cheap and stupid because the CEO and CFO will absolutely know that IT is easy.


Paul Murphy wrote and published The Unix Guide to Defenestration. Murphy is a 25-year veteran of the I.T. consulting industry, specialising in Unix and Unix-related management issues.