“Time may change me, but I can't trace time." - David Bowie from "Changes"

Recently, a younger colleague who has been reading this blog asked if I’d consider writing about the changes I’ve seen during my career in IT. He was particularly interested in the concepts that have remained relevant across the years and in how current cloud computing models are similar to the service bureau environment where I started. I hadn't thought about this topic for 100 Mondays, but it's a good one! I also admit that thinking of 100 things to write about is going to be a challenge and I appreciate the suggestion. Thanks, Lars!

My interest in computers started in high school and my first computer programs were recorded on punched paper tape and stored with rubber bands around them in a mason jar on my shelf. For some reason, I didn't actually consider a career in IT and graduated with a degree in psychology after starting out majoring in enology.  I do remember hanging out in the computer lab at university but I never actually took any computer science or IT classes before I graduated.

I began my career in IT when I needed to make more money as I considered pursuing my Master's degree. I answered an ad in the paper and ended up as a computer operator on an IBM 370 mainframe working with punch cards and green-screen terminals. I taught myself some assembler language programming and then took a class from a colleague and got my next job as an assembly language programmer in the banking industry.

From there, I continued working in IT as a mainframe systems programmer, in storage, capacity planning, quality assurance, Unix systems administration and other technical pursuits. I was hired by the university to help them migrate to an IBM 3090 mainframe and ultimately moved into management. I've managed systems programmers and systems administrators, the help desk, networking, and ultimately took on my current role as Executive Director for Technology Services. We've just hired the first CIO in our university's history and I'm very pleased with this change and looking forward to helping with the transition to get the greatest value from having IT represented on the president's cabinet.

In the course of my career, I’ve worked with many systems including IBM mainframes, Prime, PDP and VAX minicomputers, and IBM, Apple, NeXT, Sun, Silicon Graphics and Dell microcomputers. These used a wide range of operating systems, storage, memory, database, network, printing and display technologies. I’ve worked with various virtualization environments from VM/370 to z/VM, VMware and others. I’ve watched the introduction of personal computers and seen the software they use evolve from offerings like CP/M, VisiCalc, CompuServ and Mosaic to modern operating systems running programs like Excel connected by the Internet and the modern web providing social media sites like Facebook and Twitter, and cloud computing including Google offerings for just about anything. Along the way, I’ve seen the emergence of Unix, Linux and technologies that allow up to 8,000 virtual servers to run on a single IBM z13 mainframe.

Some of my colleagues will remember wiring boards for the IBM 402, mounting tapes and disk packs and not only TSO, but WYLBUR and Interact, too. For the rest of you, my lapse into nostalgic jargon above isn’t really very important. I hope you can find it more amusing than annoying!

What  is more significant is what has remained relevant across all these changes. From time-sharing systems running on mainframes, through distributed client/server computing, to the current cloud-based offerings what has remained important are things like reliability, availability, performance, flexibility, and security.  While I may find cloud computing eerily similar to the service-bureau environment I started in, with the data in the data center and the user accessing it from a terminal, IT is still mainly about people and relationships. What matters is how well we can communicate our requirements and work together to deliver value by making information out of data.

What drives people to adopt new approaches to IT, and abandon old ones, remains the same, too. We flee higher costs for the promise of cheaper, more powerful, computing. We gravitate to new solutions that provide more useful ways of analyzing, and visualizing, information. We forsake the limitations of old paradigms for the promise of innovative ways to do our work. We embrace revolutionary new capabilities that allow us to tackle problems that were beyond our ability until now.

Over and over again, we will choose to build our own solutions when waiting for others to meet our requirements takes too long. Then we will surrender that independence, or have it taken from us, in exchange for the greater efficiency of a more standard approach. Sometimes, we will get caught up in trends and leave valuable tools behind in pursuit of shiny, unrealized, promises. In March of 1991, Stewart Alsop, the editor of InfoWorld and a very insightful person, wrote, "I predict that the last mainframe will be unplugged on 15 March 1996." Sometimes, we will be wrong.

The mainframes are still here, processing the vast majority of our financial transactions and playing a key role in most major industries. Distributed computing is still here, helping us achieve the flexibility we need to meet our wide range of computing needs. Cloud computing is here now, too, offering a new way to access a wide range of computing options in a simpler way. Behind the clouds shine cheap commodity computers, super computers and mainframe power and reliability. We watch in hope that the promise of efficiency, flexibility, and power, the dream of a real computing utility, is about to be realized.

If the promise does come true, I’m confident that thinking about computing the way we do at the enterprise level will be fundamental to making that happen. The mainframe mentality that encouraged us to think about computing environments as requiring very high reliability, availability, and serviceability (RAS) is still relevant today. After a prolonged outage of one of our distributed servers lately, I have an even keener appreciation for those mainframe systems I worked with 35 years ago! High performance and capacity are needed but the ability to assure that the right capacity and performance is available for each request, so that we optimize the use of scarce resources and deliver reliable service, is fundamental. I’m grateful to my early mentors and to SHARE for teaching me to think this way.

Whether we are using a utility computing environment for the serious business of commerce, healthcare and government, or to play, connect with each other, and share our passions across the face of the planet, a service we can count on to the point that most will take it for granted will be necessary. I look forward to watching that service continue to emerge and evolve.