IN THE BEGINNING there was the mainframe, and project managers heralded a new age of computing. In each new wave of computing advances, project management software was always one of the first types of applications to appear. It's no surprise. The first waves of project management software were completely analytical in nature, and the advantage of putting these complex calculations into a computer was an obvious time savings.
PM Systems Were Originally Very Centralized. Project management was there when centralized mainframes were fed by punch cards. Some of you may be too young to remember punch cards. Imagine a card about 3 X 7 inches with a bunch of holes punched in it. Each row of holes represented a character. After the cards were punched, you'd stack them in order and bring them to a technician (read project management support staff) who would feed them into a big machine and tell you that your critical path analysis would be available on 11 X 17 inch paper in about two days. Woe to the hapless project manager who dropped his cards.
When the “mini” computer came around we could interact with the computer directly, using our own terminal in a complete and almost instant fashion. The interface was still user-hostile, of course, but with “minis” came the first notions of graphical output. The allnew plotters that appeared in the ’70s brought us bar charts, curves, and logic diagrams, all in living color and none of which had to be drawn by a staff of draftsmen. Still, the systems were so complex that only highly trained users could access them. The result was that the project management team remained a highly skilled, highly collaborative group. They were almost always co-located, almost always had a similar kind of training, and almost certainly had a similar level of training. They were not well understood throughout the organization, but they carried tremendous power nonetheless.
The PC Brought a Distributed Environment. The personal computer brought us to what I've often referred to as “the great diaspora”—the almost complete decentralization of the computing environment. With the power of a mainframe individuals could now use a computer on their own desktops, working in a comparatively user-friendly environment to create systems and manage data and calculations on their own terms.
Chris Vandersluis ([email protected]) is president and co-founder of HMS Software, based in Montreal, Canada. He is a member of PMI and the American Association of Cost Engineers. He has appeared in Fortune and Heavy Construction News, and is a regular columnist for Computing Canada. Comments on this column should be directed to [email protected].
Yes, project management software was there; the first systems were released the same year the PC was shipped, in many companies becoming the first corporate system to be installed on a PC. The result of this distributed computing power being available on desktops was a type of anarchy. Every individual, every department, every group now had the power to go their own way. Systems were cheaper and faster to implement. Friendlier interfaces resulted in less training required, which served as incentive to get a smaller system going for smaller levels of requirements.
As great as that was for the individual, the long-term effects are still being felt today. The decentralization of the project management function left many organizations with no way to bring the project management data back together for reporting to management. Individuals were empowered, to be sure, but management was not.
Networking Offered the Chance to Re-Centralize. The networking in the late ’80s offered the chance to bring system data back together by being able to share it. But the hoped-for effect has been slow in coming, at least in the project management world. Having had a taste of freedom, project managers were reluctant to let it go by having networked data re-centralized. Networking seemed to offer “you can share,” but often the effect was “you share your data with me and maybe I'll share my data with you.”
The more enterprise-oriented systems of the late ’90s had to battle an even greater interest in easy-to-use desktop systems such as Microsoft Project, which quickly became the most pervasive system used by individual project managers around the world. With MS Project taking up 70–80 percent of the project management product market, it was tough to generate growth in the recentralization efforts of an enterprise system.
There is, of course, a trade-off for bringing data together. Management is all for the idea because having data in one location and in one format makes it possible to do analysis and reporting on data across all projects. Historical data can be analyzed to refine future estimates and to prepare for contingencies. Project managers, however, will experience a loss of control. Having data in one format implies a common set of assumptions, standards, and a process that everyone who enters data into the system must adhere to. Well, there's not much argument about that, except that any 10 project managers in a room will have 11 opinions about how that process should be determined. Unless project managers believe there will be a worthwhile return on investment, the effort and perceived loss of control will more than offset the desire to see all project management data in one place.
Client/Server and the Internet Are Bringing Us Back Together. The changes of the last two years have made for a new wave of systems and for the first time in 20 years we may be seeing a movement of project management data back to the enterprise.
It is the Internet that is having the most profound affect on project management systems. Just as with every other wave of computing, the meteoric rise in popularity of the Internet has brought with it the early adopters of project management software. If a PM system isn't Web-enabled, it's either going to be in the next few months or it won't be around too much longer. There's an interesting side effect with all the systems we'd like to be Web-enabled: they mostly need to be centrally maintained.
The combination of client/server database technology and Internet-based browser-enabled interfaces has found great favor with end users. The interfaces are even easier to use than the Windows interfaces of only a few years ago. They require less training and almost all have been written from the ground up to take advantage of the inter-networked nature of the Internet for more communicative, more collaborative functionality.
The architecture of most of these systems is referred to as “n-tier”—which means that the data, the calculations, functionality, and the interface can all reside on different computers. The interface is typically the only part that needs to reside on yours. The implication, however, is that the database resides somewhere central, and that is a big seller for management. For the first time in a generation, management is receiving requests from the PM community for a system that has data residing centrally on the corporate database.
Look for New technology at PMI's Symposium in Houston. Other technology advances are having their effects. No doubt you'll be hearing lots about them at this year's Annual Seminars & Symposium in Houston. Application Service Providers (ASPs) are keen to put project management tools on their own servers and then let you subscribe to them for a user fee. (Sounds like the time-sharing of days gone by.) The advantage is partly financial (no purchase required) and partly logistical (no setup, just log in over the Internet). Whether project managers will adopt this structure in huge numbers is still uncertain. There is no doubt, however, that there are big bets being placed in the market on the success of the ASP model.
Personal Digital Assistants such as the Palm Pilot have been around for a while, but their newly connected nature is allowing such systems to serve as extensions to central enterprise systems. With the recent release of Java for the Palm Pilot, look for some new networked applications for project management this year.
Once data is centralized, I am sure you'll be seeing a new wave of functionality from two technology areas that are receiving a lot of press lately.
The first technology area is data mining. Data mining has been experiencing some growth partly, I think, because of the huge numbers of ERP implementations that have resulted in large elements of an organization's systems being stored in only one or a small number of data formats. Data mining provides functionality for sifting through massive volumes of data to find patterns and areas of data that can be used in reporting and analysis.
The second area is neural networking, also known as artificial intelligence. A number of software companies have tried to bring “thinking” to the project management realm, without success. They may want to take another try. Having data available in one format over a long period of time and/or over multiple projects is a poster-child environment for AI systems.
OVER THE NEXT COUPLE OFYEARS, you'll find more and more of your project management systems data coming back to a centralized location. Now is a good time to be thinking about how you will take advantage of it. ■
Reader Service Number 080
September 2000 PM Network