A few years ago, the University that I'm at created a mandate to unify all university systems under one system. This year was the first year where they started implementing this dream which has been
interesting. For simplicity, we will call this unified system the university portal.
My boss and her higher ups were invited to a series of consultation meetings; these talks were designed to discuss the possibility of linking our information systems with the university portal. At first, it was fairly exciting because there were a lot of new interesting possibilities that our system could gain by linking into the portal. We were initially onboard, and we wanted to be one of the first systems to link into the portal since our programming group is known as a fairly battle-tested programming group, and we like to be on the cutting edge.
After a few rounds of talks, something seemed fishy. Originally, they wanted to grab a data stream from our information system, and feed it into the portal. However, the specification for generating the feed was needlessly complicated and relied on a programming language that we don't use (Java). The feeds were to be embedded inside a java applet which is one of the slowest possible implementations. This is a very strange choice because not every web browser supports java applets. In addition, the standard way of exchanging feeds these days are via XML/RSS.
The representation in the consultation meetings were also really wierd. Departments that represented 150 students were trumping requests from other bigger departments. For example, our department represents 3,000 students, and we weren't given a prominent place in the portal. In addition, some very large departments weren't invited at all to these talks. This is a huge no-no because you need to gain support and buy-in from these other departments.
In either case, my boss had the amazing foresight to see that this was a doomed project, and our department withdrew from this portal.
There's a number of factors that hinted that the project was doomed:
- Their demo server kept crashing, so there's no proof of concept, and it was unstable.
- Bigger departments with great political clout was left out of consultations.
- Poor choice of technology.
- Lack of resources and technical expertise.
- Over ambitious deployment schedule.
The main thing that I've learned about software is that politics plays a huge part about how well the software is received. The people in charge of the portal don't understand the different politics at play in our organization, so they inadvertently stepped on a lot of people's toes.
I'm very thankful that we opted out of this project. We saved a lot of time, money, and headaches. The portal was supposed to launch this September which was completely unrealistic. Their project is now 8 months behind schedule. Furthermore, their project manager and lead programmer have both quit. It's only going to be delayed more and more.
Anyways, it looks like the project was spinning out of control, so one of the vice presidents have personally stepped in to take charge.
