Ravi Blogs about IT Topics

Many of us believe (or have believed) that home grown applications are harder to maintain, scale and support. As a result, we have gone on to purchase commercial software which to a larger extent address most of the concerns. The argument typically is that they several more resources dedicated to software development and support that is hard to match in our environment. Whereas I generally subscribe to this view (for example, it would be next to impossible for us to develop and support an ERP system such as PeopleSoft at the same overall cost), I am beginning to feel frustrated by the commercial software. This is based on some of our recent experiences.

• We began implementing the most recent PeopleSoft upgrade to version 8.9 last Fall. The approximate time between the beginning of upgrade planning and the actual rollout is 1 year. In addition, the cost of implementation is several thousand dollars. It is much higher than the actual money paid to the consultant helping us upgrade because, we need to figure in the cost of time spent on this by ITS programming staff, functional staff and the functional office staff who help us test. Oracle is asking us to prepare for some other major upgrades in the future where they are rearchitecting the back end to offer “better services”. I have argued that if they are doing this to bring efficiencies to their business, which is the truth, why should we go through the pain and pay a lot of money? Shouldn’t they be paying us for the suffering and pain?

• We recently upgraded Millennium software and it has been unstable, resulting in a couple of restarts of the web server every day. The problem was reported almost 3 weeks ago and no resolution yet. They keep sending more and more diagnostic tools , but I don’t get the feeling that they know exactly where to look. This was the result of the original company being taken over by a larger company whose expertise in Millennium may not be on par with the first company. In other words, they are still learning… and we are paying for it. And by the way, once you have upgraded, there is no easy way to roll back!
• PowerFaids, the software used by financial aid office was recently upgraded and some critical reports were not working. And I believe they are still not working…
• Last year we went through a Blackboard upgrade. Same story… took a couple of months and a bug in the installation script took a while to resolve. Again, a fair amount of resources went into this.

I understand that these are fairly large and complex systems and upgrades can be nontrivial. However, I cannot believe that with the constant advances in tools and methodologies, these companies cannot develop a paradigm by which the pain and sufferings from upgrades can be minimized.

Many of the operating systems and desktop software do upgrades seamlessly these days. I would say operating systems are pretty complex, and they have figured this out…

The problem here is that once we commit to using one of these software packages, they know we are in it for the long haul. May be there is no real incentive to make the upgrades simpler. Obviously we are not going to dump PeopleSoft and move to something else pretty easily. I am not sure there is much that can be done unfortunately, except to feel frustrated. The standard argument that you are the consumer, so you have the power to go somewhere else simply does not apply here because it is not practical…
Above all, we all (including our end users) seem to take it in stride… it is part of the deal.

There is an inherent double standard by which we seem to judge our own home grown systems relative to these commercial software. I doubt very much that we will accept a home grown system to take a whole year and tens of thousands of dollars to roll out a new version…. may be I am wrong about that…