Editor, the Gauntlet,
[Re: "If it ain't broke, don't fix it," Ændrew Rininsland, April 5, 2007.]
The article asks "How much did this cost? What ... has been gained?" The answer is that it's saving the university a great deal of money in terms of maintenance costs. InfoNet did not use a database, it worked off a flat file. And it didn't use sessions, because it tapped into the ancient dial-in phone service (that's why it begged you to log off when you were done). That also explains the cap on the number of users it can handle at any given time and the huge downtime for maintenance. Schedule data could not be changed when any users were online. But according to Mr. Rininsland, "functionality has actually decreased by several goddamned orders of magnitude." Personally, I hated getting kicked off or being told that I would only be able to log in at a specific time.
Again, I'm not saying I like PeopleSoft. I hate it, I think it's terribly designed from a usability standpoint. But I fully understand the reasons for the change, and Mr. Rininsland, finding these sorts of facts out before writing an article about it is what journalism is supposed to be about.
Anonymous Coward
Comments
I admittedly didn't know about how the InfoNet was structured, but I don't believe that changes the point of my article. Regardless of the maintenance costs it takes to keep such a system running, the opportunity cost due to bugs and having to learn such an arcane piece of technology likely offset any savings received from the upgrade.
Secondly, we have dozens of world-class software engineers on campus; does it really make more sense for the entire system to be replaced by such an inflexible and expensive piece of software? It would probably have been far cheaper and more effective to hire a group of them and have them recode the backend. At least the interface would be the same then.
-∆ndrew.