The room selection for the gradplex was supposed to be this morning. I get up before 8, log in, and try to pick out a room: none available. I screw around for ten minutes and it still says the same thing. So, I call Residence Life and they say something’s screwed up with the system and they were on the phone with IT. Then I get an email saying the whole thing just broke and they’ll try again tomorrow.
This happened last year to the undergrads and a lot of them got screwed. Happening once is less than marginally understandable since last year was the first time it was implemented and deployed on a large scale. However, this time the IT department needs some serious butt-kicking.
The graduate housing selection wasn’t quite right last year, either. The night before it was to open up, I was just messing around and found I was able to select a room before I was supposed to.
Working for five summers as a web developer for large database type systems, big time screwups like this are inexcusable. It was no joke that we had several staging servers for deployment and testing. There were tons of tedious test scripts we had to write and perform and all kinds of roadblocks to releasing a showstopping bug into the wild. Testing was a pain, but it prevented large meltdowns like these. We also made sure a subset of users extensively tested the pre-release versions on the staging server to ensure everything worked in practice. I got a few calls about stuff that made it that far, but better have the bugs stopped there than when someone’s document repository disappears. Of course, as with most large systems, there were always minor bugs that did make it out, but they were always things that could be fixed in the next release.
Edit: I just got an email reporting the whole thing was caused by an off-by-one error. Someone really didn’t do their homework before releasing this.
Recent Comments