We spent last weekend in Ocean City, New Jersey, with Jen’s family. Her aunt and uncle co-own a beach house on 3rd Street, so just about every year, in the fall, we get a free weekend at the beach. It’s a great time to go; there’s no one around and even the boardwalk is practically empty. I love it like that, because I hate crowds.
It might have been our last weekend there, though. There was talk of selling the beach house, or even rebuilding, which would take a long time. If they tear down the house and rebuild, they can only build two stories high, as opposed to the three they have now. But if they simply refab, they can stay at three stories. That means they can tear down everything but one wall. As long as one wall is still standing, the building hasn’t been demolished and it doesn’t count as rebuilding.
I asked what happens if you later replace that one wall. No one knew the answer.
Sometimes, though, it’s just easier to rebuild something new from scratch. I’m dealing with something similar at work. We have a program we’ve been using for years that was supposed to be able to let our teachers view student assessment data. We were just set to roll it out when we discovered that it was giving us completely bogus numbers. There was no obvious error or discernable pattern. They were just wrong.
So late last week I began building an assessment analysis system from scratch. Fortunately, most of the pieces were already in place: our PostgreSQL server, my copy of REALbasic, Corey Redlien’s brilliant ChartPart classes. All that remained was putting the pieces together, which I’ve been doing non-stop ever since, including at the beach.
As of this writing, the pieces are almost together. The bulk of the programming work is done. Most of what remains is moving our existing assessment data from a multitude of Excel spreadsheets into PostgreSQL. But when it’s done, hopefully our teachers will have a tool that will meet their needs. And when that moment comes, I’m going to take a well-deserved nap.