The first half of yesterday was devoted to loading the 2010 raw weather data into databases that Will and I could use on both my own computer and the second lab computer. I spent a lot of time playing around with the rawdata table and getting a feel for isolating and displaying different chunks of interesting data – all data for a certain day, times when the temperature was below -20, times when the daily rainfall exceeded 4 inches, etc. The rest of the day was spent trying to pursue the questions of what exactly a relational database is, what’s so great about it, and whether our database should be one. Everything I’ve read so far has described the relational model in pretty arcane language, so I’m still not entirely sure what the bottom line is, but the general idea seems to be that structuring the database with particular connections between fields of data in different tables makes the long-term organization and manipulation of the database more easily understandable. For example, in theory, with a relational database we could delete the faulty data for a specific date in the monthly data table and the database would automatically delete that date’s data in the yearly data table as well. The tricky part would be setting up the relations correctly, and I’m not sure how much that’s worth it for the types of tables we maintain and the ways we operate on them. Also, I would guess there’s already perfectly fine code in place for handling situations like that, and if not I think it would be pretty easy to do without adhering to the relations rules. I am going to continue to research the relational model, and hopefully I will have a better sense for its relevance to us as I become better acquainted with our database.
-Katherine