Friday, January 27, 2006
In the dark of the night
OmegaMom woke up at 3:00 AM. What to do, what to do? Most normal folks would be in bed, asleep. Semi-normal folk, when wakened at an odd hour, will grab a book or head to the computer to do some relatively mindless web cruising. Check out the blogs, check out the boards, check the email, read MSNBC or CNN. OmegaMom is busy looking for good articles explaining database normalization on a non-technical level and preparing her syllabus for the grandly named course, "Contemporary Developments: Database Uses in L2 Research". So, instead of searching on, say, Meg Ryan adopt China, OmegaMom is searching on "Database normalization". (She is actually searching for one good article she had bookmarked three years ago when she was in another position. Too bad she can't remember a little bit more about that article, aside from the fact that she considered it darned good; Google can't search on that, dammitall.) One site found has this lovely sentence: "Normalization is a ten-cent word database programmers use when they want to intimidate you, but the basic idea isn't really that hard to understand." The very most basic idea is to make sure your database has no repeating information. OmegaMom has been working with databases since before she joined Mountain University lo these many years ago. Mountain University is just chock-a-block full of little tiny databases and spreadsheets, much of it tracking the same information. When OmegaMom joined the SWAT team at Mountain University, her team leader had a dream: to ensure that all the little databases that oozed out of every little department in the university had the same information. Hah. Eight years later, it is still, in reality, mostly a dream. There's a perfectly good "student information system" that stores names, addresses, contact info (well, it's an okay system, the database is fine, but the front end [PeopleSoft] is a nightmare)...but there are still little departmental spreadsheets and databases that contain that same info, typed in by yet another staffer, and Kozmik All only knows whether that information is up to date or not. It's kind of like trying to hold freshly cooked spaghetti in your hand: you concentrate on pulling this one slithery strand of spaghetti back up into your hand, and another strand is slithering between your fingers. Anyway, the idea behind this class is for OmegaMom to provide guidance for folks who are getting fuddy-duddies in linguistics on how to create databases for their research, write some dynamic web pages to collect that data, write some more web pages to display the data, and then export the whole kit-and-caboodle into SAS or SPSS or some other statistical analysis tool. To prove what kind of a geek OmegaMom can be, she thinks this is all kind of fun. Hey! Wake up! Don't you know it's not polite to fall asleep when reading someone's blog?!?! Update: Oooh! Oooh! OmegaMom is doing the Snoopy Dance! She found the article she was looking for! Hey. It's the small things that make us happy, right?