### I Wouldn't Know What to Do If I Weren't Sitting in Front of the Computer

- They've fixed the mail server at work so that I can no longer copy 1000+ email addresses to the clipboard, paste them into the bcc line, and easily send out a bulk email to every student enrolled in college algebra. I'm going to have to figure out how to get my database to play nicely with SMTP. Since I have no clue what I'm doing, I'm going to start by having it send messages to all seven of my email addresses before I screw up and send out hundreds and hundreds (or thousands and thousands) of incorrect things. Previous attempts at mailing things from a database have led to each of the
*n*addresses on the list getting*n*copies of the message (fortunately that time*n*was less than 20). When I've had FMP go through Mail.app, it has randomly decided to use a different one of my accounts for the "From" address than the one that I specified. - Last week I went to a WebAssign training on how to code questions. Most people at the training couldn't follow it. Coding WebAssign questions involves a mix of HTML, CSS, LaTeX, and perl. Most people at the training knew at most one of those. I've offered to run a follow-up session at a more elementary level. Of course that means that I need to figure out more about how to code WebAssign questions. In reality, I suspect that Dr. Cautionary-Tale would be a better choice to lead a session, as she already knows a lot about coding WebAssign questions AND according to my preliminary data, she's a very effective teacher. But I'm not about to ask someone to do extra work for free.
- More statistics homework to do. I have to decide whether I should use a quadratic model in
*x*(*y*=*ax*^{2}+*b*-- no linear term) or a logarithmic model in*y*(ln*y*=*ax*+*b*). Based on some things, I suspect that the quadratic model fits better. However, based on the subsequent parts of the question, it seems that a logarithmic model fits the questions being asked. It's a multi-part question, and one of the later parts seems to be leading to an interpretation that works nicely with a logarithmic model. I suppose that I will fuss with the software some more and hope for the best. - The end of the semester is surprisingly near. I need to write the calculus final. I'm thinking of having 20 points of multiple choice questions about the concepts (no calculations), roughly 40 points of basic calculations (evaluating things for 5 points each), and then 40 points of word problems-ish stuff (4 problems, 10 points each).