I mean to update the site with a ton of pictures of Halloween, specifically our jack-o-lantern and Abby in her Lion outfit. There is also a movie from her school parade that Berta took. Unfortunately, you'll have to wait for those until they're all prepared. Don't worry, they're coming soon.
In the meantime, let me tell you what I've been up to over the weekend.
Saturday, we played D&D at Larry and Ken's place. Dave ran an adventure with a whole lot of barbarians. We survived pretty reasonably with my character (Grogor, a 5th-level Half-Orc Cleric) defaulting to Cure spells almost continuously.
I have some issues concerning my character and the game on Saturday. First off, I would like to point out that there is no need for a cleric to choose a specific patron. My cleric is friendly to all gods. He pays more lip service to Tymora, since that is how I have selected his domains, but he's an all-around worshipper. There is no need to leave anyone out.
And here is a question: If you're casting a spell with a somantic component (that means you have to say the spell aloud), does that mean you have to call down the heavens with a booming voice? I don't think so. Sometimes it's all about prayer. Quiet, solemn player.
Anyway, he made 6th level, although we really didn't finish the adventure. This is not good becuase it's our pick-up game, and we're supposed to finish the whole thing in a night. The idea is that next time we might not have 4 players, and then it might be difficult to fend off 8 raging barbarians with only a cleric and a rogue.
Anyway, we weren't late on Saturday, since Larry went to bed and Ken had to work. I saw Ken's pictures from the wolf reserve, which were pretty cool. We also saw some day-late college Halloweeners going to a party across the street. That's a busy neighborhood for underage college parties.
I did some work over the weekend on SpamCatcher. SpamCatcher is this transport filter that runs on the email server here. It uses a bayesian analysis to assign weights to each message as it arrives. It is able to classify a message as spam in terms of percent. So a message could be reported as 60% spam content, and then filtered if the spam content is too high.
Anyhow, I finally did some debugging and got all of the common glitches out of the system. Every message that comes through is now being filtered as expected. It's doing a pretty good job for the small word set it's using (about 14000 words), but the database is getting a little large to be properly maintained.
I have two tasks that I would like to perform to update SpamCatcher. First, I would like to switch the database over to something that works faster. Something that uses trees would be nice, since that'll make the word search go much more quickly.
Second, I want to remove all references to the Indy message component. The Indy message component is manging some messages that have attachments. I think I should be able to scan and report in two separate steps instead of loading the whole message into the component and then writing it back out with the component. I'm not sure what those Indy guys were thinking when they hardcoded some of that stuff in there (it adds two custom headers to every message no mater what).
I did some work on the anniversary edition of Asymptomatic. I've been doing some things behind the scene so that at the end of the month when the anniversary arrives, I'll be able to throw the whole thing online. It'll be four years at the end of the month.
I showed Berta the preliminary design for the new site and she approved. I really didn't want to show it to anyone until it was all done, but I couldn't resist getting a little feedback.
This week, Nana is down south visiting uncle Carter, so Berta has taken off a few days to watch Abby at home. We're going to have to figure out how she can get out of the house on her own for a little bit so that Abby's not driving her nuts all week.
I have a test in my programming class tomorrow, as well as an assignment due. I will probably work on the assignment at the last minute. It's basically a program that reads a file with a bunch of days and miles, and then computes which of two rental plans will yield the best price. This is a revision of the last program we handed in, which did it for only one trip. Tomorrow's program computes the values for several trips, then computes the total savings.
I keep thinking that I should loan myself out to some of my classmates who are having a little trouble in this class. Some of the problems they are having seem pretty easily fixed if they just had someone looking over their shoulders. I should also post my completed programs online here as a reference.
Tomorrow is election day. I wonder who the candidates are to vote for in this off year. I think that my voting strategy will be to vote for people only whom I have not heard of. This strategy works because everyone whom I have heard of is a nincompoop.
I got a letter from our current borough president the other day. He needs to have a proofreader look at the stuff he's sending out to try to get re-elected. There are many things wrong with this whole borough council thing. While there are huge flooding issues in town, we're spending money to put a metal roof on the park pagoda and install new unisex bathrooms rather than restore the ones that are already there. Somebody's wasting my money, and I think it's this guy who sent me this letter. He's not getting my vote.
I wish there was an efficient way to run for town council that didn't involve spending a ton of money on mailers to residents. It would be nice to actually talk with people in town to learn what their concerns are. Maybe I wasn't just home when all of these councilpeople stopped by my place.
Anyway, that's what's up here at home. Like I said, I hope to put those pictures online soon.
