Maybe this will help, a bit, Jared:

Friendship improves happiness and abates misery, by doubling our joy, and dividing our grief.

-Joseph Addison

We’re here.

No, not mipmaps. Just a lot in a small space (hopefully small):

• We’re starting to get some nice ice growing on the fountain:
  
  
  Once we can get the heater to run through the night consistently without tripping its GFI, it’ll be awesome.
• On the topic of the fountain, progress is being made in teaching it to speak SMTP. I have my module mostly complete, and have also found Giddings & Lewis’s official SMTP email block, which I am poring over to find the bug in mine. It was interesting, though slightly disappointing, to find a functional block from G&L that does, in fact, work, after putting several hours into writing my own.
• Google’s Russian definition of concatenate contains the word treehouse.
• I lost, yet again, to Jared in racquetball last Friday. A bit disappointing that I was up 7-3 and still lost 15-11 or so. That’s at least the 2nd time this year I’ve done that, though last time was worse: up 13-2 I still lost 15-14. I suppose it’s only fair, mason. Makes up for nearly every math test since 7th grade 😉
• Sunday night I decorated some Christmas cookies with Jeni & Lauren (and, briefly, Erin). We’ll find out tonight if any survived for the Enlight meeting… (I take full blame for the horrible attempts at rendering a fountain and the Enlight logo onto a tree and a bear, respectively. I quickly realized I’m not capable of such things and desisted.) Later Jeni and I were confounded by a tangram puzzle that I think is impossible. And Lauren solved the triangle problem in just a day. A lot less time than it took me (about a year and a half, I think. Anybody remember when I first brought this to school?):
  
• Nate had a very original, inventive, and truly inspired idea last night. But he doesn’t want me to share it. So for the record, Nate had his first idea in a long time, but it died of isolation. (kindof like the died of loneliness joke… I tried.)

I think that’s about it.

I’m so thrilled that my computer finally works again (after 36+ hours) that I decided to stay up an extra couple minutes to write about it. After all the hassle described in my last post, plus at least 2 hrs of figuring out why more packages wouldn’t update today, as of 1:30 am everything was finally done! My computer is back online and running (so far) happily.

In other news:
I’m working on a code module for the Fountain to enable it to send email messages. That should be pretty useful, once done, for getting status information like “sidewalk water overflowed, come fix it,” or “the mister DID work last night, go check it out,” or “pump 2’s vacuum sensor says there’s a problem. get ahold of tim, stat.” etc. without having to physically be at the engineering campus or in the tunnel. Plus, in working on this I am gaining a much better understanding of how PiCPro’s networking code is supposed to be put together, and why our previous code worked so precariously. Hopefully once this simpler module is complete, I can take a look at the whole server’s code and update that to work the way it’s supposed to.

Long-winded, frustrated technical post alert. Sorry, but sometimes I need to rant about geek things. (At least this one’s not trite, ey Katie?)

I forgot just how much of a PITA it is to update my kernel. I run Gentoo linux on my desktop at the moment, and decided recently, on Jon’s urging, that it would probably be a good idea to get it back up to date. Sounded easy, especially when he suggested just nice’ing “emerge -u world” in the background, so I decided to start doing that this afternoon. Well, unfortunately there’s a bit more to it than that. Seeing as I have 342 packages to upgrade (all to be recompiled from source, since Gentoo is a source-based distro), I get interrupted about every 20 packages with some problem to sort out the portage can’t handle automatically. First it was mysql. Can’t autoupgrade from 4.0.24 to 4.1.14. So, following a handy wiki, I had to first upgrade to 4.0.25, then backup and delete all my settings, then remove 4.0.25 completely, then install 4.1.14, then replace all my settings. Took like 20 minutes of manual intervention. Not a huge deal. Rolling further along. This package wants me to delete some directory, so the thing dies, then that package has an illegal /usr dynamic library link (???), so unmerge it manually, then I get to hal (Hardware Abstraction Layer). Turns out hal needs a kernel upgrade. Shit.

So I emerge gentoo-sources to get 2.6.14. No big deal. Copy over my old .config, update the new settings as needed, give it a quick once-over to make sure everything is still ok. Looks good. Compile that, mount my suse boot partition (since that’s where my bootloader actually is), copy the System.map and vmlinuz files over there, update grub/menu.lst, then reboot. Big mistake. I forgot in the heat of the moment that my motherboard has a network chip that’s not supported by the standard kernel sources. (There’s a driver in development at version 0.9, but the author says it’s still buggy…) Plus, NVIDIA’s custom kernel module dies every time you recompile, so I get uncermoniously dumped at a console full of module loading errors. Great. Cue 45 minutes of rebooting into SuSE, downloading the updated, nasty kernel module from SysKonnect, rebooting into the console of errors, patching my kernel tree, make modules && make modules_install, modprobe sk98lin, oops that didn’t work because the retarded non-functional default sk98lin tried to load already and died… reboot, readd /etc/init.d/net.eth0 to my startup scripts, reboot, finally on the net, now download & install updated nvidia drivers, reboot.

X finally comes up and lo-and-behold my entire GUI environment is dead. Seems portage thought it’d be a great idea to emerge parts of gnome 2.12 before hal (the package that caused this whole mess), so now I can’t reload gnome until everything is done updating. At the current moment, it’s compiling #6 of 243 remaining packages. Splendid.

If you actually read that this far, I hope you’re sleeping happier than I, and less fitfully than my poor computer.

Well, as of yesterday it’s official: Amphisbaena, the ACM International Collegiate Programming Contest team that I am a part of, will be representing UW-Madison at the 2006 ACM-ICPC World Finals! (See proof of our accepted status) That means a trip to San Antonio, TX, as well as some rather intense study of algorithms & programming strategy coming up next semester. The contest will be April 9-13 at the Hilton Palacio del Rio, and is sponsored by IBM and hosted by Baylor University. I expect the contest will be extremely tough, but on the upshot, it gives us a second chance to beat South Dakota School of Mines & Technology, which, if you recall from my earlier post, is my dad’s alma mater, and furthermore beat us at the regional competition.

Awesome!

Prince Rupert’s Drops

Just saw these on hackaday.com. Very cool little things. You can read up on Prince Rupert’s Drops at Wikipedia, but the basic idea is this: Drop molten glass into a cold bucket of water. A tadpole-shaped piece of glass is formed in a way that produces a huge amount of tension within the glass. On the big tadpole end, you can hit it with a hammer, and unlike ordinary glass it won’t break. (Due to the high tension and the egg-like shape distributing the force.) However, if you even scratch the tiny tadpole tail, the entire thing explodes into tiny little shards, almost instantly. (The exploding front propagates down the tail toward the head at ~4200 mph = Mach 5.5)

Even if you didn’t read the above paragraph and/or didn’t care, check out this video of Prince Rupert’s Drops exploding from museumofglass.org. It’s awesome.