The plan was to get up at 5am, leave home at 6am in oder to get to Phoenix in time for a 7:30pm flight to Seattle. That didn’t happen.

It snowed last night. This made me somewhat apprehensive about setting off pre-dawn on summer tires. We’ve only got one set of snow tires (because we almost never absolutely need more than one car to drive in the snow, so I’ve been putting off the $400 expense of having another set of snow tires. The snow tires were on the red car, but Joanna wanted to drive the blue car while I was in Phoenix. Since I’m obviously not going to need snow tires in Phoenix, we swapped them around last night so that the blue car has snow tires and the red car has summer tires. I also pulled the hard top off of the red car since without air conditioning, I’m going to need to be able to put the top down in Phoenix.

I packed and got everything ready to go last night, but didn’t load the car.

This morning, I hauled my bags down the stairs, trudged through the snow, put my key in the trunk lock only to discover it wouldn’t go in. I tried two keys, tried to investigate the problem with a flashlight. I played with various tools, and eventually managed to get the key in the lock, but it wouldn’t turn. The lock is obviously broken in some fundamental way.

Google revealed how to open the trunk. Remove the license plate, drill a hole in the precise place described. Doing so reveals a rod that runs from the lock to the latch mechanism, which when pulled with a pair of needle-nose pliers through the hole opens the trunk nicely. I do, unfortunately, now have a hole behind my license plate, but you can’t see it, and I can plug with a rubber plug. I should have left it at that, but I proceded to disassemble the back of my car in order to remove the lock cylinder to determine if I could fix it. After removing it, I discovered that I could not fix it. I called the local Mazda dealer and they said they could order the part for me. $140 and they could have it by Thursday. That wasn’t much help. I called one of my friendly internet Mazda dealers, who sold me the lock for $64. It will ship Monday, and hopefully it’ll get here by the end of the week.

By this time, it was, of course, impossible to get to Phoenix in time for my flight to Seattle, so this means I’ll be spending the weekend at home alone. Had I left shortly after actually getting the trunk open, I would have been able to make the flight, but I wasn’t thinking very clearly that early in the morning. Amusingly, I spent this morning in the shower thinking (in a rather Anathem influenced fashion) about my forthcoming trip to Phoenix and the narrative or world-track (a coherent timeline, or set of events) that I would be inhabiting later in the day. It seemed that going to Seattle or not going to Seattle presented a choice with a significant narrative fork. In other words, I had the feeling that something significant was bound to happen or not happen based on whether or not I decided to go to Seattle and drive to Phoenix. I tried to dream up significant scenarios resulting from this fork, and come up with things along the lines of sliding off an icy road on summer tires because I decided to go today. Without some sort of causal domain shear taking place, it’s hard to know what the most optimal choice is.

So it seemed awfully strange that a non-obvious and terribly inopportune mechanical failure occurred in order to force me down this narrative. I’m aware of traditional theological viewpoint, but that strikes me as narrow-minded and boring. When things like this happen I do find myself wondering why it seems like I’m being guided down one path rather than the other, but I have to admit that there’s no evidence of any guiding whatsoever. There’s still something fishy about these seemingly insignificant details that cause significant narrative forks. I’m certainly not the first one to think so.

I learned this morning that a Colgain Airlines Dash 8 flying under Continental colors crashed last night. In my foggy state, it felt like the plane didn’t crash until I found out about this morning–that in some way Dave’s mentioning it to me was equivalent to the event actually occurring. In some quantum theoretical sense this is probably true. For me, the plane didn’t crash until Dave crashed it for me.

I’m bothered by this one. It’s the first fatal airline crash since I’ve been an airline pilot. In fact, it’s the first one since the 2006 Comair crash at the Bluegrass Airport. It’s also in the type of aircraft I’m going to learn how to fly next week. Admittedly, it’s the new Dash 8 Q400, which is far bigger and more sophisticated than the Dash 8 200 series I’m going to fly, but it’s still the same basic airframe. I’ve flown the same approach (ILS RWY 23) into Buffalo many times myself.

The cause is, at this point, still entirely unknown. The media is speculating wildly about icing conditions. The impact suggests a high speed, fast descent rate from a point on the approach where the aircraft should have been able to make the runway easily. Furthermore, just prior to the crash there were relatively normal ATC communications and no distress calls suggest that the plane fell out of the sky, something that doesn’t happen often. That makes pilots think of a tail stall–something that generally only occurs in icing conditions. If it happens, it causes a violent nose-down pitching motion. If it happened at 2,300 feet (the last altitude Colgan 3407 was assigned), it might have been impossible to recover from, even if the pilot realized what was happening quickly enough. Unfortunately, while very unlikely in aircraft with anti-icing equipment that’s functioning correctly, it’s very difficult for pilots to figure out what’s happening. It’s not a scenario that we practice in the sim, and could be mistaken for a more-likely and often-practiced wing stall. The bad thing is that the corrective actions for a wing stall and a tail stall are exactly the opposite.

I listened to the Buffalo Approach/Tower air traffic control recording from LiveATC. It’s an amateur recording from a scanner, which is set for tower and approach, switching between them depending on whose talking when, so it’s possible there’s stuff missing. The last transmission from Colgan 3407 sounded normal, and they were handed off to tower. The recording doesn’t show them checking in with tower, and shortly after the handoff, approach tries to contact them again because they disappeared from the radar. It does not appear that they made a distress call at all. They went down 5 miles from the runway over the outer marker. From the sounds of the vectors they were given, at the outer marker, they would have intercepted the glideslope at 2,300 feet and started down, while configuring the aircraft for landing. If a tail stall did occur, the change in configuration (flaps, gear) and the beginning of a descent is a likely time for it.

As of yet, there’s no word of either the FDR (Flight Data Recorder) or CVR (Cockpit Voice Recorder) being recovered. The post-crash fire was intense and lasted for hours, so it’s entirely possible that the recorders did not survive. Without them, I’m not sure we’ll figure out what happened. There are reports of witnesses who said the plane sounded weird and that it was on fire prior to touchdown, but witnesses of airplane crashes are spectacularly unreliable. I’m sure it’ll be a very long time before the NTSB releases anything.

On a final note: this is very sad. It is not a tragedy. Even our well-educated president doesn’t even seem to know the meaning of the word tragedy. Or should we give up and let the popular meaning stand?

Update: Some of the newspapers are saying that “the flight data recorders were recovered.” There’s generally only one FDR and one CVR, so hopefully the plural in this case means that both survived.