SoylentNews

I know, I shouldn't expect reasonable discussion of guns here. I'm going to post this anyway to get some things off my chest.

Late last night, I was awakened by the fire alarm in my apartment building. In all the time I've lived here, since I was a Ph.D. student a decade ago, that's never happened. I found out that someone disconnected a hose from a washing machine on the third floor, water leaked down to the lower floors, and it somehow tripped the alarm.

It was no accident. Someone vandalized the laundry room, an issue that's apparently been ongoing in the building. I'm not sure if it constitutes a lease violation, but the landlord locked the laundry rooms this morning. Again, this is unprecedented since I've lived here. It's also very frustrating because it means I need to visit a laundromat to do my laundry.

I generally keep to myself, but I talked to some residents after last night's incident. Two women I talked to speculated that a resident on one of the lower levels is responsible for the vandalism. They said he's ripped down blinds from windows and he's been verbally and physically aggressive toward them. I haven't encountered him, but they said he's a large man whose aggression may be driven by mental illness. Apparently he was also aggressive toward the property manager. I don't feel safe with him in the building. Maybe he only goes after women, but I have no way of knowing that for sure.

I live on the third floor. If someone's breaking into my apartment, I could escape by jumping off the balcony, probably sustaining significant injuries when I collide with the ground. Otherwise, I need to hope that the police arrive in time to protect me. Sure, I have non-lethal force like a stun gun and pepper spray, but there's no guarantee that would halt an attack. I'm horrified at the thought of shooting someone with a gun, but if my life is in danger, I want the lethal force of a gun to improve my chances of surviving and not being seriously injured. I've never owned a gun, but this has been a wake up call that it's not safe here, and that I need to protect myself.

In my state, I have to apply to the sheriff to obtain a permit to purchase a gun. The website for the county sheriff's office says this could take up to three weeks to complete. That's three weeks that I have limited protection from someone who could be dangerous toward me. Maybe he'd never try to attack me, but I just don't know. I have no criminal record, have never been arrested, and would have a really boring background check.

I'm all for keeping guns away from dangerous people. If this individual is as crazy as residents speculate, I wouldn't want him to have access to guns. I understand the need for brief waiting periods to prevent suicides and to check the necessary databases. Still, three weeks is a long time. Surely there's a better way to keep guns away from people who are at high risk of using them to commit crimes while making it simple for law-abiding people to buy guns for self defense or sport.

Is this likely to take three weeks, or am I likely to get a permit sooner? I'm also interested in any suggestions on guns that would allow me to protect myself while minimizing the risk of harm to other residents. Maybe I should have applied sooner, but I really hate the idea of lethal force and never wanted to own a gun. Still, this process seems onerous, and there has to be a better way to do gun control.

I intend this journal as an Ask Soylent prompt. I'm trying to understand how to implement collision detection in 2D and 3D games. It seems like a complicated topic, so I'm hoping that maybe some people here have experience or could at least point me in the right direction.

Detecting that a collision has occurred actually seems very simple. It's just testing if there's an intersection between two shapes. I'm old enough to remember that collision detection was done in hardware on some old computers. As I recall, the VIC-II chip detected collisions between sprites. Now it's done in software, but it's still simple enough.

It seems much more confusing how to handle a collision once it's occurred. In order for there to be a collision, shape A (e.g., a player) must overlap shape B (e.g., a wall or a platform) by some amount. One approach is continuous collision detection, which sounds like it's computationally intensive and isn't practical for real-time use with large numbers of objects. Otherwise, it seems like the position of shape A has to be moved back to a point where it no longer intersects with shape B. I'm sure there has to be a way to do this efficiently because it was done on hardware that was orders of magnitude less powerful than current computers. How was this implemented in 2D games like platformers and in 3D games like Doom?

Let's say I have a number of balls that are bouncing around inside a box. The balls can collide with the walls of the box and with other balls. if a ball collides with a wall, its direction should change, but its position also needs to be updated so it no longer intersects with the wall. However, the act of updating its position could also cause it to collide with another object like another wall or perhaps a different ball inside the box. Within a single time step in the loop, it's definitely possible to have multiple collisions, and the act of resolving a collision could also cause the object to collide with another object. Aside from doing continuous collision detection, how would this be handled in an orderly manner?

Maybe I'm totally missing something, but handling collisions appears to be a topic that sounds very simple in principle but is actually rather complex. I'd like to learn how this was done in old games that were designed for hardware where continuous collision detection simply wasn't practical to implement.

I saw some comments in the article about the quad-state tornado that were dismissive about a possible climate change link. Just five days later, we're looking at another very unusual high impact severe thunderstorms event, this time in the central Plains.

First, I want to give a bit of meteorological background. If you know there's going to be a thunderstorm, there are three main factors that modulate its strength and severity: moisture, instability, and vertical wind shear.

Moisture is linked with instability in that, all other things equal, more low-level moisture will make the atmosphere more unstable. It's also an important factor in tornado potential, with most strong tornadoes (EF2+) occurring with dewpoints in the 60s or greater.

Instability causes rising air in thunderstorms to accelerate upwards. With strong instability, air inside thunderstorms can rise at speeds of 100+ mph. Generally speaking, more instability will produce stronger storms.

Vertical wind shear is the change in wind speed and direction with height. We often talk about the bulk wind differential, which is the vector difference between the winds at two levels in the atmosphere. Strong low-level vertical wind shear is favorable for tornadoes if the conditions are favorable. There tends to be much less instability in the cool season, but vertical wind shear tends to be much stronger during that part of the year. The strong shear can somewhat compensate for the lack of instability, also accelerating air upward in thunderstorms.

In the warm season, most tornado outbreaks occur with high moisture and instability, but the vertical wind shear tends to be weaker than during the cool season. The opposite happens in the cool season, with tornado outbreaks generally having less moisture and instability, but really strong vertical wind shear. In either case, you need sufficient amounts of all of the ingredients to get a tornado outbreak.

During the cool season, it's extremely rare to get enough moisture and instability in northern states to get tornado outbreaks. The shear is really strong, but there's almost never sufficient moisture and instability to get severe thunderstorms. Today is different... very different.

Temperatures over parts of Nebraska and Iowa are close to record highs for the entire month of December. Those records are generally in the low to mid 70s. More surprising is the amount of moisture, with dewpoints around 60 in Iowa and in the upper 50s in eastern Nebraska. This is about as much moisture as you'll ever see in these states in mid-December, if not completely unprecedented. The instability forecast this afternoon is also on the very high end of what's possible this time of year. The vertical wind shear isn't unprecedented, but it's strong, even for this time of year.

Put these together and you have the recipe for a tornado outbreak over the central Plains today. From one of the Storm Prediction Center's convective outlooks issued this morning:

This should result in potential for at least a few tornadoes, mainly after sunset. One or two of these may be strong, particularly across western to northern IA and southeast MN. This threat appears unprecedented for this region this late in the year.

Because of climate change, we expect moisture and instability to increase during the cool season. We expect that extremely rare events like this will become more frequent as the climate changes. These are weather conditions I'd expect in March or April, not December. I used to chase storms and I've seen plenty of tornadoes. The strong winds, the warmth, and the moisture make this feel like what I would expect on a chase day in March or April. I don't expect this in the middle of December.

To be clear, this isn't a repeat of Friday's tornado outbreak. There still isn't enough moisture and instability to get the type of storms that caused those tornadoes. But there's still a threat of tornadoes, some strong. Storm motions may get close to 80 mph, so tornadoes won't need to be on the ground particularly long to have long tracks. The winds just above the surface are extremely strong, and it won't take a very strong thunderstorm to bring those winds down to the surface. There are high wind warnings up over most of the central US. It's very possible there will be widespread 75+ mph wind gusts, enough to cause a lot of damage over a wide area.

Cool season outbreaks are particularly dangerous when they happen. Storm motions tend to be really fast, like today. Much of the severe weather is expected to happen after dark. It's a time of year when people aren't expecting a severe weather outbreak.

Again, this type of event just does not happen in December in the central Plains. How many "unprecedented" events have to occur before we accept that there's a link to climate change, and that these events are no longer so unprecedented?