Does the Duck Accident Count?

[Paul in Saudi]

The Commissioner seeks advice.

The annual Mass Killing List has always included vehicular homicides (along with arson and poisonings). It is not and was never meant to be another list of gun crimes. In fact, that is why it is unique.

There are problems with vehicular homicides. If the driver is killed, he cannot be charged with a crime. Therefore the list only includes a small subset of such killings. Further, such cases are hard to detect in the news flow. Nonetheless, the Commissioner has ruled that vehicular homicides are to be included. He is resistant to changing the rules and so making the lists different from year to year. He aims for perfectly comparable data year after year.

So what is the best way to handle the Branson, Missouri Duck Accident? On Thursday 19 July 2018 a tourist truck entered the water despite weather warnings. The weather did in fact change with remarkable speed, swamping the vehicle. Seventeen people were killed.

Reports indicated the driver was also killed, so I lost interest. No it seems the driver lived and has been charged with "17 counts of misconduct, negligence or inattention to duty by a ship's officer, resulting in death."

I suppose it has to be included, I can think of no reason why it ought not to be. The Commissioner appreciates your thoughts.

Further, as the end of the year approaches, I am open to discussion that this annual thread be moved or posted elsewhere. I realize it is a depressing subject.

 

[Bootstoots]

I'm a big fan of your thread. Morbid body-counting is my kind of thing.

I'm not really a fan of counting vehicular homicides towards the total, except where they are intentional. I fairly strongly suspect that a really thorough search for drunk driving and reckless driving accidents would swamp the intentional homicide total and inflate the numbers.

But the Commissioner sets the rules with absolute authority, much as the King (or, in practice, the testosterone-addled Crown Prince) sets the rules in the Commissioner's adopted home. I would advise that if he wishes to continue including vehicular homicides that he include reckless duck boat drownings in Ozark tourist traps, for consistency's sake.

 

[warpus]

What you first need to figure out is what the data is being used for. What are the data reporting needs of the Commissioner? Why does the list exist? What are the lists' parameters? What needs do those who have requested the list to exist have? Will the list drive any other processes? What are the parameters of those processes?

I recommend an interview of the Commissioner by a data analyst and potentially a data modeler and/or backend programmer. Afterwards a report can be compiled to present several possible ways of implementing the list, depending on the given parameters and operating budget.

 

[Paul in Saudi]

The Commissioner appreciates the input. He figures he would prefer to err by inclusion than by exclusion. As long as the nature of the killing is noted for each case, a reader may pick and choose as he likes.

 

[Timsup2nothin]

C'mon man...accidents happen. If you set this precedent consider the can of worms you are opening.

What about a plane crash? There's always some zealot shrieking about how pilots are supposed to be perfect and they should be charged with the death of their loved one.

How about fires? Exhausted mom leaves a burner on and the house goes up, killing the four outright monsters she is trying to raise into humans. Does she belong on your list?

Heck, a long time ago on an oil rig seventeen people suffocated in a self righting escape pod because for it to actually self right it needed their body weight strapped in the seats for ballast and they hadn't bothered because it was "only a drill." All kinds of shrieking ensued, and lots of it was insisting that firings and wrongful death suits weren't enough and someone needed to go to jail...even though no one even had a good guess as to who would be appropriate.

 

[Paul in Saudi]

Plane crashes and house fires do count if a person is charged with a crime. There are instances of both in the older lists.

All in all, The Commissioner has ruled in favor of inclusion. This ruling cannot be appealed.

 

[Bamspeedy]

If the the killer with the gun is killed, he and his victims are counted. Yet if the killer with a vehicle is killed, he, nor any of the victims are counted. You say the killer with the vehicle can't be charged with a crime if he's dead, yet the killer with the gun can't be charged with a crime either as he's dead too.

You want consistency, it sure isn't there.

I'll say it before and I'll say it again. You list should be 'intentional' killings.

 

[Silurian]

DUKW?

 

[uppi]

What you first need to figure out is what the data is being used for. What are the data reporting needs of the Commissioner? Why does the list exist? What are the lists' parameters? What needs do those who have requested the list to exist have? Will the list drive any other processes? What are the parameters of those processes?

I recommend an interview of the Commissioner by a data analyst and potentially a data modeler and/or backend programmer. Afterwards a report can be compiled to present several possible ways of implementing the list, depending on the given parameters and operating budget.

The data scientists says: collect the data first, worry about its application later (or never). It's better to have the data and not need it than to need the data and not have it. The only condition is that the Commissioner is vigilant about data quality. A heap of bad data is useless, despite management belief that *technology-of-the-day* can turn crap into gold.

 

[Farm Boy]

Do we have any suggestions on sources of vehicular homicide records with which to assist the commissioner in accurately grasping their entirety?

 

[warpus]

The data scientists says: collect the data first, worry about its application later (or never). It's better to have the data and not need it than to need the data and not have it. The only condition is that the Commissioner is vigilant about data quality. A heap of bad data is useless, despite management belief that *technology-of-the-day* can turn crap into gold.

The problem with this approach is that until you know he exact data requirements, you could very well miss something in the data collection. You could put it in a format that will not work with the data reporting requirements. A lot of things could go wrong. You collect requirements first, data second.

I'm actually involved in a project here at work where we decided to collect the data first, because.. well... because we suck. And now we have problems delivering certain things that the project requires, because the data is faulty wrt to the requirements (but could have been just fine were the requirements different)

 

[uppi]

The problem with this approach is that until you know he exact data requirements, you could very well miss something in the data collection. You could put it in a format that will not work with the data reporting requirements. A lot of things could go wrong. You collect requirements first, data second.

I'm actually involved in a project here at work where we decided to collect the data first, because.. well... because we suck. And now we have problems delivering certain things that the project requires, because the data is faulty wrt to the requirements (but could have been just fine were the requirements different)

Good data can always be transformed into the format you need. If you already know exactly what you are going to do with the data, then it is of course prudent to think about the requirements first (but throw in a few extra fields if you can). But if you don't know how exactly you are going to use it, put it in a good structured, extensible format and you can always transform the data later. If you always want to think about requirements first, you are going to end up in the madness of big companies where you need to specify the data in a 100-page document and then pay hundreds of thousands of Euros to implement it exactly like that (and then ten thousands of euros when you want a field changed). Better start collecting it, analyse the first samples and then modify the collection process if necessary.

 

[JollyRoger]

I am against any accidents being counted.

 

[warpus]

Good data can always be transformed into the format you need.

It's not about formatting, it's about the way the data's logically laid out and collected.

 

[Paul in Saudi]

All in all, the Commissioner does not want to "not count" a person's death. It seems each is a tragedy. To be killed by a criminal act seems even worse.

The Commissioner thanks you all.

 

[Bootstoots]

Could The Commissioner maybe make a second tally for the deliberate totals? Maybe something like:

Events 134 (77)
Killed 671 (398)

Where the headline number is the total number of cases of criminally culpable mass killings of four or more people, and in the parentheses are the killings that are deliberate?

 

[Paul in Saudi]

The Commission waives any right to his work and welcomes you to slice, dice, reproduce or rebroadcast it in any way you like. Frankly, he just does not know how to do a better job with reliability.

 

[TheMeInTeam]

All in all, the Commissioner does not want to "not count" a person's death. It seems each is a tragedy. To be killed by a criminal act seems even worse.

The Commissioner thanks you all.

At the same time, if you accept this as a "mass killing" you would indeed wind up having trouble differentiating it from routine car accidents that result in 4+ deaths. In both cases, the deaths were not anybody's intention. That doesn't make them less tragic, but it really does appear to be a distinct category from incidents where someone inflicts harm/death intentionally.

 

[Farm Boy]

Eh. The kids with the virus in NJ are just as dead as the ones in the bar in CA. If we can strip mens rea down such that you can be criminally charged for homicide for doing something accidentally, then that counts as legal intent.