Show HN: Identify car crash editorial anti-patterns using NLP • Hacker News Board

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#71

Post by Zigurd » Tue, Oct 12, 2021, 6:29 PM UTC

fermigier wrote: post #18

I like the general idea, and the use of NLP to implement it. But I am dubious about some of the principles behind it. - For example, I find it inaccurate to say "the driver hit the pedestrian", which to me suggests a collision between two people, not between a person and a vehicle. (Of course this does not apply to a phrase like "the vehicle fled the scene" - it is clear that it was the driver who fled the scene). Wh…

If I hit you with a wrench, saying "the wrench hit you" is clearly awkward and obscuring what matters.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#72

Post by lkrubner » Tue, Oct 12, 2021, 6:31 PM UTC

bluGill wrote: post #26

dymk wrote: post #9

This is neat, but something bugs me about the framing of the intended goal versus how you propose to get there. If the intent is to raise public awareness and to put pressure on leadership to make roads generally safer, why do almost all the corrections follow a pattern of shifting blame from the individual pedestrian onto the individual driver? Many of the suggested fixes are just their own form of counterfactual -…

One way to make roads safer is to make better drivers. There is a lot of training and continuing education that can be done to make better drivers. It is far too easy to drive, and once you get your license it is very hard to lose it even as your body degrades to the point you cannot see. People think of bad driving tickets (mostly speeding) as just a price you pay to drive. I've never heard of someone getting a tail…

I live in the USA. My friend Kristin is from Germany. She got her driver's license in Germany when she was 28. She trained awhile for it and then took 3 tests: city driving, highway driving, and nighttime driving. From how she described it, the training and the testing was much more rigorous than anything that has existed in the USA.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#73

Post by seltzered_ » Tue, Oct 12, 2021, 6:32 PM UTC

bluGill wrote: post #26

dymk wrote: post #9

This is neat, but something bugs me about the framing of the intended goal versus how you propose to get there. If the intent is to raise public awareness and to put pressure on leadership to make roads generally safer, why do almost all the corrections follow a pattern of shifting blame from the individual pedestrian onto the individual driver? Many of the suggested fixes are just their own form of counterfactual -…

One way to make roads safer is to make better drivers. There is a lot of training and continuing education that can be done to make better drivers. It is far too easy to drive, and once you get your license it is very hard to lose it even as your body degrades to the point you cannot see. People think of bad driving tickets (mostly speeding) as just a price you pay to drive. I've never heard of someone getting a tail…

This is a complicated topic. One thing I can recall is that there's more pedestrian collisions as there's been more safety technologies and larger cars and average vehicle miles travelled (can't find the study, this might be a lead: https://safetrec.berkeley.edu/news/new-release-2020-safetrec... )

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#74

Post by dfxm12 » Tue, Oct 12, 2021, 6:34 PM UTC

dash2 wrote: post #60

ihaveajob wrote: post #41

Earlier quoted context omitted.

The problem with using "accident" as the default term is that it preempts any evidence-finding, implying that it was nobody's fault. If someone shoots their gun randomly and someone gets hurt by just being there, nobody will call it an "accident", but somehow if the weapon is a car, the driver gets the benefit of the doubt regardless of the facts.

Of course if there is evidence that the driver deliberately tried to kill someone, nobody will call it an accident. I would imagine that the vast majority of car deaths are indeed accidental, not murders. This seems to contrast with the situation of gun deaths. No?

I'm not sure. I read a lot of gun violence reporting like "person a was struck by a bullet and later died due to complications" vs "gunman shot and killed person a". It's kinda the same thing.

I would make no assumptions about if car accidents are accidental (which for some reason seems to be the default) or not - and that's more or less the point of the article.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#75

Post by kazinator » Tue, Oct 12, 2021, 6:35 PM UTC

praxulus wrote: post #66

chiefofgxbxl wrote: post #53

Earlier quoted context omitted.

Thanks for the feedback. There seems to be some conflict with how to define "counterfactual". Merriam-Webster defines it as "contrary to fact" [1], so that comes down to a statement being either true or false. But the way I'm using it (and the research my work is based off) aligns more with the Cambridge dictionary usage: "thinking about what did not happen but could have happened, or relating to this kind of thinkin…

Your counterfactual explanation makes sense in the helmet example, but I think most people wouldn't be able to figure out that connection without such an explanation. Even if it's correct, it's still confusing. For example, I can't figure out what the relevant counterfactual is in the alcohol example. That's not to say that these are purely value-neutral facts, it still makes sense for them to be called out by this t…

I think the issue is this: there is a bias when certain news sources only mention facts about what drivers did right (like not being impaired) and about what pedestrians or cyclists did wrong (like not wearing a helmet).

We cannot easily detect this pattern of bias from just a sample of one article from the given source, however, even from one article, there can be a hint of this bias. For instance, a statement about the impairment level of the driver is not made, but the status about the bicycle helmet is made.

To avoid, or at least reduce biases, the reporter has to have a standard template of all relevant fact types, fill it in with everything that is known and then report on everything. If the cyclist was wearing a helmet, report that; if not also report that.

(The available facts may be biased, like what the police and other on-scene responders take in and communicate.)

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#76

Post by Bermion » Tue, Oct 12, 2021, 6:37 PM UTC

dymk wrote: post #9

This is neat, but something bugs me about the framing of the intended goal versus how you propose to get there. If the intent is to raise public awareness and to put pressure on leadership to make roads generally safer, why do almost all the corrections follow a pattern of shifting blame from the individual pedestrian onto the individual driver? Many of the suggested fixes are just their own form of counterfactual -…

We need to limit speed and the amount of interactions between cars and pedestrians. This will necessarily be at the expense of the drivers who will have to drive slower and get less space since they have historically been prioritized.

It is much easier to implement the necessary measures (against the will of many drivers) if it is clear that drivers run over people, not that people accidentally end up under cars.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#77

Post by uDontKnowMe » Tue, Oct 12, 2021, 6:40 PM UTC

Thank you so much for this, it's great work!

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#78

Post by dymk » Tue, Oct 12, 2021, 6:51 PM UTC

Bermion wrote: post #76

dymk wrote: post #9

This is neat, but something bugs me about the framing of the intended goal versus how you propose to get there. If the intent is to raise public awareness and to put pressure on leadership to make roads generally safer, why do almost all the corrections follow a pattern of shifting blame from the individual pedestrian onto the individual driver? Many of the suggested fixes are just their own form of counterfactual -…

We need to limit speed and the amount of interactions between cars and pedestrians. This will necessarily be at the expense of the drivers who will have to drive slower and get less space since they have historically been prioritized. It is much easier to implement the necessary measures (against the will of many drivers) if it is clear that drivers run over people, not that people accidentally end up under cars.

I agree that roads need lower speed-limits and to be made safer by their geometry. That's done by reporting on the unsafe conditions of the road.

It does no good to play this game of rhetoric, reframing the matter so the driver is at fault (which is not always the case). A driver does not decide to not run over a person because they think a news story will be more accusatory of them.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#79

Post by FabHK » Tue, Oct 12, 2021, 6:57 PM UTC

JackFr wrote: post #30

whimsicalism wrote: post #20

Earlier quoted context omitted.

Yes, I do not think you bear responsibility if you collapse and someone strikes you with their car and then flees the scene.

Don’t be obtuse. A person collapses into the street and is struck by oncoming traffic - no one is to blame. It is an accident.

> no one is to blame. It is an accident

That is not the mindset with which aviation arrived at its amazing safety record.

Re: Show HN: Identify car crash editorial anti-patterns using NLP

#80

Post by lelandbatey » Tue, Oct 12, 2021, 6:59 PM UTC

dgfitz wrote: post #67

ecshafer wrote: post #31

Earlier quoted context omitted.

If someone collapses into the street, the driver should be able to stop in time, unless they are going to fast. Who I can think to blame here: * The auto manufacturer for not having sensors to detect this. * The city for making car dependent areas * The city for making roads that have cars drive so fast by pedestrians, that the pedestrian can die if hit * The driver for not driving more reasonably * The city for not…

> If someone collapses into the street, the driver should be able to stop in time, unless they are going to fast. Or they collapse mere seconds before the car strikes them making the __accident__ unavoidable.

The duration between when the pedestrian collapses and when the driver runs over the pedestrian is not as relevant as you seem to be implying. If someone is standing in a crosswalk in broad daylight and falls over, even if the pedestrian collapses 0.1 seconds before someone runs them over that doesn't leave the driver blameless. Why, in broad daylight, in the middle of a crosswalk, was a driver 0.1 seconds away from running over a pedestrian? That sounds like poor road design, poor driving, or another factor. The same logic applies even as we worsen the conditions: if it was dark, where was the lighting? If the pedestrian "suddenly appeared" from behind an obstacle, why is there an obstacle that close to a road and why can a pedestrian get so close that they can "appear" from behind that obstacle?

Roads should be designed so that pedestrians cannot accidentally end up 0.1 seconds (or pick your duration) from being crushed by a driver. And drivers, state-licensed machine operators, should face an incredibly high bar of scrutiny when it comes to damaging pedestrians. Yes this probably means very different road designs than the ones we have now, that is one of the end goals of this kind of effort.