35 ms·

"Close Reading 2.0" (https://curmudgucation.blogspot.com/2014/01/close-reading-20.html https://curmudgucation.blogspot.com/2014/01/close-reading-20... ) is only

"Close Reading 2.0" (https://curmudgucation.blogspot.com/2014/01/close-reading-20.html https://curmudgucation.blogspot.com/2014/01/close-reading-20... ) is only important for authors of standardized tests and for the "look at me, I'm so clever" crowd. You page of statistics suggests you were in the second. If that's not what you want, please take this comment as a cautionary warning.

kragen 5y ago
No idea what any of that means. I accumulated a page of statistics in order to try to understand what "1 in 10000" means in practice, just as I read your comments to try to understand what Hertz's policies mean in practice, which were very helpful, thank you. Then I shared that page of statistics in order to share that understanding, just as you were kind enough to share your understanding with me.
- eesmith 5y ago
  "Close Reading 2.0" refers to students who are required to take a standardized test for 'English' wherein they analyze a document without any context. Indeed, these texts often are chosen so there is no well-known context, purportedly to prevent bias in favor of students who might have studied that context. As a result, these are rather pointless tests, as they don't reflect the skills people need to understand literature. For a certain class of student, it doesn't matter - the goal is to get a high score. These people often consider themselves "clever" for doing that. In this story about Hertz, your response focused on one detail in the story, laboriously worked out, on the apparent assumption that you didn't expect people here to know about micromorts, and had to see a full table rather than follow a link to a more full description. This is the sort of thing I see in people who think they are more "clever" than others. Further, your comments appeared to only be interested in the surface-level details of the linked-to article ("isn't mentioned in the article or my comments"), rather than considering context readily available outside that frame, and earlier mentioned in my comments - thus reminding me of "Close Reading 2.0". And when I objected to your comments, you constructed a hypothetical ("I'm guessing that reporting car thefts to the police is a standard industry practice") rather than re-evaluate the certainty of your conclusion. Something common to everyone, including me, but what "clever" people do is build on their hypotheticals to come to a conclusion which is irrelevant to the issue ("Hertz's process may actually be heavily biased in the direction of type II errors"). My interpretation can be entirely wrong. If you get no feedback you'll get no chance to re-evaluate the effectiveness or appropriateness of your sharing.
  - kragen 5y ago
    I appreciate the feedback! But that wasn't my motivation, and I don't agree that the risk of getting reported for theft if you rent a car from Hertz is "one detail of the story" or that the actual error rate tradeoff of the process is "irrelevant"; rather, I think it's almost the only actionable information about the whole story unless you're working in law enforcement or Hertz management, or defending an accused person in court. Apart from that, it's just a sad human-interest story: look, as a result of this Kafkaesque bureaucratic procedure, a terrible thing happened to this guy and this other guy you don't know. I try not to spend my time on that kind of thing both because the depths of human suffering are limitless and because stories like this are often used to manipulate rather than enlighten. It's true that nearly all inferences from information like this are necessarily based on additional hypotheses which are not necessarily stated up front; in this example, the hypothesis that reporting car thefts to the police is standard industry practice, which I still think is true. Surfacing those hypotheses when reasoning is questioned, as I did, is often a useful way to resolve disagreements, because either the hypotheses or the reasoning based on them can be wrong, and making them explicit allows you to discuss which. It doesn't imply that the hypotheses were constructed after the fact to rationalize the previously stated conclusion :) The only experience I have with this process was from a brief period of time when I lived with a programmer named John in 02000; we had been working at the same startup, which had rented some shared housing for us. John had a cocaine problem and a profound lack of consideration for others. He rented what I seem to recall was a truck and at some point left his Basenji locked up in it, returning to find the dashboard chewed to pieces. He evidently decided he needed to repair the vehicle before returning it, but didn't succeed in doing that very quickly (I'm not sure what the difficulty was), so the rental car company sent a repo man around. The truck was inside the garage with the garage door closed, and rather than answering the doorbell, John poured orange juice on the repo man from the second-floor window to make him go away. When the police started leaving messages on the answering machine about auto theft, I moved out into a hotel. As far as I know, he was able to resolve the dispute without getting arrested, though he did take a long trip to Colorado about that time. This experience probably biases me to think about true accusations of rental car theft rather than false ones. As for being clever, I have met people who are dumber than I am, but I have no reason to think you're one of them, and standardized tests are at best limited in their ability to measure such things.

kragen 5y ago

No idea what any of that means. I accumulated a page of statistics in order to try to understand what "1 in 10000" means in practice, just as I read your comments to try to understand what Hertz's policies mean in practice, which were very helpful, thank you. Then I shared that page of statistics in order to share that understanding, just as you were kind enough to share your understanding with me.

eesmith 5y ago
"Close Reading 2.0" refers to students who are required to take a standardized test for 'English' wherein they analyze a document without any context. Indeed, these texts often are chosen so there is no well-known context, purportedly to prevent bias in favor of students who might have studied that context. As a result, these are rather pointless tests, as they don't reflect the skills people need to understand literature. For a certain class of student, it doesn't matter - the goal is to get a high score. These people often consider themselves "clever" for doing that. In this story about Hertz, your response focused on one detail in the story, laboriously worked out, on the apparent assumption that you didn't expect people here to know about micromorts, and had to see a full table rather than follow a link to a more full description. This is the sort of thing I see in people who think they are more "clever" than others. Further, your comments appeared to only be interested in the surface-level details of the linked-to article ("isn't mentioned in the article or my comments"), rather than considering context readily available outside that frame, and earlier mentioned in my comments - thus reminding me of "Close Reading 2.0". And when I objected to your comments, you constructed a hypothetical ("I'm guessing that reporting car thefts to the police is a standard industry practice") rather than re-evaluate the certainty of your conclusion. Something common to everyone, including me, but what "clever" people do is build on their hypotheticals to come to a conclusion which is irrelevant to the issue ("Hertz's process may actually be heavily biased in the direction of type II errors"). My interpretation can be entirely wrong. If you get no feedback you'll get no chance to re-evaluate the effectiveness or appropriateness of your sharing.
- kragen 5y ago
  I appreciate the feedback! But that wasn't my motivation, and I don't agree that the risk of getting reported for theft if you rent a car from Hertz is "one detail of the story" or that the actual error rate tradeoff of the process is "irrelevant"; rather, I think it's almost the only actionable information about the whole story unless you're working in law enforcement or Hertz management, or defending an accused person in court. Apart from that, it's just a sad human-interest story: look, as a result of this Kafkaesque bureaucratic procedure, a terrible thing happened to this guy and this other guy you don't know. I try not to spend my time on that kind of thing both because the depths of human suffering are limitless and because stories like this are often used to manipulate rather than enlighten. It's true that nearly all inferences from information like this are necessarily based on additional hypotheses which are not necessarily stated up front; in this example, the hypothesis that reporting car thefts to the police is standard industry practice, which I still think is true. Surfacing those hypotheses when reasoning is questioned, as I did, is often a useful way to resolve disagreements, because either the hypotheses or the reasoning based on them can be wrong, and making them explicit allows you to discuss which. It doesn't imply that the hypotheses were constructed after the fact to rationalize the previously stated conclusion :) The only experience I have with this process was from a brief period of time when I lived with a programmer named John in 02000; we had been working at the same startup, which had rented some shared housing for us. John had a cocaine problem and a profound lack of consideration for others. He rented what I seem to recall was a truck and at some point left his Basenji locked up in it, returning to find the dashboard chewed to pieces. He evidently decided he needed to repair the vehicle before returning it, but didn't succeed in doing that very quickly (I'm not sure what the difficulty was), so the rental car company sent a repo man around. The truck was inside the garage with the garage door closed, and rather than answering the doorbell, John poured orange juice on the repo man from the second-floor window to make him go away. When the police started leaving messages on the answering machine about auto theft, I moved out into a hotel. As far as I know, he was able to resolve the dispute without getting arrested, though he did take a long trip to Colorado about that time. This experience probably biases me to think about true accusations of rental car theft rather than false ones. As for being clever, I have met people who are dumber than I am, but I have no reason to think you're one of them, and standardized tests are at best limited in their ability to measure such things.