4 ms·

> "I am from Germany where time goes from 0:00 to 24:00 - so to me they are just arbitrary numbers." I'm very comfortable with (and prefer) a 24-hour clock mys

by grzm 9y ago

> "I am from Germany where time goes from 0:00 to 24:00 - so to me they are just arbitrary numbers." I'm very comfortable with (and prefer) a 24-hour clock myself. The advantage the 24-hour clock provides is that a.m. and p.m. are no longer needed to distinguish between morning and night, and there's no ambiguity with hour 12. They're not arbitrary numbers with respect to the (approximate) position of the sun in the sky. A 24-hour vs 12-hour clock is orthogonal to the issue of time zones. I think the author's done a bit of a disservice by adding this detail here, as the issue is that you don't know from the hour whether it's morning or night. > "Compute the difference of the solar altitudes." This isn't a convenient calculation to make, and something that time zones already give us to a reasonable approximation. For example, (using a 24-hour clock): It's 9:00 in New York (and Tokyo, since everywhere it's the same). Do I call? First, I have to look up the relative positions of the sun in London and Tokyo. How do I easily do that without any assistance? One quick (?) approximation would be to look up the longitudes. New York City is at 74°W, Tokyo is at 143°E, that's a difference of 217°, or 60% of a day. So the sun is 60% further along in Tokyo than in NYC. Since we're not talking about changing the length of the day, how many hours is that? About 14½. How am I going to make that meaningful to me? Likely add it to the time, so I get 23:30. You could argue that they could put the differences on the map in terms of hours rather than doing the longitude calculations, but then you've just got (roughly) a time zone map. If you have other thoughts on how to do this, please do share. I haven't been able to come up with a scenario that doesn't boil down to this, doesn't rely on a more sophisticated tool which isn't always available, and takes into account human psychology. With time zones, I can easily find out how many hours Tokyo is ahead of London which can be found in a simple table or map (15). I add that to the current time: it's 24:00 in Tokyo. I know that it's the middle of the night. I still need to take into account what I know of but I have to do this anyway, regardless of knowing the relative position of the sun, and that's usually mental calculation based on the relative position of the sun. Time zones are a convenient short hand for computing the difference in solar altitudes and putting it into human understandable terms.

wolfgke 9y ago
> A 24-hour vs 12-hour clock is orthogonal to the issue of time zones. I think the author's done a bit of a disservice by adding this detail here, as the issue is that you don't know from the hour whether it's morning or night. I disagree: I don't know how native English speakers think about this, but the wording that am and pm stand for are references to sun positions while numbers are not.
- grzm 9y ago
  They're orthogonal because you can change one without changing the other: 24-hour vs 12-hour clock is about how we label the passage of time throughout the day locally, roughly mapping the local position of the sun to an hour table. Time zones are a method of being able to use the same labels locally and translating between them. We can change to a 25 hour day, partitioning that into 5-hour segments we label α, β, γ, δ, and ε and still use time zones to know that 2β (because we can also choose this arbitrarily with varying measures of utility) is approximately the time when the sun is highest in the sky. Another example: we could use 0 as noon and label hours + or - relative to noon, so 11 a.m. would be -1 and 1 would be 1 p.m. That would have the utility that you would know, roughly, relative darkness based on the absolute value of the hour. And again, you can change this independently of the choice whether to use time zones. This arguably makes more sense with respect to sun position: while the a.m. and p.m. label periods leading up to and after midday, the number used as part of that label is still arbitrary. This makes it even more clear that we do fine some kind of measure of relative position of the sun in the sky useful. We want to be able to understand relative to the location of a point on the earth. Hours set locally do this. Using UTC everywhere does not. Yes, you can find situations where this isn't as important, but they're generally outside of the everyday human domain. I encourage you to carry your ideas further than a one sentence description of disagreement: How do they apply across cultures? How are they practical? How would they be used in every day life? What are the implications? What are other reasonable alternatives? Please note I'm not saying they're not, but it's really hard to figure out exactly what you're getting at. My goal is not to construct a straw man to argue against, but any model that I may make to explore the issue further may be inaccurate because you've provided so little to base it on. In particular, I'd like to hear about your ideas on computing the difference in solar altitudes.

by grzm 9y ago

wolfgke 9y ago

> A 24-hour vs 12-hour clock is orthogonal to the issue of time zones. I think the author's done a bit of a disservice by adding this detail here, as the issue is that you don't know from the hour whether it's morning or night. I disagree: I don't know how native English speakers think about this, but the wording that am and pm stand for are references to sun positions while numbers are not.

grzm 9y ago
They're orthogonal because you can change one without changing the other: 24-hour vs 12-hour clock is about how we label the passage of time throughout the day locally, roughly mapping the local position of the sun to an hour table. Time zones are a method of being able to use the same labels locally and translating between them. We can change to a 25 hour day, partitioning that into 5-hour segments we label α, β, γ, δ, and ε and still use time zones to know that 2β (because we can also choose this arbitrarily with varying measures of utility) is approximately the time when the sun is highest in the sky. Another example: we could use 0 as noon and label hours + or - relative to noon, so 11 a.m. would be -1 and 1 would be 1 p.m. That would have the utility that you would know, roughly, relative darkness based on the absolute value of the hour. And again, you can change this independently of the choice whether to use time zones. This arguably makes more sense with respect to sun position: while the a.m. and p.m. label periods leading up to and after midday, the number used as part of that label is still arbitrary. This makes it even more clear that we do fine some kind of measure of relative position of the sun in the sky useful. We want to be able to understand relative to the location of a point on the earth. Hours set locally do this. Using UTC everywhere does not. Yes, you can find situations where this isn't as important, but they're generally outside of the everyday human domain. I encourage you to carry your ideas further than a one sentence description of disagreement: How do they apply across cultures? How are they practical? How would they be used in every day life? What are the implications? What are other reasonable alternatives? Please note I'm not saying they're not, but it's really hard to figure out exactly what you're getting at. My goal is not to construct a straw man to argue against, but any model that I may make to explore the issue further may be inaccurate because you've provided so little to base it on. In particular, I'd like to hear about your ideas on computing the difference in solar altitudes.