3 ms·

The "bizarre" part is that the statement "San Jose is bigger than San Francisco" is technically true while also self-evidently ridiculous. In other words, I ho

The "bizarre" part is that the statement "San Jose is bigger than San Francisco" is technically true while also self-evidently ridiculous. In other words, I hope nobody thought I was attacking tptacek! What he said is fine! Again, it does have the attractive quality of being true, even if it obviously isn't true :) A more revealing example is the statement "Austin is bigger than Boston" which is, again, only true in the most narrow sense. What any reasonable person thinks of as "Boston" is much larger than Austin. I mean this in the most common-sense way imaginable. Say your dream was to move to a "big city" and your choices were Austin and Boston. If you used the statistics within each city's political borders to decide which to move to, you'd paradoxically end up in the smaller city. That "Austin" has more people within its arbitrarily-drawn political borders is mostly just trivia. That fact doesn't help us make useful comparisons; in fact, it pretty clearly misleads. (At bottom, there's no other reason to have this conversation than this last point. We're just looking for a way to compare cities that adds value and clarifies. A system of comparison that leads you to believe Austin is bigger than Boston fails that test.)

tptacek 9y ago
That's a similarly arbitrary distinction that seems, like SJ and SF, to be the product of Austin's proximity to San Antonio, and the way San Antonio slurps up a lot of towns south of Austin that would otherwise be within the same range as, say, Worcester or Manchester is from Boston.
- dionidium 9y ago
  No, sorry, this only further confuses. First of all neither Worcester nor Manchester are in the Boston MSA, so they're not great examples. They are in the larger designation -- the CSA -- which includes cities as far afield as Providence. The whole reason CSAs exist is to help us make comparisons that do include cities that far outside of the core urban area. Unfortunately for us, the OMB hasn't defined a CSA for San Antonio-Austin, so it's hard to compare here, but almost certainly the CSA of Boston-Worcester-Providence would still be larger than San Antonio-Austin. The whole reason the OMB would define a CSA for San Antonio-Austin is to let us make these kinds of comparisons in a fair way without having to guess where San Antonio ends and Austin begins. (I wish they would!) Additionally, these are not arbitrary designations. The whole point of MSAs and CSAs is that the OMB is trying to come up with reasonable ways to compare regions that aren't arbitrary. (I mean, the conditions they decide are most important are arbitrary in the sense that they could choose some other conditions, but that's kind of a silly way to say that it's arbitrary. What slurps up what, in other words, is well defined.) All of this is just to say that there are many ways to compare regions and, again, the worst of all of them is using political borders. There is a movement afoot in St. Louis to combine the City of St. Louis and its surrounding County. Nothing about "St. Louis" actually changes the day after that measure passes (if it ever does), but St. Louis would jump way up in city rankings. The results of an 1876 law that set St. Louis's borders where they remain today is a very silly way to decide how big "St. Louis" is. Edit: Another way to say this is that using political borders is like resolving the question by taking a vote. Which city is biggest? Let's just let the people vote! MSAs and CSAs exist to help us make better comparisons than that. Edit: yet another way to say this is that your follow-up comment notices that it's hard to make comparisons because it's hard to know where one thing ends and another thing begins and which things should "count" in comparisons. That's the whole question and it's why we should use MSAs and CSAs. They're at least trying to answer those questions in a way political borders don't even address.
  - tptacek 9y ago
    This is all fine and makes sense, but to bring it back to the thread: people who buy houses are more attached to the south bay job market, which is unofficially centered on San Jose (and Santa Clara) than to San Francisco, which is the point I was making originally. People don't generally buy houses in Sunnyvale to commute to SF from (that is a brutal commute), but they do buy in Sunnyvale and commute to San Jose/Santa Clara.
    - dionidium 9y ago
      Totally. I was merely making an aside about the claim, casually made, that San Jose is bigger than San Francisco.

tptacek 9y ago

That's a similarly arbitrary distinction that seems, like SJ and SF, to be the product of Austin's proximity to San Antonio, and the way San Antonio slurps up a lot of towns south of Austin that would otherwise be within the same range as, say, Worcester or Manchester is from Boston.

dionidium 9y ago
No, sorry, this only further confuses. First of all neither Worcester nor Manchester are in the Boston MSA, so they're not great examples. They are in the larger designation -- the CSA -- which includes cities as far afield as Providence. The whole reason CSAs exist is to help us make comparisons that do include cities that far outside of the core urban area. Unfortunately for us, the OMB hasn't defined a CSA for San Antonio-Austin, so it's hard to compare here, but almost certainly the CSA of Boston-Worcester-Providence would still be larger than San Antonio-Austin. The whole reason the OMB would define a CSA for San Antonio-Austin is to let us make these kinds of comparisons in a fair way without having to guess where San Antonio ends and Austin begins. (I wish they would!) Additionally, these are not arbitrary designations. The whole point of MSAs and CSAs is that the OMB is trying to come up with reasonable ways to compare regions that aren't arbitrary. (I mean, the conditions they decide are most important are arbitrary in the sense that they could choose some other conditions, but that's kind of a silly way to say that it's arbitrary. What slurps up what, in other words, is well defined.) All of this is just to say that there are many ways to compare regions and, again, the worst of all of them is using political borders. There is a movement afoot in St. Louis to combine the City of St. Louis and its surrounding County. Nothing about "St. Louis" actually changes the day after that measure passes (if it ever does), but St. Louis would jump way up in city rankings. The results of an 1876 law that set St. Louis's borders where they remain today is a very silly way to decide how big "St. Louis" is. Edit: Another way to say this is that using political borders is like resolving the question by taking a vote. Which city is biggest? Let's just let the people vote! MSAs and CSAs exist to help us make better comparisons than that. Edit: yet another way to say this is that your follow-up comment notices that it's hard to make comparisons because it's hard to know where one thing ends and another thing begins and which things should "count" in comparisons. That's the whole question and it's why we should use MSAs and CSAs. They're at least trying to answer those questions in a way political borders don't even address.
- tptacek 9y ago
  This is all fine and makes sense, but to bring it back to the thread: people who buy houses are more attached to the south bay job market, which is unofficially centered on San Jose (and Santa Clara) than to San Francisco, which is the point I was making originally. People don't generally buy houses in Sunnyvale to commute to SF from (that is a brutal commute), but they do buy in Sunnyvale and commute to San Jose/Santa Clara.
  - dionidium 9y ago
    Totally. I was merely making an aside about the claim, casually made, that San Jose is bigger than San Francisco.