"Just entering your zip code is sufficient to obtain a list."
This is not even, sadly, close to true.
In the US, there are types of political districts that split down the middle of apartment buildings, etc. The federal districts are pretty sane. At the state and local level, pretty much anything goes.
While it's true there are a few areas where you can use zipcode and get reasonable results for some types of representatives, it's simply not correct for the US writ large for most state level political positions.
We once started generating shape files for different types of political districts we had, but for some types in the US, they aren't closed polygons, or they require 3 dimensions (the apartment building split), or ...
In some cases they are self-intersecting, due to border disputes among states, etc. Or you can prove the data is wrong, or ... Heck, in some cases, the local and state authorities have official data that disagrees on the political districts and nobody has ever really noticed!
This is one of those things people think should be really easy, and in pretty much every country but the US it is. In the US, it just varies wildly in difficulty.
Doing representatives at least means people are willing to accept some error rate, which makes it a little easier, but not much.
(For reference: I started Google's politics and elections engineering team, and co-started the voter information project to try to open up some of this kind of political data for real )
Where they intend to do it, they just draw the lines in 2d by lat/long or whatever, and if it goes through the middle of a building, oh well. The number of states still doing this is small -10 or less I believe.
When they don't, you see cases like New York city, where they don't theoretically split by Apartment, but if you look at the official data, you discover addresses in the same apartment building have been given different voting districts or whatever (this happens for all political district types). This is quite common. Remember that they use the same data, so they give you wrong answers too (until you get it corrected)! For reps you just give a wrong answer or nothing if you can detect it, but for things like voting, you tell them to call their state or whatever, because you can't risk it.
This is not even, sadly, close to true. In the US, there are types of political districts that split down the middle of apartment buildings, etc. The federal districts are pretty sane. At the state and local level, pretty much anything goes.
While it's true there are a few areas where you can use zipcode and get reasonable results for some types of representatives, it's simply not correct for the US writ large for most state level political positions.
We once started generating shape files for different types of political districts we had, but for some types in the US, they aren't closed polygons, or they require 3 dimensions (the apartment building split), or ... In some cases they are self-intersecting, due to border disputes among states, etc. Or you can prove the data is wrong, or ... Heck, in some cases, the local and state authorities have official data that disagrees on the political districts and nobody has ever really noticed!
This is one of those things people think should be really easy, and in pretty much every country but the US it is. In the US, it just varies wildly in difficulty.
Doing representatives at least means people are willing to accept some error rate, which makes it a little easier, but not much.
(For reference: I started Google's politics and elections engineering team, and co-started the voter information project to try to open up some of this kind of political data for real )