sunlightpolicy / State-Open-Data-Census

Working towards a US State Open Data Census
11 stars 3 forks source link

Define Dataset: State-Owned Land (Earth Observation) #37

Open emily878 opened 9 years ago

emily878 commented 9 years ago

Define the essential substantive elements of the core State-Owned Land dataset. What are the components that it must minimally include? Do we have a dataset that we could hold up as a model?

waldoj commented 9 years ago

For reference, Virginia started providing this data last year, and they publish the following fields in their Excel file:

I'm not in any position to say whether this is a model, but it seems like a good sign for its modernity that Virginia only started tracking its own real estate holdings a couple of years ago, which is to say that somebody thought through quite recently what this should look like. (It's a terrible sign for Virginia that it didn't know what real estate that it owned until a couple of years ago, but that's neither here nor there.)

waldoj commented 9 years ago

I'm seeing that some states go beyond the list of what they own, and also list all real estate that they rent (e.g., Florida). That brings with it a set of requirements not faced by state owned real estate, such as specifying from whom the real estate is leased, for what price, for what duration, and square footage. I suggest that we stick with state owned real estate, and regard any inclusion of state occupied real estate as a bonus.

waldoj commented 9 years ago

Some states also include the year that the facility was built (if the real estate includes physical infrastructure), the size of buildings in square feet, the condition of the physical infrastructure (e.g., poor, fair, good, excellent), monthly/annual utility costs, year of construction, year last overhauled, occupancy status, date acquired, acquisition method, and structure size broken down by usable square feet and utilized square feet.

Some states only provide aggregate data (e.g., Oregon), which I want to specifically call up as being insufficient.

waldoj commented 9 years ago

These are the fields that seem to be common among state-owned real estate datasets, and that I think are a good framework for a minimum viable dataset:

There are fields that I'd like to require, but I don't think that we reasonably can. Square footage of buildings, year of construction of buildings, geodata for holdings, GNIS IDs for containing municipalities, and unique IDs for each holding would all be awfully helpful, but this data is still useful without that information.