Out of curiosity, I wondered what sort of data my home county (Salt Lake County) kept and if I could get access to it.
It was easy to find... but apparently will cost me $1000 dollars to access it...
Perhaps I don't understand exactly how the Assessor's office works, but it seems pretty wrong that public data from an organization funded by taxpayers isn't freely available for download.
The article doesn't mention it, but it looks like LA County charges as well.
I started a company that works extensively with geospatial data and I often interact with cities from across the country (https://angel.co/civic-insight). Learning how to get access to government data was probably one of most challenging parts of starting up, but I've seen a significant evolution in the thinking of governments in the past few years.
I was a Code for America Fellow in 2012 (I am still heavily involved in the space) and back then any talk of publishing datasets, shapefiles or having "APIs" was thought to be an exotic topic reserved for fancy cities with large budgets like San Francisco and Boston. It was something that smaller cities would love to have but don't have the resources to do anything about it.
But that's not the case anymore. Even a lot of the smaller cities now understand that opening up now gives them access to a new ecosystem of tools and technologies that are rapidly developing.
It is really easy to be cynical about government policies and feel like there is little that can be done to change them, but it seems like what you are doing and orgs within the government, like 18F, are having success in changing the game, at least when it comes to tech.
It depends. Taxpayer dollars may only barely cover the cost of getting the data together, but distributing it could be fairly significant as a budgetary item. To cover that cost they may just charge some fee that has several uses:
- Keep taxes down and only support people who really have a serious need for it
- Cover the cost of packaging and distribution (and other associated costs)
I don't necessarily agree with this, but these are the kinds of justifications that are generally tossed around inside the government about these kinds of things.
A better question for the people that live there is 'are the cost of distributing the data artificially inflated?'.
In my own county the costs of document scanning and record storage were in the 1 to 2 million dollar range for years. Oddly this coincided with the county commissioner having a husband that work for Xerox at the time, the Xerox office also handled the counties scanning and document storage. Amazingly after the time that commissioner was no longer in the office the contracts were renegotiated and the price dropped over 25%. The biggest issue I have seen working at the local government level for over a decade now is very few people without a stake in the game ever review the use of technology by the counties. Politics has far more sway than best practices.
I'm assuming you meant the prices dropped to 1/4 of their previous level. Even then, pretty much any civic minded organization or "citizen scientist" is priced out of the market. In fact, the only people who can access it, in that range, are people who can turn a profit on it, i.e., likely not looking out for the public interest as a primary concern.
Prices didn't drop, but oddly enough they were not excessively priced in the first place, as compared to many other counties. The operations of the records department are funded from the general fund and not self supporting. Still if you were purchasing many thousands of pages like someone in the data sciences would, the cost would be excessive.
In retrospect probably one of the dumber things I've ever done is not keep 2TB of county records from 1850 to 2008 after a company closed down operations. There wasn't any particular reason I couldn't keep them, they were paid for and the company didn't sell them off after closing. The server was likely neglected for years after that and eventually formatted.
In a roundabout way, that answers the original question, doesn't it? This sort of data is so seldom used, that efforts at 'standardizing' and 'broad distribution' don't scale. You didn't see any value or use in the data for years, only in hindsight, maybe, it would have been fun to look at. Because that's what it is - it's fun (for those from outside the field) to read posts like the OP every once in a while, but there's very little actual value in it, and it's too much work for the vast majority of people to do themselves.
I work in geospatial modelling and I use data like this every day; have been for 15 years. So I know a thing or two about availability of this sort of data, how hard it is to assemble and maintain, and how many people actually do something with it. In the EU, there are directives that create obligations for governments to offer all sorts of data freely to everybody. The amount of money this costs is staggering, and although I'm very happy with it (because it has made my life 100 times easier in the data respect), the useful stuff that is done with it by people who didn't have access to it before is minimal; and certainly disproportional to the amount of money that goes into it. The consultants are laughing all the way to the bank, of course (and hey, I get paid indirectly from it as well, so it's not like I'm complaining).
So that makes the 'data should be open' mantra an entirely ideological disposition. I used to believe it as well, but then why shouldn't we also say 'governments can only spend money on vegan stuff because some people believe in 'animals rights'?' I no longer see pet peeves like 'free data' as black and white as I used to. Bureaucracies are expensive, partly because their goal is not efficiency (which is something that is usually overlooked by the naive) and partly because of other effects inherent to them. We need to restrain ourselves in what and how much we demand from them.
A group from Code For America came to my town recently (Tempe, AZ) and they address a lot of the issues with access to public data. Was really glad to see things like this exist. http://www.codeforamerica.org
$1000 doesn't sound expensive to me, for the first request. Somebody has to think about the question whether there is confidential or copyrighted information in there, somebody has to to write a decent disclaimer, think about any security implications, etc.
Having said that, a government nowadays should have an open data policy. You shouldn't need to make a freedom of information request for 'bureacratic' data.
Also, any 'for the people' government should allow the free download of such information, and 'the people' should realize that that costs some money. That, I think, is a problem in the USA, from both sides.
> it seems pretty wrong that public data from an organization funded by taxpayers isn't freely available for download.
One really nice thing about the USA is how much public and open data there is. In Europe you can't even get access to maps, which is why OSM was created in Europe.
Whilst the Open Street Map is free its important to note that the Ordinance Survey maps are some of the most detailed in the world. They offer a free access option and a paid option. Whilst the organisation is operated out of public money I am not opposed to its usage terms that are in effect a tax on business use.
You can get access to a lot of data in Europe, including maps (ordinance survey). In fact the UK has more open data available than the US, according to OpenDataBarometer:
That doesn't appear to be a very good counter point, your own link confirms what the parent said.
There's one country ahead of the US in that list, half of Europe averages middling scores, and the other half is either not even listed (eg Bulgaria, Moldova, Croatia, Belarus) or posts incredibly bad scores (eg Poland, Greece, Hungary, Belgium, Iceland).
The interesting thing with charges like this is that they tend to be decided by one person. (in the case of things like court fees and record access, Clerks of Court often ignore state mandates, because they can) The easiest way to fix this might be to make an issue of this the next time the County Assessor has an election. Which might cost more than $1000, but if it were your main business...
my previous county listed lots, primary owner, and assessed value, when they first went live many years ago. Since then I think they only removed the owner's name; it usually was just last name.
why was it important, having someone who you don't want contact with could determine your home and possible income simply by searching the county tax records.
>Perhaps I don't understand exactly how the Assessor's office works, but it seems pretty wrong that public data from an organization funded by taxpayers isn't freely available for download.
Hear, hear!
I have looked at various online records offices for various properties located in cities across the USA and most every single one was a horrifyingly slow, blast from the past type experience. I'm talking minutes for basic queries to complete. And expensive to boot. Usually $5 to $10 per query.
> Perhaps I don't understand exactly how the Assessor's office works, but it seems pretty wrong that public data from an organization funded by taxpayers isn't freely available for download.
I think those costs are there because the provision of that data is considered above and beyond the mandate of the Assessor's Office.
>>public data from an organization funded by taxpayers isn't freely available
Perhaps it was free, they wouldn't have resources to compile it? Plenty of things that tax payer money supports aren't free, for example suburban swimming pools or perhaps more egregiously sports venues.
Whoa. That escalated fast. I thought I was reading an simple article about property data. And then, BOOM! All the sudden we're discussing file formats, and then making database queries. And out of nowhere this monster appears: http://robrhinehart.com/wp-content/uploads/2015/01/spherical...
That equation is rather unnecessary showing off, I don't see it being used later on, and the author doesn't bother trying to explain it. It looks like (going by the GM/r) that it might be the gravitational potential for a mass distribution described by a bunch of spherical harmonics.
Yes, especially as one of the key advantages of using a projection is that you don't have to worry about this stuff. You can just work with a Cartesian coordinate system and the results will probably be right.
And immediately preceding that equation is this sentence:
"If it sounds simple, it is not, but if you’re interested it’s a great reason to learn spherical harmonics."
The entire point of showing off that equation is to say "This stuff is actually really deep and complicated and it's a rich subject to dive into on its own if you want to go for that, but I'm not going that route."
I have the firm belief that Los Angeles is actually the largest city by population in the USA (as opposed to NYC). Due to an accident of history, what people colloquially call "LA" is still divided politically into something like 80 independent cities. New York City consolidated into the five boroughs in the late 19th century, prior to that Queens, Brooklyn, etc. were separate cities. If you look at the city borders of official Los Angeles, it's a complex mapping of the spaces around and in-between the various independent cities.
In most of the US, unincorporated land is extremely rural. In LA, "unincorporated" land blends seamlessly into the urban landscape. Often if you're in an unfamiliar area you can't tell what "city" you're in without looking at the street signs.
While not 100% of LA county residents can be counted as living in "LA", it's a close-enough approximation. ~10 million vs NYC's 8 million.
If you're going to include all that into LA, shouldn't NYC lay claim to Yonkers, Jersey City, and dozens of other places? The New York metropolitan area is certainly larger than the Los Angeles metropolitan area, and I'd wager that there's no circle of radius R in the US, for R between 1 mile and 1000 miles, that contains more people than one centered appropriately on lower Manhattan.
A key difference is that the actual City Of Los Angeles is quite sprawling and surrounds many of the areas that are incorporated as their own municipalities.
It's amusing -- in Australia we list cities populations like this by default. I never consider the population of Melbourne to just be the area covered by Melbourne City Council - it has always included the greater metro area.
That's generally true here in the US as well. I guess the distinction is being made in this case because the city data is split into separate jurisdictions and databases... there is no "greater metro area" database for land use and population.
however, i don't think that includes large parts of orange county, which is adjacent to LA. OC has a distinctly different feel but it's all smushed together, there isn't like a large open expanse of nothing like between OC and SD.
If you're going to go down that route, you'd also have to consider sections of NJ and Westchester County. The density of say, Jersey City is 16,736.6/sq mi. There's the Hudson River in between, but so what?
Greater NYC still includes a lot of very dense conurbation well beyond the city limits. For example, some of the densest cities in the US are those in Hudson County, NJ, and some of the suburban downtowns (for example, White Plains) have skylines that rival midsized Midwestern cities. Newark, though down at its heels, is linked to NYC by a 35-minute rapid transit trip at 10-minute headways, the commuter rail network is even more extensive than that, and bus commuters come to NYC from Allentown and the Poconos in Pennsylvania.
You can't just count an entire massive county as 'LA' while cutting off NYC at its city limits. The stub-ends of the subway lines at the Bronx border are dramatic, but the city doesn't just end anywhere— I wouldn't be shocked to see the subway extended beyond NYC at some point.
You could do the same with LA. "Greater LA" also includes the Inland Empire, Orange County, Riverside/San Bernadino. Total population (according to Wikipedia) is around 18 million.
I agree that you can draw the line arbitrarily to support whichever city you prefer as "winning". My sense is that a) if the City of Los Angeles consolidated as NYC did over 100 years ago it would have the highest official population in the US, b) if you do a more fuzzy "metro area" analysis the result is up for grabs depending on how you choose to draw the borders.
An interesting bit of naivete from Rob Rhinehart here:
Most, if not all, counties use GIS (geographic information systems) to maintain this data
Most people have no idea how appallingly corrupt, backward, wrong, inaccurate, inefficient, unjust, etc. etc. record keeping and government in general can be at the county level in some places. For the purpose of brevity let's put it this way: some counties maintain a poor paper trail, deliberately. Those counties also tend to be poor, so you might just assume they cannot afford better systems, but sometimes there is a bit more than that going on.
I actually think it would interesting to know how many counties make use of GIS.
Regarding GIS data specifically, the first thing I thought of was a story from a friend how owns a gas well in rural West Virginia. There's a framework for determining who owns the rights to extract gas at a given depth on a given property.
There are engineering-based ways to cheat the system (oops! there's totally accidentally a hole in my well above the depth where I have the right to extract gas) but the thing that struck me about her story was how bad all the basic county record keeping was. People die, mineral rights (and other property) need to be handled by the court, and ambiguities or errors on the record of who owns what where that might have easily been corrected when people were still alive become a matter of dispute, and some are a lot better equipped than others to legally deal with that kind of dispute. Her complaints had more to do with an obviously corrupt method of informing interested parties of things like auctions, whose inefficiencies always seemed to benefit a small number of people equipped to exploit them.
I'm sure someone with more experience with real property could give lots of examples. My experience is more with county court systems, and I just assume that if counties haven't computerized their criminal and civil records - which is dead simple, conceptually - they haven't got around to implementing relatively more sophisticated GIS systems. Maybe I'm wrong.
Working with landmen and title companies in Texas has taught me that even the best kept computer GIS records are probably wrong. This goes for regular property as well as mineral leasing. The amount of records for one sublease are simply staggering. We are currently doing a digitizing and OCR of historical well records from 1980 to current and the amount of paperwork fills a 40x30 storage building ceiling to floor. This is only for a few thousand wells. I've seen title searches for large families where there wasn't a will set up end up with a sack of near 1000 printed pages. Detangling property rights in areas with a lot of history, and mineral rights separated from land rights is a damned nightmare.
So you can't figure out just how much land the county commissioners wife really owns...
Literally. And if you should attempt to disambiguate all the trusts and LLCs and special purpose vehicles to figure out who really owns what in your county; you can expect a visit from law enforcement.
The US may look clean and consistently score low on measures of corruption; but it's there, especially at the county level.
I think it's a bit far to say that most counties are deliberately obfuscating data. That said, many counties, such as the one I live in, run on a shoestring budget and can't keep up with the current request load - the overhead for modernization or even corrections doesn't seem to exist, and the whole situation is made worse by the standard "pace of government." There's nothing like local politics to delay any kind of large-scale project.
The end result is that I wouldn't be surprised if, nationally, enormous binders are more common than GIS. The property records in my county are in the form of yellowing pages in 20-pound leather post binders on those rolling high-density shelves. People complaining about 2 minute query times? Here you drive to the county clerk's office, ask them your question, and they might be able to tell you which of the giant binders to look in first. The upside is that if you ask nicely they'll let you use their photocopier.
Perhaps we're unusually backwards, this is New Mexico after all, but we're not the smallest county in this state by far - and certainly not in the nation.
#1 Children's Hospital Los Angeles: Non-profit teaching hospital.
#2 Kaiser Foundation Hospitals: Non-profit hospital
#3 Cedars-Sinai Medical Center: Non-profit hospital
Are non-profit hospitals property-tax exempt? I know non-profits normally aren't, but there are exceptions (e.g., low-income student housing co-ops in Texas).
Yes. You pay taxes on profits. Since a non-profit entity makes no profit, it pays no taxes.
The bigger issue I come across is whether a donation to a non-profit organization is tax deductible. For that, the non-profit entity must apply to the IRS for IRC 501(c)(3) status and receive it before a donation to a non-profit entity is tax deductible.
Property tax is a tax on property, not on profits. Even a business that's losing money (negative profit) has to pay property tax, if it owns property. Some states do exempt charities (or some types of charities) from property taxes as well, but it doesn't follow automatically from non-profit status.
>Since a non-profit entity makes no profit, it pays no taxes
Is this true in the strictest sense? I mean I've seen a lot of non-profits make more money than they spend, but they're somewhat obligated to reinvest the money next year.
My impression was that non-profit was sort of a "pinky promise" about how you'll use your profits (with the IRS flying in if this is not the case)
That's true. If profit is defined as income less expenses, some orgs may have more income than expenses at the end of the year, which could be considered "profit". However, a non-profit is organized in such a way as there are no shareholders or other owners to distribute the profit. So, the precise term should be non-profit-distribution, I guess. Of course, the principals of a non-profit still take a salary, and you could probably sneak in a reasonable bonus as well, and not call it a profit distribution.
That table of most expensive cities is interesting. Given how Proposition 13 works, the assessed value of places in various cities might be more a reflection of how actively things are turning over than how much they are actually worth. The old house we rent in Manhattan Beach has an assessed value of like $88,000 since the landlord has owned it forever but if we bought it the value would jump to like $1.7 million. It would be cool to integrate this data with estimated house values from Zillow to see how much that skew is per city.
You could just to a bileaner filtered surface over recently sold houses in your town to more or less fake the same result. Just subtract the assessed value from the height of the approximate surface.
Fascinating article. I'm no GIS wonk, but I've moved in real estate circles for a while and understand how messy (truly truly messy) property data can be. If Rob wants parcel data across the country, he can pay for it, otherwise its a standartization nightmare.
I also love that this is a subtle pitch for a chief database architect (with 3-5 years of experience(?)).
I've built GIS databases w/ parcel data, owner data, tax data and zoning layers of high growth Virgnia and North Carolina jurisdiction. It's a very time consuming and ad hoc process. Some jurisdictions charge a lot of money for this data. If anyone has any interest in exploring a parcel based CRM startup concept, contact me. There is a lot of demand for this service. I have a few paying consulting customers and thinking about building it into a SaaS tool to power real estate development.
It shows that long term planning will indeed win out in the long term.
Meaning, if the extremist privatization politicians got their way, the gov't would own zero. But, if the gov't holds the commons in receivership over a very long term, the commons (as the gov't is ours if done well) will still hold a strong position over time.
I heard the largest landholder, once you strip back the entities, is actually the Church of Scientology. Maybe that was just hearsay, or only visible in Hollywood, where it's almost certainly true.
Your statement, which is just accepting an internet-rumor as true, contrasts very strongly with the article itself, which is about going to great effort to learn things from hard data.
I don't see how the article addresses this point at all. The owner_name is simply text in a field, and could represent a subsidiary of another entity, a front company or an LLC established simply to own that piece of property (common in California business real estate).
Even a simple typo or difference in conventions at the time of data entry will obfuscate that multiple parcels are owned by the same entity.
The article's method simply performs exact matches on owner_name, which is a place to start. It makes no attempt to delve deeper to piece together the true ownership structures that this data represents, an admittedly far harder problem.
A surprising number of homes in my LA neighborhood are owned by trusts. I used to assume they were trusts for the people living in them, but then I noticed many of the trusts own a number of properties. There's probably an interesting story or three to be told there with some data digging.
My understanding is that the smart way to sell real estate in California is to simply not sell it, and rather to sell control of the company that owns the real estate.
There are non-tax benefits, such as bypassing probate, but as the value of a property increases the 1% property tax becomes increasingly worrying. I suspect this is part of the reason why it's more common with commercial real estate.
The goal is to avoid tax reassessment when a property is sold, which is a big deal since the average price of a house in CA in 1940 was $36,700 and is now $211,500 (and much, much higher in places like Los Angeles and San Francisco). Even at the average house price, it's the difference between a $367/yr bill versus a $2,115/yr bill. Presumably commercial real estate has seen an equivalent rise over time.
I have no need to become an expert on California property law, but the exclusions at your link are very narrow. This related page discusses the transfer of ownership interest in a legal entity that owns California real property. triggering a reassessment.
If the originator of the trust maintains a controlling ownership in the trust, no sale has taken place. If they transfer control of the trust, the page I linked says that triggers a reassessment.
There are certainly cases where churches own a vast portion of urban real estate, all tax-free. I'd be amazed if churches didn't own a plurality of the land in the greater Oklahoma City area, for instance. It's funny how "The power to tax is the power to destroy" doesn't apply to any of the other entities mentioned in the First Amendment.
Scientology is probably too small to own much of LA. Clearwater, FL, on the other hand, may be a different story.
It was easy to find... but apparently will cost me $1000 dollars to access it...
Perhaps I don't understand exactly how the Assessor's office works, but it seems pretty wrong that public data from an organization funded by taxpayers isn't freely available for download.
The article doesn't mention it, but it looks like LA County charges as well.