9,866,539 buildings in the Netherlands, shaded according to year of construction
dev.citysdk.waag.org
dev.citysdk.waag.org
Kind of gives you an idea of how large parts of it were destroyed during the second World War and then has had multiple levels of reconstruction and changes taking place throughout the decades since.
- 1738 - Vijfharingenstraat 5 6211EC Maastricht
- 2002 - Grote Staat 5 6211CT Maastricht
http://dev.citysdk.waag.org/buildings/?utm_source=buffer&utm...
(For non-Dutch people: Maastricht is one of the oldest continuously urban centers in the Netherlands, which makes it nice to show off this kind of technology but it's also very far away from the rest of the country. Nice city though, definitely worth a visit).
Basically whenever a house is sold, (I never actually bought a house so someone more knowledgeable should correct me) information about the purchase and the state of the house is submitted to the Kadaster. As far as I know the kadaster is something that has its roots in the middle ages, but I could be mistaken.
As for the digitization, that probably happened somewhere in the late 60's early 70's out of necessity along with many other governmental databases (like taxes).
What is news is that one apparently can query the entire kadaster database, this is something done by http://dev.citysdk.waag.org/, a European project so maybe we'll get info like this from other countries as well.
I know some municipalities maintain extensive GIS and financial datasets that include valuation, sale data, ownership, maps, etc. Here's an example from the Boston assessor's database: http://www.cityofboston.gov/assessing/search/?pid=0503500000
But I don't know if smaller/poorer towns have created or maintain digital archives, what standards are used, or whether data is shared at the county or state level. One question I have: Is this a state or local requirement that such records are created and posted online, and how does it differ from municipality to municipality and state to state?
I would be surprised if a federal agency maintained a national database like the Netherlands, although these days it's not really clear what data our government is gathering/sharing, and for what sorts of purposes.
The land registry is woefully incomplete. Things only get into the registry when there's a change of ownership, and some land has been in the same family since the Domesday book. This was a problem when I used to work on software processing planning applications as some councils thought they could use the land registry to validate addresses.
The planning applications had great data in them if it could be opened up (eg they are generally submitted with a property boundary map!) but generally the councils wanted to either monetize this data or not release it at all; so you just get pdfs&metadata out, even when the user submitted in a vector format.
2. 'Kadaster' is dutch for the English 'cadastre' (http://en.wikipedia.org/wiki/Cadastre)
3. The Dutch cadastre was initiated in 1811, by decree from Napoleon Bonaparte (the Netherlands was part France at the time). First ?complete? set was available in 1832 (http://www.kadaster.nl/web/Themas/themaartikel/dossierartike...; in Dutch)
No, instead we have a very limited series of copyright-protected maps by a company called Sanborn that was hired to do this style map from 1867 to 2007 for fire insurance companies - but not in every town and city, not regularly, and rarely for rural areas.
On top of that we have 3100 county-like bodies which will each tend to have a local plat map of who owns which properties (which will occasionally mark buildings), in varying stages of digitization and privatization (some of them charge ridiculous amounts, some are public access).
We also have an irregularly updated USGS quad dataset that is not very precise, and doesn't attempt either building shape or urban areas.
Lastly, there are modern aerial-collection LIDAR returns that tend to produce high accuracy building corner references.
Source: Dutch Wikipedia http://nl.wikipedia.org/wiki/Basisregistraties_Adressen_en_G...
I wonder why this map was done with a divergent scheme though. The data has no midpoint, it makes no sense to diverge like that. And when you zoom out it, due to the diverging lightness ramp it highlights the lightest colors which are 1945 to 1975.
It looks gorgeous but the color choice is not perfect for exploring the data.
When you click on a house, the click is converted to lat/lon and compared with the database to show which house you clicked.
> Originating as a small fishing village in the late 12th century, Amsterdam became one of the most important ports in the world during the Dutch Golden Age, a result of its innovative developments in trade.
So I doubt there are a lot of buildings from 700AD left. But I agree, they could've gone back to ~1500AD.
Still, building materials mean that buildings from before the 16th century are incredibly rare.
Alternatively, download the data from https://data.overheid.nl/data/dataset/basisregistratie-adres...; it is open data.
At night, from the ground: http://straatkaart.nl/2286JK-Albert-Schweitzerlaan/media_fot...
At night, from space: http://www.flickr.com/photos/astro_andre/6769497883/in/set-7... (the large bright spot are greenhouses with artificial lighting to speed up plant growth)
Which doesn't detract from the fact that this is a beautiful visualisation. And also I doubt the general public can query those data in the bulk amounts required for this map :)
But to properly display that is a real hard problem. First, you need way more data, a complete history of the changing shapes of buildings as they were being reconstructed/renewed/etc, over the ages. Second, how to visualize all of that? Layer all of it on top of eachother, with old at the bottom and newer shapes covering it, maybe? or ... stacked in 3D, like a geo/archeological excavation!
Right now it seems a bit of a compromise, mashing up two datasets that don't always properly "fit" together. But on the whole, on average, and for many aesthetic purposes, it does a very good job.
This is definitely a beautiful and useful visualisation and something we've been thinking about for a while too. We've found that in an average real estate search people very much appreciate information about the neighbourhoods they come across when finding relevant properties. The thing I love about this is -- like some others pointed out already about Rotterdam -- is that it tells a genuine story about a neighbourhoods history in a way a real estate agent wouldn't be able too.
In the US there's Trulia Hindsight built by Stamen Design which includes animation: http://hindsight.trulia.com
Fortunately, the site links to this page where you can report incorrect data: http://www.kadaster.nl/web/Themas/themaartikel/BAGartikel/BA...
It has uncovered some interesting issues with addresses - some student housing for instance has complicated numbering schemes that don't fit in the database (i.e 10 B 5). Municipalities that don't cooperate are cut national contributions.
There's a building older than my country sitting right next to one built while I was in college. The casual blend of old and new is one of the things I love about European culture.
I'm sure it's found in many other places on our humble planet, but I've only been to Europe (so far!).
>Building
>Constructed: 1300
>Address: Keizer Karelplein 6 6211TC Maastricht
>Area: 3,306 m²
>Function: Public
Here is the street view:
As well as the Netherlands map there is the Portland map, a map of Ljubljana, Slovenia (similar to the Portland map) and the Brooklyn map.
http://www.mapsmaniac.com/2013/08/view-age-of-10-million-glo...
http://bklynr.com/block-by-block-brooklyns-past-and-present/
I published some instructions on how to generate such a map using the recently released PLUTO data:
http://bdon.org/2013/06/29/working-with-pluto-and-pad/
To the best of my knowledge, San Francisco doesn't have a public building age dataset. There is a building footprint dataset on https://data.sfgov.org/, but the quality of the shapes isn't very good.