The technology demo of Dutchstats as presented during Hack de Overheid last year has been a nice trigger for further development along that axis. For now the project under codename Dutchstats2 —a new name and identity is forthcoming— will be underway.
We got the announcement a couple of weeks ago that our proposal for subsidy had been accepted. We spent December gauging interest, taking in the project and building a team that can execute this in Q1 2011. Team introductions forthcoming after we’ve kicked it off.
The assignment is still the same one that prompted the original Dutchstats:
Given a set of values for a set of geographical regions visualize the mapping from the values to the regions in a way that is interesting, useful and pleasant.
Simple enough to be doable. Broad enough to be generally applicable.
The original Dutchstats was mainly concerned with Dutch municipalities as geographical regions and election results as values and we will be continuing along that line, but we will be looking into opening up both the values and also the geographical regions for anybody who has something to contribute to either. The idea is to create a generative atlas.
A generative atlas mostly to see if we can give the concept of an atlas new currency in the online world.
In the Netherlands there is an atlas called the Grote Bosatlas which still is the standard atlas for everybody in and out of school. But asking people around the question: when is the last time you have even thought of an atlas, let alone got and leaved through a Bosatlas, everybody draws a complete blank. Google Maps has supplanted most of the topographical and wayfinding functionality of paper maps and atlases to the extent that it has wiped out the original concept out of people’s heads.
The social geographical function of the atlas has been replaced by a ton of projects working either with or on Google Maps/Earth using GIS or placing points on the map (using location or geocoded data), Stamen’s Dotspotting is a good example of that. Besides those web centric approach there’s also a slew of closed/semi-closed mapping tools from statistical offices, government bodies etc. that are built on poor and closed technology and are limited to the task at hand (which they usually do poorly at that).
We’re going to determine as we go the technology that we’re going to use, but the project needs to be webcentric and is allowed to be bleeding edge (though perhaps not as bleeding as the original prototype) so I hope we can avoid using Flash completely.
Depending on how much of the base components are already available (data stores, tile servers, rendering engines), we will be focusing more on the application part. But if such components are not yet available or up to par, we will be investing in building them ourselves.
In our practice we believe in standing on the shoulders of giants, sharing alike and giving credit where credit is due. We will be doing this project completely in the open not because we don’t have a customer for it but because everybody is a potential customer and they should be able to see and participate from the earliest stages on.
Any software that we produce will be released under a very liberal open source license. So that anybody can use our stuff and we hope to advance the state of mapping online in our own modest amount. Also all our design research and progress will be posted to this blog in chunks of a week or a bit more (depending on our sprints).
Fully open is the only way we can imagine doing this. We hope you will join us.