Spanish Cadastre import

From OpenStreetMap Wiki
Jump to navigation Jump to search

We will try to summarize the steps followed by the Spanish community in order to import the Cadastre data. We will follow the structure of the Import/Guidelines. We want to STRESS that Cat2Osm2 is going to be a tool to help mappers in mapping of places with little data or to improve, with very specific data, places which are already well covered. The objective of the tool cat2osm will never be to perform the actual data uploads.

Create a community

The idea of importing the Spanish Cadastre data arises with the publication of the license and access criteria to the massive download of data from Spanish Cadastre on March 2011. Since then, we've had several meetings around this idea and a tool (Cat2Osm2) is being developed for the import. For instance: Jornadas_OpenStreetMap_Deusto_2011, VII Jornadas SIG Libre where we had a workshop with 30-40 attendants (including several enterprise representatives). During the first Night of the Living Maps we also had several meetings to test the script.

There is already a community of around 20 active persons actively following the discussion in talk-es list, testing Cat2Osm2 and giving feedback about it, as well as collaborating in the edition of the wiki. We know there are more silent listeners, waiting for the guidelines to be finished and to participate in the upload.

Make sure data license is OK

We have already discussed this with a lawyer.

Discuss import with community and Document your import on the wiki

As we have said we have had already 2 big meetings and several VERY large discussion threads in talk-es (around 500 mails). We had to create a specific tag for the discussion because we have increased the normal traffic in the list in more than 300%. See CATASTRO tag in talk-es mail list or Herramienta de fusión de los datos actuales con el catastro thread for example. Moreover, we have documented several wiki pages summarizing the discussion in the list (Spanish only).

The code of the scripts is in github with contribution from several people already.

Use a dedicated user account

We have already created them. We have created one account by province (50 in total). We send user and password to the person (or persons) that will be in charge of the uploading process. See Spanish_Cadastre/results for more datils.

Define your own tag prefix

We have already created one (catastro:), but it will only be used for the cadastre reference. The rest of the metadata will be using standard tags. We have followed these rules:

  • If a metadata did not have ALREADY a tag for it, this metadata is of not use.
  • If a geometry did not have any useful metadata this geometry will not be created.

Keep server resources in mind

We are discussing in the import list how to proceed but our estimations are that it will take SEVERAL years to finish even ONE province and there are 50 of them in Spain. The data have been splitt in small chunks (in the order of a block) in order to be easilly modificable and uploaded.

Don't screw up the data!

The data is not intended to be uploaded without being contrasted with existing data. There will be zones that would be easy (because there is no data) and in other zones there wouldn't be ANY import at all, because the data is good enough. The final decission will be taken BY THE LOCAL community that is ALREADY responsible for their part of the map.

WE NEVER EVER will import any data over the top of existing data. ALL that we are doing is PROVIDING a tool that the LOCAL COMMUNITY can use to speed up the creation of data or completing the existing data. The primary goal of this tool is provide a VERY GOOD starting point for places where there wasn't any or little data.

'We want to insists on this, this is not an automatic import tool!!!'

By the way, the authors of Cat2Osm2 think that the actual editors should change to a more context based approach where they only show the tools and data necessary to resolve the task and not ALL the data and ALL the tools. The iD editor its a good start but we think that a context base flow-work would be better suited for the task as this could lead to a editor more tailored to mobile devices with a tactics interface. We are currently talking to other groups in the DeustoTech, University of Deusto as well as to other people in the Spanish list to make such a tool. We will discuss the details shortly in the josm dev list or other places.

Consider simplifying

The program is already simplifying the geometry and the tags. Anyway, the data would have to be checked manually in JOSM before further action. Moreover, we can only export parts of the data (by year, by type, etc). Every municipality will be discussed BY the local comunity to decide what type of information will be of interest. There is a wikipage per province with tables to summarize the state in Spanish_Cadastre/results.