[Imports] Spanish Cadastre ELEMTEX
Cruz Enrique Borges
cruz.borges at deusto.es
Thu Jun 7 07:46:12 UTC 2012
Fist of all I want to notices that I'm one of the developers of the cat2osm
> - Some nodes have only name, source and source:date tags while others have
> those and place=locality
It is better to have these tags in all nodes? This can be made easily.
> - Some have absurd names (e.g. name=-I+I and name=P-1, P-2, etc). These
> could be dealt with manually but it would be worth seeing how common they
> are and dealing with them in the conversion
These are tags that SHOULD have been put in other places in the original
files. In the normal processes (not the discussed here) the tool try to
correct this types of mistakes and place them where they belongs (in this
case their denote the height of the construction and should be in the relation
with the tags bulding=yes) but as the type of mistakes vary enormously between
files it is very difficult. Anyway we will try to filter as much as we can,
and the rest will be manual corrected by the local member that want to import
> - Some of the names could be translated to tagging, e.g. name=GASOLINERA
> could be turned into amenity=fuel. Again, this would depend how common they
> are and how consistent they are.
We have currently planed something like that for street data that can be
extended to elemtex too. The idea is to use a external file with the
translations that can be update manually if a local mapper find a patter in
the files they are processing. We have a preliminary one in  but we have
find that it will not be easy as every file have they particularities.
> CanVec has convinced me that going this method with imports requires some
> sort of QA plan or one importer will not check any of their work and cause
> duplicates, disconnected data, bad data, etc.
> The problem is not that there aren't tools for checking your own work, the
> problem is that some people will not use them and cause significant damage.
> I can't suggest a solution for this, and I think it is not as significant
> for this particular data set, but it is definitely an issue when you start
> getting into others like roads and landuse.
> If you find a good solution for this problem, please document it and let
> everyone know as it is a big problem with imports done by multiple people
> who may have varying standards.
We have planed a review process of all data. It is documented in the wiki
(see for example ). I will post a summary here:
- We have created separate accounts for every province.
- For every province there will be a person in charge of distribute the
passwords and coordinate the actions (and possibly to upload the data).
- A web page it is also created with all the municipalities and info
about the local particularities that the locals mappers have agree.
For example, in some provinces there are discussion on what
information should be imported for every municipality and the like.
- Local mappers take a empty municipality and process with the cat2osm
tool and upload the results to the wiki page. Then it is mark that
municipality the (usually) in yellow in the map.
- Then, the output file is hand fixing by one or several local mappers.
The osm file at the moment is manually split but in the next version
of the tool we plan to automate this process. Then it is mark that
municipality (usually) in green in the map.
- Finally the corrected osm file is uploaded to the wiki waiting for review
of other mappers. Only after one (or several) person review the data,
the osm file is uploaded. We in the process of discussing how we can
upload the data in the less invasive way for the servers.
- We plan to be VERY strict with the QA process and will ask to BAN every
users that repetitively do not respect this process.
I hope this will clarify better the import and, of course, comments are
more than welcome for all the process.
Cruz Enrique Borges Hernández
Email: cruz.borges at deusto.es
Telefono: 944139000 ext.2052
Avda. Universidades, 24
48007 Bilbao, Spain
More information about the Imports