Hi Maarten,
At the time we though that we were using a unique dataset to import these
state boundaries but they did not match in some places. We were not able to
(mannualy) fix the divergences since then.
If you can help in fixing then you effort will be welcome.
thanks
Claudomiro Jr.
On Sat, Nov 6, 2010 at 2:46 PM, Ulf Mehlig wrote:
> Hello Maarten,
>
> at least in part the double borders are due to double import from IBGE
> (Brazilian Institute for Geography and Statistics) public domain data. I
> don't know details about the gml2osm import, but for Pará state, for
> example, imports where done independently by Claudomiro Nascimento
> Junior and me. I tried to fix a few things manually but (obviously) did
> not finish. A bot solution might be interesting, if manual changes to
> the border would be preserved. I'll forward this message to the
> Brazilian talk list.
>
> Thanks alot!
> Ulf
>
> em português:
>
>(Maarten Dean perguntou sobre fronteiras duplas no Brazil com
>fonte "IBGE", em parte com rótulo "criado por gml2osm". Esta
>última parece mais detalhada, mas não é utilizada em relações.
>Ele sugere solucionar o problema através de um programa que
>corrige os dados automaticamente (bot).)
>
> Pelo menos em parte as fronteiras duplas tem origem na importação dupla
> de dados em domínio público do IBGE. Não sei detalhes sobre a importação
> por gml2osm, mas pelo menos no estado do Pará, dados foram importados
> independentemente por Claudomiro e por mim. Tentei concertar umas coisas
> manualmente, mas obviamente não terminei. Uma solução automática poderia
> ser interessante, caso preservasse mudanças feitas manualmente. Estou
> mandando este email à lista brasileira do OSM.
>
> On Sat, 2010-11-06 at 16:45 +0100, Maarten Deen wrote:
> > I noticed that Brazil has a lot of double boundaries. You can see them on
> the
> > map, for instance the north-east point of Brazil (but it's everywhere).
> >
> > Looking at the data, both boundaries are source=IBGE, but there is an
> older,
> > mostly more detailed border with created_by=gml2osm which is not used in
> > relations and a newer, less detailed one that is used in relations.
> >
> > I don't know what happened here, but is there something which can be done
> by a
> > bot here, or is the older one to be deleted or should that one be used to
> make
> > the border relations?
> >
> > Maarten
> >
> > ___
> > talk mailing list
> > t...@openstreetmap.org
> > http://lists.openstreetmap.org/listinfo/talk
>
> --
> Ulf Mehlig
> --
>
>
> ___
> Talk-br mailing list
> Talk-br@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-br
>
___
Talk-br mailing list
Talk-br@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-br
