[OSM-talk] Global test area

80n 80n80n at gmail.com
Sat Dec 9 08:51:35 GMT 2006


I have a .osm file containing some rendering test cases.  It would be very
helpful to be able to load it into a database where it could be shared with
other users.  A separate test database would be my preferred choice for this
kind of thing.

Perhaps it should be named Second Planet ;-)

I think we should also, ultimately, have a Sandbox Planet where new users
can play on a copy (or subset) of the real thing without causing and
damage.  It could be refereshed every week to restore it to a decent state.

80n


On 12/9/06, Jochen Topf <jochen at remote.org> wrote:
>
> There are always issues like "how to best tag certain features", "how to
> best join those two streets when one is of type x and the other of type
> y",
> or "how does renderer x renders this map feature" etc. Currently there are
> some wiki pages addressing those issues, but their are hard to maintain,
> because it has to be done manually, including getting screen shots of
> relevant features from a renderer etc.
>
> To help with all those issues I propose we create a "global test area", an
> area on the map of the earth where we can put all these special (and
> normal) features in their context and in all variations necessary. This
> area could be located on the North pole or somewhere in an ocean where
> it doesn't collide with real features.
>
> Data for this area would be stored in the normal OSM database. But to
> get it out you would have to set a special flag which must be added to
> the API. If an application doesn't set that flag, it will get the normal
> view of the real earth. If id does set the flag, it will see the special
> area in addition to the normal view. So nothing would change for legacy
> clients, only clients supporting the new flag would see the data.
>
> The easiest way of adding this flag would be to use a different base URL
> for the map with the test data. That way no client changes are necessary,
> because in most clients the base URL is probably configurable already. It
> might still be better to have a button in the client which says "add test
> area" or so, but for the moment thats not strictly necessary.
>
> The database itself needs no changing, only on exporting the data, the
> flag has to be checked and the test area has to be clipped if the flag
> is not present.
>
> Another option would be to create a completely separate planet for this.
> That is certainly easier, because we only need to set up a separate
> database and no software changes are necessary.
>
> I am not sure whether the database integration with flag or the separate
> database would be better. The difference would be most relevant when
> you want to see both real and test data in the same application. I would
> probably want to look at them side-by-side. I don't know about other
> applications, but JOSM doesn't support different windows or views yet.
> Using a separate database would mean that you have to set a source for
> each window, with only one database this would not be needed.
>
> Still another option would be to just have an .osm file with all this data
> in it. This could be downloaded from the wiki. Advantage of this approach
> is that no new software or services would be needed. But the big
> disadvantage would be that we loose the interactive and collaborative way
> to edit the data, which we are used to.
>
> If such a thing would we deemed helpful and if somebody creates the
> database
> or puts support for this in the current one, I would volunteer to start
> populating that area with features. (Of course everybody can help doing
> that, thats the point of the whole exercise.)
>
> Jochen
> --
> Jochen Topf  jochen at remote.org  http://www.remote.org/jochen/
>   +49-721-388298
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20061209/5fc397c2/attachment.html>


More information about the talk mailing list