[OSM-talk] Mappers and apps should focus on relations at the very start

[Simon Poole]

Am 28.06.2015 um 08:34 schrieb Jo:
...
> 
> As far as the datatypes go, I would be all in favor for the area
> datatype. I hear a lot of talk about it, and I don't understand why it
> doesn't materialise. While we're at it, we should 'formalise' a few more
> of the things we now use relations for.
...

Because it is not just a question of hand waving?

On the one hand there's the task of finding a way of modelling areas
that is net an improvement over what we have today. We have lots of
proposals that improve certain aspects of our current model, but at the
cost of worse behaviour for other aspects. I yet have to see a proposal
that is convincing enough overall*.

On the other hand, actually migrating to a new area type is going to be
a major undertaking in itself and my gut feeling is that it will be the
only dev thing we will be working on for at least a year once we start
and there will be a lot of stuff that will never be migrated. And that
all without actual gain in what we can model in OSM.

Simon

* note: for large areas I would consider Zveriks area references a good
solution that would generalize and replace the current magic tags
(coastline and others) and could be implemented without actually
introducing an area type.



-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20150628/d6abd669/attachment.sig>