[Tagging] Development of Tag Replacements (was: Feature Proposal - RFC - landuse bush)

Vincent van Duijnhoven vincenttemp at vanduijnhoven.xyz
Thu Feb 4 08:15:50 UTC 2021


I personally also feel more for using and extending the landcover tag. It already exists so why introduce a new one. The question will be what "master" tag should be used or introduced in combination with the landcover tag.

I think barrier=hedge + landcover=scrub would be fine. 

For the other, would landuse=greenery + landcover=* be sufficient? If it would be introduced as value for natural=*, a more specific value like urban_greenery (or with a proper alternative for the word urban) would be needed. This to avoid confusing with the other values of natural

4 feb. 2021 08:47 van mail at jeroenhoek.nl:

> On 04-02-2021 06:23, Brian M. Sperlongano wrote:
>
>> Yup, I think you've hit the nail on the head.
>>
>> IMO, the only way this could be solved is with a technical solution. 
>> There would need to be a system of backwards compatibility in which
>> existing clients can continue to access the data using legacy tagging
>> schemes, with a stated end of life date, and clients could choose to
>> upgrade on their own timeline to the newer scheme.  This implies a layer
>> of translation that would need to exist between the underlying data and
>> what is actually presented to the user.  Now obviously this is glossing
>> over A LOT of technical details about all of the tools in the ecosystem
>> and how users interact with the data.  But I don't see any other way we
>> could have a systematic mechanism for tagging standardization (if indeed
>> the community even desires it) without some kind of backwards
>> compatibility translation layer.
>>
>
> I don't think it needs to be that complex. We know that landuse=grass,
> natural=scrub, and landuse=forest will remain abused for a long while.
> But a better alternative for the use case of mapping landscaping can
> gradually become the default if a few concerns are met:
>
> * Solid proposal with community backing
> * Good documentation
> * Rendering support in Carto, JOSM, and iD
> * Presets in JOSM and iD
>
> Mappers will follow soon enough if that foundation is there.
>
> The landcover-tag has a good head start already in terms of use, and
> should be at the heart of such a proposal:
>
> landcover=grass
> landcover=trees
> landcover=scrub
> landcover=flowerbed
>
> And perhaps landcover=scrub + barrier=hedge may then be acceptable for
> hedges mapped as areas?
>
> At nearly 30.000 uses for landcover=grass and 160.000 for
> landcover=trees, can rendering in Carto be on the table by this point?
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20210204/af21cd54/attachment.htm>


More information about the Tagging mailing list