On 2020-02-27 10:04, Mateusz Konieczny via Tagging wrote:
> 27 Feb 2020, 09:55 by colin.sm...@xs4all.nl:
>
>> If it was semantically important, we should be scanning for and flagging up
>> waterways with out-of-order ways.The fact that we are not, shows that the
>> ordering of the ways is not essential for a correct geometrical
>> interpretation....QED
>
> This proof assumes that OSM validators
> are completed and detecting all possible
> semantic issues.
>
> This is not true, just look at
> JOSM/osmose/iD issue tracker, every
> one has plenty of open issues suggesting
> new validator checks.
I'm sure there are thousands of ideas for validators. But this one has
apparently never been thought important enough - if it is indeed buried
in the issue trackers somewhere. Anyway, the validators should ideally
follow the agreed rules, not create them.
>> it is possible that re-ordering the members of one relation, causes the
>> members of another relation to become out-of-order.
>
> Why reordering members in one relation would
> reorder members of other relation?
You are right, I retract this statement. Sorry, I hadn't thought it
through and was thinking of the order of the nodes in a way.
_______________________________________________
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging
