[OSM-talk] Good practice, and should we rely on defaults?

[Simon Poole]

Am 06.04.2022 um 19:57 schrieb Yves via talk:

> ,,
> All in all I'd prefer explicit mapping.
> ..

Could you provide us with an example of a road that you consider to be 
completely explicitly tagged?

> Regards,
> Yves
>
> Le 6 avril 2022 13:40:02 GMT+02:00, Martin Koppenhoefer 
> <dieterdreist at gmail.com> a écrit :
>
>     I just noticed a recent edit to the good practice page and it made
>     me reflect about the current consensus.
>
>     https://wiki.openstreetmap.org/w/index.php?title=Good_practice&diff=next&oldid=2276559
>     <https://wiki.openstreetmap.org/w/index.php?title=Good_practice&diff=next&oldid=2276559>
>
>     The edit was surely executed in good faith, and while I generally
>     agree, it touches a sensitive topic.
>     The changes to the title seem ok, loosing "your" is probably an
>     improvement, while saying that "objects in reality" means
>     "specific objects" is probably the same.
>
>     Regarding the U-Turn-Restrictions in Brazil, it may be fine to
>     encourage not mapping them explicitly (it is already on the
>     restrictions page), but there is a slippery slope (in this
>     instance it seems ok, because mapping restrictions is not only
>     tedious and the precious mapper time could be spent better,
>     relations also increase the data complexity for everyone and make
>     the following edits more complicated).
>     On the other hand, from a legal point of view the situation is no
>     different from the absence of maxspeed signs in the second
>     example, where we do map implicit limits, or similar to the turn
>     restrictions we map because of specific divider markings (and in
>     some places you may even cross a continuous divider line to turn
>     left into a property). There is a specific object (an
>     intersection) because of which the U-turn is forbidden.
>
>     For example, highway=pedestrian (at least if signed as a
>     pedestrian zone with no additional signs), may allow cyclists to
>     use the road, or may not, it depends on the country. Generally we
>     rely on people "translating" the local rules to normalized
>     universal tagging (ideally, in reality there are local tagging
>     preferences to model the same situation). While we could leave
>     this all to the routing engines and defaults, it typically
>     produces bad results, so that mappers who want to use the data
>     now, are adding tags to describe explicitly what might in theory
>     be resolved by encoding language and region specific rules.
>
>     For example if a country has a rule, no spitting on the street, we
>     should not add this to every street. Or "no stealing" to a shop
>     (potentially any shop).
>
>     The second example is somehow contradicting the rule i.e. it
>     weakens the rule by showing an example where it is not applied, so
>     that's ok as it adds relativity ;-)
>
>     What are your thoughts about this rule, what is it meant to
>     prevent, and what is beyond the intentions?
>
>     Cheers,
>     Martin
>
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20220406/1411a09c/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: OpenPGP_signature
Type: application/pgp-signature
Size: 495 bytes
Desc: OpenPGP digital signature
URL: <http://lists.openstreetmap.org/pipermail/talk/attachments/20220406/1411a09c/attachment-0001.sig>