[OSM-talk] OSM tagging validation lib
john whelan
jwhelan0112 at gmail.com
Sun Dec 24 19:31:21 UTC 2017
The problem with HOT inspired maperthons is half the participants will not
return. So restricting their choices to the most common is helpful.
Typically I see highway=primary between two small settlements,
highway=living_steet by mappers mapping from imagery in African countries
that do not have any in legal existence.
Whilst there is a mechanism of validation to correct errors it's only very
recently that it has been possible to restrict validation to mappers with
some experience. Even today with TM3 I can see tiles have been validated
by mappers with very little or even no experience and it is not possible to
identify the tiles easily to revalidate them and to be honest not that much
HOT mapping gets validated.
Even with mappers who use JOSM and do not map through HOT there are
problems. One cheerful mapper managed to upload over 400 untagged ways
with JOSM and they would have been warned when uploading them.
It isn't possible to make things idiot proof new improved idiots arrive all
the time but by suggesting the most common tags to new mappers might make
it easier for them to select the correct values.
Cheerio John
On 24 Dec 2017 1:51 pm, "Andy Townsend" <ajt1047 at gmail.com> wrote:
> On 24/12/17 18:41, john whelan wrote:
>
>> True but it's to do with data quality and avoiding the need to validate.
>>
>
> I don't think that different presets will "avoid the need to validate" in
> any scenario (HOT or otherwise) - the only thing that will do that is
> education and training - and having those educated and trained people come
> back again to do more mapping.
>
> Best Regards,
>
> Andy
>
>
> _______________________________________________
> 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/20171224/dbf149a4/attachment-0001.html>
More information about the talk
mailing list