[talk-ch] Fwd: Changes to HikingNetwork wiki page

René Buffat buffat at gmail.com
Wed Jul 14 18:39:09 CEST 2021


The email was blocked the first/second time as the attached screenshots
were too big. I put them in a google drive folder:
https://drive.google.com/drive/folders/1Q8pOvKHSfIXzZnoUIJ1MOokJMb96vw1c

----------

I see having a name / meaningful from/to tags as something for the next
mapper that looks at the data to help him understand what my intention was
when I mapped the relation. In the past I found it many times beneficial to
find this information.  The situation is a bit different if you can just
look into a GIS.

>From my experience, a requirement such as "Each way should appear in a
single hiking path network relation only" makes it near impossible to have
meaningful from / to tag.

As previously mentioned, there is another aspect of name/from/to tags.  The
name tag is displayed in iD, which makes mapping 1000x more usable. The
Rapid editor uses to/from tags if there is no name available.  See the
attached screenshots for the region Davos. If somebody uses one of these
editors in Graubünden, it will be really painful.

I understand that this is no concern for the power users that uses JOSM,
but for the majority of users that don't, this is an issue. As long that
there is not an improvement in the usability of iD (or iD catches up with
Rapid and there are widespread meaningful from/to tags), there will be
users that do things to make their life easier.

Personally, at the moment I do not really care if there is a name tag or
from/to. Both can be parsed by humans and there is no validation tool yet.
But as from/to tags could useful for algorithm checking consistency, I
would be more in favor of from/to tags (as soon as the issue with iD is
solved). Thus, I started to use from/to tags for all new relations I tag,
but didn't bother to update all relations I have mapped in the past.

lg rene

On Wed, 14 Jul 2021 at 17:41, Sarah Hoffmann <lonvia at denofr.de> wrote:

> On Wed, Jul 14, 2021 at 04:53:01PM +0200, René Buffat wrote:
> > > In beiden Beispielen muss man vor Ort gucken, wie die Markierung in der
> > Realität verläuft. Ob man das dann mit überlappenden Routen einträgt oder
> > nicht, ändert daran erst einmal nichts.
> >
> > Having relations with metadata allow to detect if there is something
> fishy
> > going on without being at the present location.
> > The elephant in the room is that the vast majority of the base network
> have
> > meaningful names. Which could be interpreted as that the community
> prefers
> > to have metadata of hiking relations.
>
> Niemand bezweifelt, dass es nützlich ist, grobe Angaben zu haben, von
> wo nach wo die Relation führt. Deshalb ja der Vorschlag mit den from und
> to-
> Tags, um das ganze noch etwas 'offizieller' zu machen.
>
> Wogegen ich mich verwehre ist, die Argumentation umzudrehen: weil from und
> to eingetragen werden sollen, wird daraus geschlossen, dass die Relationen
> an einem benannten Wegweiser anfangen und Ende müssen. Das ist einfach
> nicht
> so. Ich habe Dutzende Relationen gemappt, die nicht an benannten Wegweisern
> angefangen haben und trotzdem eine entsprechende note=<from>-<to> hatten.
> Das war nie ein Problem und hat der Nützlichkeit der Information keinen
> Abbruch getan. Das war übrigens auch der Grund, warum diese Information
> in den 'note'-Tag kam bevor hecktor alles umgetagt hat. Es war immer ganz
> klar eine Information für den Mapper und kein Name.
>
> Sarah
> _______________________________________________
> talk-ch mailing list
> talk-ch at openstreetmap.ch
> http://lists.openstreetmap.ch/mailman/listinfo/talk-ch
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.ch/pipermail/talk-ch/attachments/20210714/ea738ea4/attachment.htm>


More information about the talk-ch mailing list