I do not recommend trail visibility in a case like this. I think it is
meant for real, usable trails, that just happen to be hard to see on
the ground. To use it in this case, is almost troll tagging. Basically
you are saying: there is a path here, but it isn't actually a path.
less advanced data-users (i.e. almost any app) will not show it any
differently.
This is why I would recommend the lifecycle tags. If the official path
is still there, but it is just razed by the agricultural works, you
can use razed:highway. If remnants of the path are still there, you
could use disused:highway. If the situation persists for a longer
time, it might be best to delete it all. But as you said above:
strictly "mapping what's there" means you should delete and remap a
lot of paths a few times every year. This is what creates clutter and
makes the data less readable to me!
The use of lifecycle tags implies that a data user has to know about
this stuff, and hence it is a concious choice, not an accident, to
show them to the data user. Osmand for example does this. Which I
really like, because I like exploring the woods. It would be
relatively easy to make a "switch" for Osmand rendering to show or not
show "lifecycle related" stuff. If you're not interested in ruined
buildings, future bridges or disappeared paths, you can swithc them
all off :)
If used reasonably (as in "oh this is weird situation, how should I
map it", not "I have an old atlas, let's map ALL of the disappeared
paths"), then I don't see how it "clutters" the map. And even where it
does, people don't seem to mind much anyway. Check out:
https://www.mapcontrib.xyz/t/6d1770-Trage_wegen_als_Note
It shows a bunch of ways with no properties except a note="some
buurtweg here". I shared it a few times here, and nobody bothered to
delete or fix them...
Op wo 21 aug. 2019 om 13:57 schreef Francois Gerin
<francois.ge...@gmail.com <mailto:francois.ge...@gmail.com>>:
Thanks for the comments, it confirms that it was relevant to share
on this.
It's already time to share a little more on my own conclusions then.
@Marc Marc:
Thanks for using option 3. The global/general idea to map only the
reality is good and important, but what appears a contradiction
here is not, IMHO. (See here below.)
PS: You're right for the highway path/footway. I fully agree, and
this what I do in my area. But in the area of the example, another
habit is in place... So I respected it. This is another issue,
which is a consequence of the French translation of the web editor
menus, according to me.
Thanks for the comment on the description tag, good point, I'll
add it to the other case.
@Marc Gemis:
I fully agree with the general rule "map the existing" and was
applying it in these cases too until recently. In fact, this is
the reason of my mail... I extend on this here below.
Thanks for the disused tag, I missed it. It will be useful in some
other cases, but here it cannot apply. (See here below.)
I'm contributing also to balnam, which is an organization that
monitors those paths and footways, which is absolutely not the
same purpose as OSM, and both are very useful, each one in its
area. Also, the official administration in charge of this
monitoring is so slow (years!) than the life cycles with OSM would
result in a complete mess.
Also there are several administrations for several purposes, and
quite inefficient in many ways, even if some have real good intents.
@Tim Couwelier:
Indeed the user's perspective is critical, and this is part of the
various items I integrated in my own analysis of this issue.
Thanks for the confirmation, it also goes in the direction I
expected. But this is more related to the rendering than the data
itself.
So, since we "agree", a little more from my own conclusions...
- Yes, I fully consent to the "map the current reality" approach.
And in fact, this is what I was doing before I had to reconsider
my way of thinking and finally change my mind. This rule must be
kept as the main lead. However, like all rules, especially the
"global" and "generic" ones, there are exceptions... And here it
is one that, IMHO, requires a specific attention, so as to
document it for the (probably many) contributers who face this.
- An important aspect, that is missed by the general rule and
fully part of the exception, is the timing: The path *appears and
disappears very periodically*, according to the cultures on the
field... If someone removes the path from the map, I'll add it
again soon after, when the path is back. This would lead to big
frustrations and/or litigations, as well as a lot of noise in the
database... Resulting in a situation that is negative for
everybody. (While having all the data in the DB and rendering
properly would lead to a positive situation fro everybody.)
- The comment from Tim about the users is particularly important,
but it is more a question of rendering than data in the DB. (That
was what I pointed to in my original message, "Side issue" note.)
A flag, being trail_visibility or another, makes it possible for
cheap, and it satisfies the software development rule "issues must
be solved at their root cause".
- We prefer not to add yet another tag just for this. The disused
tag does not match either, it would change every few months. The
trail_visibility much better matches matches the case, even if not
perfect... Think of a street closed periodically, here and then,
for the time a building (1-4 years) is made in a city. It would be
strange to see a tag "trail_*" for a street in a city.
=> This is just to mention that the notion is wider, I'm not
asking for a solution for this case, the solution of the
trail_visibility is just fine for me. But if something new has to
be made, probably it should be made generic enough to also cover
more generic cases. Maybe just adapt the trail_visibility to make
it more generic.
That's it for now on my side. And I guess sufficient to bring the
point to everyone...
While waiting for a possible other option/consensus, I'll continue
to proceed with solution 3, which is not contradicting the
important "map the real state" rule, according to me. It does not
contradict because the official way still exists in reality, even
if it is sometime hidden for a few weeks/months a year, in a
cyclical way.
Thanks for your participation and comments. If some have
meetings/discussion sessions, I think it would be a good topic...
Regards,
François
On 8/21/19 11:42 AM, Tim Couwelier wrote:
I'm with 'second marc' on this one - I chose to map ground truth.
In part because that's generally 'how things should be mapped',
in part because otherwise we receive criticism from avid users,
who are highly annoyed to get stuck / at dead ends because they
saw a path on their map and it's nowhere to be found.
While I fully support efforts to keep such paths functional /
accessible / known to the public, mapping them when they aren't
to be found in the field does not seem like the way go.
Op wo 21 aug. 2019 om 10:46 schreef Marc Gemis
<marc.ge...@gmail.com <mailto:marc.ge...@gmail.com>>:
Seems my opinion is different from the other Marc.
AFAIK, the OSM consensus is to map what is on the ground, in
this case
only the by-pass. You could keep the "official" path, with
some tag
disused:highway or so, but IMHO, that is just clutter that
makes it
harder for others to edit. When your local council does not
bother to
re-instantiate the official path, it will soon loose that
status, not?
As far as the removal of the "official" path is concerned, it
probably
depends on what "official" means. If it is e.g. in the Atlas der
Buurtwegen and was not officially removed by the council, you
should
contact your council and describe the problem. I did that
once and the
day after, the track was open to the public again.
On Tue, Aug 20, 2019 at 8:59 PM Francois Gerin
<francois.ge...@gmail.com <mailto:francois.ge...@gmail.com>>
wrote:
>
> Hi,
>
> Here is a probably subjective issue, that has certainly
already been
> discussed, but I cant' find a search engine for the mailing
archives.
>
> Problem:
> It's very frequent, in Belgium and certainly in many
places, that a
> private or farmer steals a footway because he dislikes
people pass there
> or just to extend his field for free.
> The **official** path is then often no more visible and,
sometime, there
> may have an **unofficial** by-pass in the area.
> The official trace MUST be kept because, well... it is
official. :-)
> And also because the by-pass MAY disappear at any time.
>
> Envisioned solutions:
> 1. Keep official path only. =bad because it does not
reflect the
> reality (which may stand for many years!)
> 2. Delete the official one, draw the by-pass. =rejected,
because the
> official must be kept, or we may loose both
> 3. Keep both, but flag the hidden one with trail_visibility
tag. =best
> option found up to now, which seems accepted widely+officially
>
> Questions:
> A. Is there any OSM consensus for a solution, at the
global/worldwide
> community level?
> B. If not, is there any Belgian community consensus?
> C. If not, is there any widely accepted option?
> D. If not, is there any better solution than option 3?
>
> (Side issue: the current rendering on OSM does not express
that this
> path is poorly visible. But at least the flag is there for
other
> rendering tools/layouts.)
>
> Two examples I had to do:
> https://www.openstreetmap.org/way/700172645
> https://www.openstreetmap.org/way/629096505
>
> Thank you in advance for any pointer/doc/wiki/consensus! :-)
>
> Regards,
> François
> (aka fgerin on OSM)
> (aka fge1 on balnam)
>
>
> _______________________________________________
> Talk-be mailing list
> Talk-be@openstreetmap.org <mailto:Talk-be@openstreetmap.org>
> https://lists.openstreetmap.org/listinfo/talk-be
_______________________________________________
Talk-be mailing list
Talk-be@openstreetmap.org <mailto:Talk-be@openstreetmap.org>
https://lists.openstreetmap.org/listinfo/talk-be
_______________________________________________
Talk-be mailing list
Talk-be@openstreetmap.org <mailto:Talk-be@openstreetmap.org>
https://lists.openstreetmap.org/listinfo/talk-be
_______________________________________________
Talk-be mailing list
Talk-be@openstreetmap.org <mailto:Talk-be@openstreetmap.org>
https://lists.openstreetmap.org/listinfo/talk-be
--
Joost Schouppe
OpenStreetMap <http://www.openstreetmap.org/user/joost%20schouppe/> |
Twitter <https://twitter.com/joostjakob> | LinkedIn
<https://www.linkedin.com/pub/joost-schouppe/48/939/603> | Meetup
<http://www.meetup.com/OpenStreetMap-Belgium/members/97979802/>
_______________________________________________
Talk-be mailing list
Talk-be@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-be
