[Tagging] Route maintenance tagging

Mon Dec 31 12:57:29 UTC 2018

A quick progress report. The idea has taken on in Nederland. It appears to
be workable because most of the longer routes have been split into sections
of one day's walking.

The sectioning into relations of <= 300 ways had already been done to make
the shirtload of walking route relations manageable. The national operator
uses the same sectioning. This makes life so much easier for routemongers,
I would encourage all of them to put in the effort and section all the long
routes. Especially newcomers will welcome the smaller easy2handle chunks.
(The hierarchy is not complicated: the lowest level is a sack of ways,
which may be bundled in sacks of sacks, etcetera. A route change is made
only on the lowest level. That's it!)

Next step: creating a webpage which lists (a selection of ) walking route
relations sorted by survey:date ascending. Using an osm-query as data
source.

I'll need help for that, because I probably can put a sorteable wikitable
on a wikipage but I have no idea how to fill a table from a live query. I
know it can be done, because I see applications do things like that all the
time! I have no site and zero web programming skills. There, I said it!

Op do 19 jul. 2018 om 23:58 schreef Peter Elderson <pelderson at gmail.com>:

> We section long routes because it is very hard, if not impossible to keep
> big relations intact. Survey info would be entered on manageable sections.
>
> Mvg Peter Elderson
>
> > Op 19 jul. 2018 om 23:41 heeft Warin <61sundowner at gmail.com> het
> volgende geschreven:
> >
> > There is a national route near me.. it is some 5,000 km long. Not many
> walk/ride the entire length.
> >
> > It is not complete in OSM, nor upto date. But there is a route .. broken
> in places ...  I think of it as a guide rather than truth.
> >
> > The survey:date would have to be added to each way of the relation and
> be modified to survey:date:releationxxxx=*
> > Sorry but I'm not doing for the 100s of ways involved ... too much work.
> >
> >> On 20/07/18 07:06, Philip Barnes wrote:
> >>
> >>
> >>> On 19 July 2018 20:57:20 BST, Peter Elderson <pelderson at gmail.com>
> wrote:
> >>> Just saw https://wiki.openstreetmap.org/wiki/Key%3Asurvey%3Adate
> >>> Since survey:date is a documented tag, I will start using it to record
> >>> route survey dates.
> >>> Not on ways, but on sizeable hikingdare route relations.
> >> A date on a hiking route relation is likely to be meaningless for the
> reasons already mentioned by DaveF. That being they will rarely be walked
> in their entirety but mappers will do sections here and there.
> >>
> >> Using an example of my local long distance route, the Shropshire Way, I
> have systematically walked about half of it, so far that has taken over two
> years. Which date do I put on the relation?
> >>
> >> Phil (trigpoint)
> >>
> >>> See if I can get fellow mappers and walking route operators to join the
> >>> effort.
> >>>
> >>> 2018-07-19 18:39 GMT+02:00 Peter Elderson <pelderson at gmail.com>:
> >>>
> >>>> Thanks for the warning. Of course it is not the idea to delete
> >>> anything
> >>>> except when proven wrong.
> >>>> I meant: information from outside sources, such as gpx-trackings,
> >>> which
> >>>> are older then the last completed survey, should not be entered into
> >>> OSM.
> >>>> Also remember that I'm talking about route information, not mapped
> >>>> physical objects. We're not mapping individual waymarks, but routes
> >>>> indicated by waymarks. Even if you remove the route relation, nothing
> >>>> physical is taken from the map.
> >>>>
> >>>> The survey date is the key data element here, if any kind of
> >>> systematic
> >>>> maintenance to the route relations is setup. Will it take? I don't
> >>> know.
> >>>> We'll see. The check&maintenance system for cycle node network and
> >>> walking
> >>>> node networks (vmarc.be) works like a charm, so I have good hope)
> >>>>
> >>>> 2018-07-19 17:02 GMT+02:00 Kevin Kenny <kevin.b.kenny+osm at gmail.com>:
> >>>>
> >>>>> On Thu, Jul 19, 2018 at 7:22 AM Peter Elderson <pelderson at gmail.com>
> >>>>> wrote:
> >>>>>> The goal of the idea is to tag the date of the last reality check.
> >>> The
> >>>>> best thing I have now is the date of the last edit, which most of
> >>> the time
> >>>>> results from e.g. a mapper's action (cut or remove) on a way that's
> >>> part of
> >>>>> the route relation.
> >>>>>> I want to ensure that the route in the field and the route
> >>> relation
> >>>>> stay in sync, and when they don't (which is a 100% certainty) that
> >>> you can
> >>>>> tell at what point in time it did match.
> >>>>>> Information older than that date (e.g. gpx-tracks) can be
> >>> discarded,
> >>>>> newer information can be entered, and edits after the survey date
> >>> are new
> >>>>> info which should be kept.
> >>>>>
> >>>>> Keeping the field survey up to date is a laudable goal, and I've no
> >>>>> objection to some sort of tagging that reports "this geometry was
> >>>>> field surveyed on <date>." Making it fit with the data model will be
> >>>>> challenging; it's not something that can be easily automated, given
> >>>>> the variety of mappers' workflows.In the current world, to make
> >>>>> something like this a reality you have to have an individual or
> >>>>> organization that becomes the de facto 'owner' of the route and
> >>> keeps
> >>>>> track of its own surveys - and that isn't very OSMish. I think this
> >>>>> could be worked around with sufficient cleverness.
> >>>>>
> >>>>> But please, please, don't discard data older than a certain date.
> >>> OSM
> >>>>> is a very young project as geography goes. While out-of-date data
> >>> can
> >>>>> be misleading, the right thing to do is to inform, not to delete,
> >>>>> particularly in cases where the out-of-date information is the only
> >>>>> information that is available. It may also be the only information
> >>>>> that can guide in recovering from an act of vandalism or a
> >>>>> badly-considered import.
> >>>>>
> >>>>> Perhaps I'm coming at this from the 'wrong' perspective. since a
> >>> fair
> >>>>> amount of my mapping is of features that nobody has yet seen fit to
> >>>>> map at all, or that were once imported from external data that I
> >>>>> consider hallucinatory. If someone with a GPS found a route passable
> >>> a
> >>>>> decade ago, that's a piece of information that I now have that I
> >>>>> wouldn't have had otherwise. It could be that the route is no longer
> >>>>> passable, has been relocated, or has been demolished, but without
> >>> the
> >>>>> old data, what reason do I have even to go and find out?
> >>>>>
> >>>>> Moreover, the land remembers. I've been on trips where abandoned
> >>>>> tracks and the grades of dismantled railroads, a century old and now
> >>>>> grown to trees, have been important landmarks. I have no qualms
> >>> about
> >>>>> not showing them on a general-purpose map, but to an off-trail
> >>> hiker,
> >>>>> they are waymarks for eyes to see that can.
> >>>>>
> >>>>> The right thing to do with 'stale' data - perhaps even 'proven
> >>>>> incorrect' data - is to inform, not to discard.
> >>>>>
> >>>>> _______________________________________________
> >>>>> Tagging mailing list
> >>>>> Tagging at openstreetmap.org
> >>>>> https://lists.openstreetmap.org/listinfo/tagging
> >>>>>
> >>>>
> >>>>
> >>>> --
> >>>> Vr gr Peter Elderson
> >>>>
> >
> >
> > _______________________________________________
> > Tagging mailing list
> > Tagging at openstreetmap.org
> > https://lists.openstreetmap.org/listinfo/tagging
>

-- 
Vr gr Peter Elderson
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20181231/ad401edb/attachment.html>