[Tagging] New tag proposal: 'addr=milestone'

Paul Allen pla16021 at gmail.com
Wed Oct 2 11:32:00 UTC 2019


On Wed, 2 Oct 2019 at 03:40, Jorge Aguirre <jorge.aguirre at kaart.com> wrote:

>
> In all fairness, I think it should not be as difficult to find a good way
> to facilitate entering a complementary address tag, one that is very much
> needed in our part of the world - one which applies to and needed in most
> of the world actually.
>

I haven't seen anybody say such a tag is not needed.  I've seen people in
various parts of the
world say a similar situation applies there.  What people are saying is
that your proposed tag
is not well named.

>
> > I think we're close to hitting the record for how misleading a tag name
> can
> > be.
>
> Getting into analyzing the true definition of an actual ‘milestone’ - in
> this case - is needless.


One of the main reasons this list exists is to try to discourage tags with
misleading names,
because they end up being misunderstood and misused.  The true definition
of milestone,
as it is understood outside of OSM, is VERY relevant: people new to OSM
will interpret
the names of tags according to their common meanings, not OSM "this doesn't
have
the same meaning as in ordinary life" meanings.


> I feel we cannot all become ‘purist’ and try to find the proper
> definitions for terms to be then used as ’universal tags' applicable to the
> entire world.


And yet we must, else we end up with tags that are misunderstood and
misused.  And tags
that mean one thing in Brazil, a different thing in France, and something
else in China.  OSM
is a map of the world, not a collection of country maps.

The best we can all do is adapt and apply what we have at hand.


NO.  NO, NO, NO, NO, NO.  That's how we end up with bad tags.  Like
highway=milestone.
Outside of OSM, a milestone is a stone with a distance in miles marked upon
it.  Outside
of OSM, this is a milestone:
https://en.wikipedia.org/wiki/File:Milestone@Penrhosgarnedd_2.jpg
It's made of stone and the distances are in miles.  That's why it's called
a milestone and not
a kilometremarker.  But when somebody wanted to tag road markings that were
in kilometres
and not on stones, he or she decided to "adapt and apply what we have at
hand."


> So, things may not be perfect, but they are good enough to use and and
> apply to many different
>
cases.


MIsapplying tags to different cases leads to confusion and errors.  If the
cases are different
they need different tags.  If the same tag really does apply to both then
they aren't different.
Your definition of "good enough" is a lot less stringent than mine.


> Most people have come to appreciate the simplicity of using and how
> versatile the OSM project
>
really is.
>

Except that OSM isn't as simple as it could be because people keep coming
up with bad
names for tags.  OSM is stuck with landuse=grass when it should be
landcover=grass
because of a past bad decision.  Because of that past bad decision, people
invent other
landuse tags that should be landcover tags based upon the fact that
landuse=grass exists.
And somebody else insists on addr:milestone because of a past bad decision
about
highway=milestone.

The existent tag known as ‘highway:milestone’ and it’s definition found
> here:


If your friend put his hand in the fire, would you do the same thing?
highway=milestone is
a badly-named tag.


> [https://wiki.openstreetmap.org/wiki/Tag:highway%3Dmilestone] has been
> previously created, accepted and is currently being used by the entire OSM
> community. The point I have based this ‘addr:milestone’ proposal on, is
> what the general concept of the ‘:milestone' stands for: a location marker
> - which may be made of any material including ’stone’ - and which it’s
> primary function is to indicate a location on a road - which coincidentally
> can match (or not match) the distance - in kilometers (or miles), as these
> are internationally accepted measuring units mostly useful for these longer
> distances. But the ‘milestone’ as such, is still only a reference point on
> any given road.
>

Then propose addr:road_marker if, as you say later, that component of the
address is only loosely
based upon the position along the road and the marker is not at the true
distance it claims to
represent.

>
> The ‘milestone’ markers usually indicate a location - and not necessarily
> the distance - but in either case are used as reference points.
>

Wow, that makes them really useful.  It's a road marker that's not made of
stone, shows kilometres
rather than miles, and the number shown is wrong.  So you want to call it a
milestone.

>
> Unfortunately, DISTANCE could require being too exact in a very subjective
> and too ambiguous ‘milestone’ related addressing system.  The concept of
> either of these is not and cannot be an exact science. Most any known
> address system I’ve heard of is just based on proximity to other known
> references.


Are you deliberately being obtuse?  It has already been stated here that a
house between
marker 9 and marker 10 might have "9.5" in the address because it is a
distance of
APPROXIMATELY 9.5 km from some starting location.  But you say distance is
too exact,
even though that number is an approximation of a distance.  Nobody
suggested that you
measure it exactly and make the address 9.4.76638 rather than 9.5; instead
they're saying
that you should not use the term "milestone" but something like
addr:distance or
addr:road_marker  or whatever, because there are no milestones, as the
world outside of one
OSM wiki page understands them, in Guatemala.


> A tag name using the term ‘milestone’ to locate an address - as opposed to
> using a tag with the term ‘distance’ - in which that distance is measured
> from a ‘milestone’ (which may or may not be even there) makes no sense to
> me.


Hahahahahaha.  So your addr:milestone refers to a thing that isn't a stone,
shows kilometres
rather than miles, shows kilometres from an unspecified location (unlike
real milestones,
which show distance to a specified location), and may not even be there,
but you're fixated
on calling this thing a milestone.

As occurs in real life, these distances are not even close to being exact
> on any addressing system - they are only approximate distances from a
> reference point and usually expressed in round one (with any luck 2)
> decimal numbers - give or take a few hundred meters…
>

You are completely missing the point.  You're arguing against things that
nobody has said or
even implied.  You have constructed enough straw men to form a straw army.
The only precision
anyone is concerned with is precision in language.


> If the tag name [addr:milestone] I have suggested to use for this concept
> is not liked or is in any way unacceptable for use is just a side-tracked
> problem to be resolved.


Nobody is saying a tag is not needed, or that it's not useful, just that
you have chosen a bad
name for it.


> - The whole purpose of it all is to find a tag with which anyone can
> easily enter an address, including (or not) the ‘Km/Mi’ nomenclature it
> contains into the OSM data in an orderly fashion.
>

And that the tag is not confusing.

-- 
Paul
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/tagging/attachments/20191002/615d5cfd/attachment.html>


More information about the Tagging mailing list