Re: [Tagging] How to tag small canals?

[SelfishSeahorse]

Hi

I've written an issue request on openstreetmap-carto regarding the too
thick canal rendering:
https://github.com/gravitystorm/openstreetmap-carto/issues/3354

Regards
Markus
On Thu, 16 Aug 2018 at 18:20, Christoph Hormann  wrote:
>
> On Thursday 16 August 2018, Martin Koppenhoefer wrote:
> > >
> > > All of this together has its origin in the fact that in the UK and
> > > other early OSM countries large artificial waterways are almost
> > > always for navigation and small artificial waterways are almost
> > > always for transporting away undesirable water.
> >
> > +1
> > That’s my analysis as well. What do you think, shouldn’t we fix the
> > wiki to make it more universally applicable to all kinds of
> > waterways?
>
> As usual i don't think you can change the definition of tags that are
> already used hundreds of thousands of times except by extending their
> scope which would make the tags more vague than they already are.
>
> What would make sense is to explicitly mention that waterway=canal does
> not have a lower size.
>
> Inventing a separate tagging scheme for irrigation systems might be an
> option too but irrigation is performed in very different ways in
> different parts of the world so it might not be too easy to create an
> universal tagging system for that.
>
> > The canal definition was changed in March 2018, before it said to use
> > canal only for „the largest waterways created for irrigation
> > purposes“
>
> Yes, that was the obvious attempt to expand the narrow scheme to other
> parts of the world in a superficial way oriented at the standard style
> rendering but not at the actual semantics.
>
> Going this route would essentially mean turning all of ditch/drain/canal
> into one big uniform catch-all.  Removing the above is the attempt to
> salvage some of the semantic value in the data.  Not sure how
> successful that is without support from the standard style though.
>
> Unfortunately waterway=canal has only 13k combinations with width=* and
> only 1.8k combinations with usage=* at the moment.  Extending that
> would be the best way to move forward IMO.
>
> --
> Christoph Hormann
> http://www.imagico.de/
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Warin]

On 19/08/18 04:01, Martin Koppenhoefer wrote:



sent from a phone


On 17. Aug 2018, at 13:43, Christoph Hormann  wrote:

If you want to map the river width tag width=*, if you don't want to map
the width then don't create classes based on width thresholds.


river size is not about width alone, the most generic approach to measure size 
is looking at the quantity of water per time. It accounts for width, depth, 
flow speed in one parameter. For the significance, length is probably also 
interesting.
The width alone depends on factors like the terrain and its geology.

These numbers are most useful in relation to other waterways nearby. The 
context matters, the same amount of water in one area might be a stream and in 
another a river.

Usually at the confluence the name of the waterway with the most water is the 
one whose name prevails, but sometimes there are exceptions, e.g. for historic 
reasons (e.g. they didn’t know exactly which river brought more water, or 
considered one more important for other reasons)

For navigation it is interesting to know if you can cross the waterway without 
bridge, tunnel or ford by jumping over it (width) or by walking through it 
(depth).



Of course there are exceptions.
The Todd River in Alice Spring Australia has had 0 width and 0 flow every time 
I have been there.
And you can walk across it, camp in it .. play games in it. Most of the time.
Yet it is a river as can be seen by the width of the sandy river bed and the 
length of the bridges crossing it.
And when water does flow in it you don't want to be in it .. even in a boat.


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Martin Koppenhoefer]

sent from a phone

> On 17. Aug 2018, at 13:43, Christoph Hormann  wrote:
> 
> If you want to map the river width tag width=*, if you don't want to map 
> the width then don't create classes based on width thresholds.


river size is not about width alone, the most generic approach to measure size 
is looking at the quantity of water per time. It accounts for width, depth, 
flow speed in one parameter. For the significance, length is probably also 
interesting.
The width alone depends on factors like the terrain and its geology. 

These numbers are most useful in relation to other waterways nearby. The 
context matters, the same amount of water in one area might be a stream and in 
another a river.

Usually at the confluence the name of the waterway with the most water is the 
one whose name prevails, but sometimes there are exceptions, e.g. for historic 
reasons (e.g. they didn’t know exactly which river brought more water, or 
considered one more important for other reasons)

For navigation it is interesting to know if you can cross the waterway without 
bridge, tunnel or ford by jumping over it (width) or by walking through it 
(depth).


Cheers,
Martin


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Peter Elderson]

Without getting into scientific, expert or philosophic discussions, I think it 
would not hurt to have a standardized way of recording say a bandwidth for 
dimensions like width and ele. The example in this case was width=3 m - 6 m or 
so, with the assumpion this notation would not be usable to data users. I’m 
just saying that it would not be that hard to find a processable notation for 
this, and simply asked if it already exists or has been proposed. 

Agreeing on it is a different matter, especially in OSM circles.

Mvg Peter Elderson

> Op 18 aug. 2018 om 10:44 heeft Christoph Hormann  het 
> volgende geschreven:
> 
>> On Saturday 18 August 2018, Peter Elderson wrote:
>> Sure. But is there a standard method to indicate this uncertainty in
>> OSM, which can be processed by data consumers?
> 
> This is outside the scope of of OSM IMO and not practical for mappers to 
> determine in a meaningful way.  You can specify the method used to 
> determine the width (using source:width=*) and different methods will 
> have different inherent measurement accuracies but the accuracy of the 
> values depends on lots of other factors beyond that.
> 
> Practically mappers have usually no way to verifiably determine the 
> accuracy of their measurements unless they have a reference value of 
> known high accuracy in which case the whole excercise is pointless.  
> Theoretically mappers could practice their measurement skills and 
> assess their personal measurement accuracy based on test measuring a 
> statistically significant number of cases under comparable 
> circumstances with a known accurate reference value but practically 
> this is not feasible.
> 
> -- 
> Christoph Hormann
> http://www.imagico.de/
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Warin]

On 18/08/18 18:44, Christoph Hormann wrote:

On Saturday 18 August 2018, Peter Elderson wrote:

Sure. But is there a standard method to indicate this uncertainty in
OSM, which can be processed by data consumers?

This is outside the scope of of OSM IMO and not practical for mappers to
determine in a meaningful way.  You can specify the method used to
determine the width (using source:width=*) and different methods will
have different inherent measurement accuracies but the accuracy of the
values depends on lots of other factors beyond that.

Practically mappers have usually no way to verifiably determine the
accuracy of their measurements unless they have a reference value of
known high accuracy in which case the whole excercise is pointless.
Theoretically mappers could practice their measurement skills and
assess their personal measurement accuracy based on test measuring a
statistically significant number of cases under comparable
circumstances with a known accurate reference value but practically
this is not feasible.



You can usually find set distances measured out for taxis to test their 
equipment on, those are fairly accurate.


The equipment is one source of uncertainty. Farlarger sources can be the 
way in which measurement is performed and the ambient conditions.



I did come across one case where 100 1 meter steel rules were submitted 
for testing. One was rejected as being 1 mm out in that 1 meter ... the 
machine that made it had missed a single 1 mm mark along the length. 
That said, equipment accuracy is fairly good for most items, you would 
have to go really cheap to get some apparatus that would have a large 
uncertainty.



While the width of a canal may be seen as important to visual 
inspections when water is present,


in terms of the water you would want to know the depth of the water, 
shape and smoothness of the walls and the rate of fall.


If your not going to state those things then there is probably not much 
point in the uncertainty of the width.



-- In summary.

Uncertainty statements in OSM are going to be fairly useless.

I don't see data consumers wanting them.

I don't see mappers wanting to state them.

I don't see either of them being able to effectively determine them or 
use them.





___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Saturday 18 August 2018, Peter Elderson wrote:
> Sure. But is there a standard method to indicate this uncertainty in
> OSM, which can be processed by data consumers?

This is outside the scope of of OSM IMO and not practical for mappers to 
determine in a meaningful way.  You can specify the method used to 
determine the width (using source:width=*) and different methods will 
have different inherent measurement accuracies but the accuracy of the 
values depends on lots of other factors beyond that.

Practically mappers have usually no way to verifiably determine the 
accuracy of their measurements unless they have a reference value of 
known high accuracy in which case the whole excercise is pointless.  
Theoretically mappers could practice their measurement skills and 
assess their personal measurement accuracy based on test measuring a 
statistically significant number of cases under comparable 
circumstances with a known accurate reference value but practically 
this is not feasible.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Warin]

How are data consumers going to use the data ???

Not much point in it for them I'd think.

For mappers that come along later it may be a usefull guide for 
comparison with their measurements.
But few mappers will be bothered entering it so I again don't see much 
point.


I'd think tagging like
width:uncertainty=*
ele:uncertainty=*
etc

Or
uncertainty:width=*
uncertainty:ele=*
etc

might be the way to go?
The uncertainty coverage factor of 1 (or standard deviation for a 
Gaussian distribution) could be stipulated as part of any documentation.


On 18/08/18 17:53, Peter Elderson wrote:
Sure. But is there a standard method to indicate this uncertainty in 
OSM, which can be processed by data consumers?


Mvg Peter Elderson

Op 18 aug. 2018 om 01:35 heeft Warin <61sundow...@gmail.com 
> het volgende geschreven:



What you are trying to refer to is 'measurement uncertainty'.

For a non professional rough guide;
https://en.wikipedia.org/wiki/Measurement_uncertainty

Naturally formed water way widths may have a great deal of variation 
along their widths .. and so the uncertainty will be very high unless 
specified along short segments.


On 18/08/18 09:11, Peter Elderson wrote:
It would not be that hard to add a precision to a measurement. Any 
measurement. Maybe there already is a standard method for that?


Mvg Peter Elderson

Op 17 aug. 2018 om 20:50 heeft SelfishSeahorse 
mailto:selfishseaho...@gmail.com>> het 
volgende geschreven:


On Friday, August 17, 2018, Christoph Hormann > wrote:


On Friday 17 August 2018, SelfishSeahorse wrote:

> Of course we could just use width=*, but it's not always easily
> possible to measure the width (e.g. in a forest) and sometimes it
> changes often.

I would translate this into "i want a subjective non-verifiable
classification system but i hide this by defining pro forma
verifiable
criteria for the classes".


A classification based on width is arbitrary, but i don't see why 
it be subjective.


If you want to map the river width tag width=*, if you don't
want to map
the width then don't create classes based on width thresholds.


Imagine a stream/brook in a forest, not visible on satellite 
imagery. You can't measure its width on site (because you don't 
have the equipment or because the soil at its sides is marshy), but 
you know (estimate) that it's wider than 1 metre, but less wide 
than 3 metres. In my opinion it's better to have that information 
that none.


If you enter width="1 m - 3 m", data users very likely won't 
understand it. However if you enter width="2 m", the width value 
pretends to be exact. Besides it is very unlikely that someone else 
verifies that value, considering the fact that less than 1% of 
waterway=* tags have a width=* tag.

___



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Peter Elderson]

Sure. But is there a standard method to indicate this uncertainty in OSM, which 
can be processed by data consumers?

Mvg Peter Elderson

> Op 18 aug. 2018 om 01:35 heeft Warin <61sundow...@gmail.com> het volgende 
> geschreven:
> 
> What you are trying to refer to is 'measurement uncertainty'.
> 
> For a non professional rough guide;
> https://en.wikipedia.org/wiki/Measurement_uncertainty
> 
> Naturally formed water way widths may have a great deal of variation along 
> their widths .. and so the uncertainty will be very high unless specified 
> along short segments. 
> 
>> On 18/08/18 09:11, Peter Elderson wrote:
>> It would not be that hard to add a precision to a measurement. Any 
>> measurement. Maybe there already is a standard method for that?
>> 
>> Mvg Peter Elderson
>> 
>> Op 17 aug. 2018 om 20:50 heeft SelfishSeahorse  
>> het volgende geschreven:
>> 
 On Friday, August 17, 2018, Christoph Hormann  wrote:
 On Friday 17 August 2018, SelfishSeahorse wrote:
 
 > Of course we could just use width=*, but it's not always easily
 > possible to measure the width (e.g. in a forest) and sometimes it
 > changes often.
 
 I would translate this into "i want a subjective non-verifiable 
 classification system but i hide this by defining pro forma verifiable 
 criteria for the classes".
>>> 
>>> A classification based on width is arbitrary, but i don't see why it be 
>>> subjective.
>>> 
 If you want to map the river width tag width=*, if you don't want to map 
 the width then don't create classes based on width thresholds.
>>> 
>>> Imagine a stream/brook in a forest, not visible on satellite imagery. You 
>>> can't measure its width on site (because you don't have the equipment or 
>>> because the soil at its sides is marshy), but you know (estimate) that it's 
>>> wider than 1 metre, but less wide than 3 metres. In my opinion it's better 
>>> to have that information that none.
>>> 
>>> If you enter width="1 m - 3 m", data users very likely won't understand it. 
>>> However if you enter width="2 m", the width value pretends to be exact. 
>>> Besides it is very unlikely that someone else verifies that value, 
>>> considering the fact that less than 1% of waterway=* tags have a width=* 
>>> tag.
>>> ___
>>> Tagging mailing list
>>> Tagging@openstreetmap.org
>>> https://lists.openstreetmap.org/listinfo/tagging
>> 
>> 
>> ___
>> Tagging mailing list
>> Tagging@openstreetmap.org
>> https://lists.openstreetmap.org/listinfo/tagging
> 
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Warin]

What you are trying to refer to is 'measurement uncertainty'.

For a non professional rough guide;
https://en.wikipedia.org/wiki/Measurement_uncertainty

Naturally formed water way widths may have a great deal of variation 
along their widths .. and so the uncertainty will be very high unless 
specified along short segments.


On 18/08/18 09:11, Peter Elderson wrote:
It would not be that hard to add a precision to a measurement. Any 
measurement. Maybe there already is a standard method for that?


Mvg Peter Elderson

Op 17 aug. 2018 om 20:50 heeft SelfishSeahorse 
mailto:selfishseaho...@gmail.com>> het 
volgende geschreven:


On Friday, August 17, 2018, Christoph Hormann > wrote:


On Friday 17 August 2018, SelfishSeahorse wrote:

> Of course we could just use width=*, but it's not always easily
> possible to measure the width (e.g. in a forest) and sometimes it
> changes often.

I would translate this into "i want a subjective non-verifiable
classification system but i hide this by defining pro forma
verifiable
criteria for the classes".


A classification based on width is arbitrary, but i don't see why it 
be subjective.


If you want to map the river width tag width=*, if you don't want
to map
the width then don't create classes based on width thresholds.


Imagine a stream/brook in a forest, not visible on satellite imagery. 
You can't measure its width on site (because you don't have the 
equipment or because the soil at its sides is marshy), but you know 
(estimate) that it's wider than 1 metre, but less wide than 3 metres. 
In my opinion it's better to have that information that none.


If you enter width="1 m - 3 m", data users very likely won't 
understand it. However if you enter width="2 m", the width value 
pretends to be exact. Besides it is very unlikely that someone else 
verifies that value, considering the fact that less than 1% of 
waterway=* tags have a width=* tag.

___
Tagging mailing list
Tagging@openstreetmap.org 
https://lists.openstreetmap.org/listinfo/tagging



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Peter Elderson]

It would not be that hard to add a precision to a measurement. Any measurement. 
Maybe there already is a standard method for that?

Mvg Peter Elderson

> Op 17 aug. 2018 om 20:50 heeft SelfishSeahorse  
> het volgende geschreven:
> 
>> On Friday, August 17, 2018, Christoph Hormann  wrote:
>> On Friday 17 August 2018, SelfishSeahorse wrote:
>> 
>> > Of course we could just use width=*, but it's not always easily
>> > possible to measure the width (e.g. in a forest) and sometimes it
>> > changes often.
>> 
>> I would translate this into "i want a subjective non-verifiable 
>> classification system but i hide this by defining pro forma verifiable 
>> criteria for the classes".
> 
> A classification based on width is arbitrary, but i don't see why it be 
> subjective.
> 
>> If you want to map the river width tag width=*, if you don't want to map 
>> the width then don't create classes based on width thresholds.
> 
> Imagine a stream/brook in a forest, not visible on satellite imagery. You 
> can't measure its width on site (because you don't have the equipment or 
> because the soil at its sides is marshy), but you know (estimate) that it's 
> wider than 1 metre, but less wide than 3 metres. In my opinion it's better to 
> have that information that none.
> 
> If you enter width="1 m - 3 m", data users very likely won't understand it. 
> However if you enter width="2 m", the width value pretends to be exact. 
> Besides it is very unlikely that someone else verifies that value, 
> considering the fact that less than 1% of waterway=* tags have a width=* tag.
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Friday 17 August 2018, SelfishSeahorse wrote:
> [...] You can't measure its width on site (because you don't have the
> equipment or because the soil at its sides is marshy), but you know
> (estimate) that it's wider than 1 metre, but less wide than 3 metres.

If you know it is wider than 1m and less wide than 3m you have a pretty 
precise measurement.

> However if you enter width="2 m", the width value
> pretends to be exact.

Why do you assume that?

In OSM individual data points - be that in tags or in coordinates - come 
with no implication of accuracy of the data.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

[Tagging] How to tag small canals?

[SelfishSeahorse]

On Friday, August 17, 2018, Christoph Hormann  wrote:

> On Friday 17 August 2018, SelfishSeahorse wrote:
>
> > Of course we could just use width=*, but it's not always easily
> > possible to measure the width (e.g. in a forest) and sometimes it
> > changes often.
>
> I would translate this into "i want a subjective non-verifiable
> classification system but i hide this by defining pro forma verifiable
> criteria for the classes".


A classification based on width is arbitrary, but i don't see why it be
subjective.

If you want to map the river width tag width=*, if you don't want to map
> the width then don't create classes based on width thresholds.
>

Imagine a stream/brook in a forest, not visible on satellite imagery. You
can't measure its width on site (because you don't have the equipment or
because the soil at its sides is marshy), but you know (estimate) that it's
wider than 1 metre, but less wide than 3 metres. In my opinion it's better
to have that information that none.

If you enter width="1 m - 3 m", data users very likely won't understand it.
However if you enter width="2 m", the width value pretends to be exact.
Besides it is very unlikely that someone else verifies that value,
considering the fact that less than 1% of waterway=* tags have a width=*
tag.
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Paul Allen]

On Fri, Aug 17, 2018 at 12:54 PM, Philip Barnes 
wrote:

> Creek is certainly used in Essex.
>

I didn't say it wasn't used.  Or even that it wasn't used to denote a wide
stream.  However, it also has other meanings.
Such as a wide tidal bay to a marsh on an estuary.  Like one near me.  It's
as wide as the river it feeds into but what
feeds into it is a narrow stream.  And it's called, by the wildlife trust
that owns that area, a creek (which it is, by one
definition).

The problem with creek is the different meanings.  You can guarantee that
some would use it to mean small stream, as
it would be defined in the wike and others would use it to map tidal inlets
because they're named "Foo Creek" and therefore
waterway=creek is the obvious and natural tag to use.

-- 
Paul
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Philip Barnes]

Creek is certainly used in Essex.

Phil (trigpoint) 

On 17 August 2018 12:40:38 BST, Paul Allen  wrote:
>On Fri, Aug 17, 2018 at 12:16 PM, SelfishSeahorse
>> wrote:
>
>>
>> Actually, it's only drain that doesn't seem to make sense
>> semantically, but ditch seems to be fine for smaller canals used for
>> drainage and irrigation, at least according to the definitions by
>> Wikipedia[^1] and the Cambridge Dictionary[^2].
>>
>
>As used by Ordnance Survey in the UK, "drain" is what most of us (even
>in
>the UK) think of as a ditch.
>
>I'm not sure if a (smaller) mill race can also be called a ditch,
>
>
>I wouldn't think of a mill race as a ditch.  To my mind a ditch is not
>about transporting water from A to B and
>especially not for generating mechanical power.  I wouldn't call a mill
>race a canal, either, but it seems to be
>closer in definition than ditch is.
>
>
>> but i think it makes sense to have at least two tags for man-made
>channels
>> of different width.
>>
>
>Not if width=* can do the job.  Not if it's possible to draw an outline
>of
>the bank.  Both are possible.  It's just that
>current OSM carto doesn't honour them in this situation.
>
>* waterway=creek - small to medium-sized natural stream (1-3 m wide)
>>
>
>That is not UK usage.  See https://en.wiktionary.org/wiki/creek
>
>The only word I can find meaning large stream is "beck."  Or you could
>have
>"brook" for small stream.  Giving
>either river/beck/stream or river/stream/brook.  Or we just stick with
>river/stream.  Especially as we don't have a
>meaningful definition of the difference (how far can you jump).
>
>All this is complicated by the fact that what starts out as a brook
>often
>turns into a stream which can turn into a river, and is
>named "River X" or "Afon X" (if you're Welsh") all the way from the
>mouth
>to the dripping tap in somebody's back yard (spot
>the old TV comedy reference).
>
>-- 
>Paul

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Philip Barnes]

A brook is bigger than a stream. A beck, or Clough, is the same as a brook, 
just a different regional dialect.

But true a stream will turn into a brook will turn into a river.

Phil (trigpoint) 

On 17 August 2018 12:40:38 BST, Paul Allen  wrote:
>On Fri, Aug 17, 2018 at 12:16 PM, SelfishSeahorse
>> wrote:
>
>>
>> Actually, it's only drain that doesn't seem to make sense
>> semantically, but ditch seems to be fine for smaller canals used for
>> drainage and irrigation, at least according to the definitions by
>> Wikipedia[^1] and the Cambridge Dictionary[^2].
>>
>
>As used by Ordnance Survey in the UK, "drain" is what most of us (even
>in
>the UK) think of as a ditch.
>
>I'm not sure if a (smaller) mill race can also be called a ditch,
>
>
>I wouldn't think of a mill race as a ditch.  To my mind a ditch is not
>about transporting water from A to B and
>especially not for generating mechanical power.  I wouldn't call a mill
>race a canal, either, but it seems to be
>closer in definition than ditch is.
>
>
>> but i think it makes sense to have at least two tags for man-made
>channels
>> of different width.
>>
>
>Not if width=* can do the job.  Not if it's possible to draw an outline
>of
>the bank.  Both are possible.  It's just that
>current OSM carto doesn't honour them in this situation.
>
>* waterway=creek - small to medium-sized natural stream (1-3 m wide)
>>
>
>That is not UK usage.  See https://en.wiktionary.org/wiki/creek
>
>The only word I can find meaning large stream is "beck."  Or you could
>have
>"brook" for small stream.  Giving
>either river/beck/stream or river/stream/brook.  Or we just stick with
>river/stream.  Especially as we don't have a
>meaningful definition of the difference (how far can you jump).
>
>All this is complicated by the fact that what starts out as a brook
>often
>turns into a stream which can turn into a river, and is
>named "River X" or "Afon X" (if you're Welsh") all the way from the
>mouth
>to the dripping tap in somebody's back yard (spot
>the old TV comedy reference).
>
>-- 
>Paul

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Friday 17 August 2018, SelfishSeahorse wrote:
> >
> > Yes, that was the obvious attempt to expand the narrow scheme to
> > other parts of the world in a superficial way oriented at the
> > standard style rendering but not at the actual semantics.
>
> Actually, it's only drain that doesn't seem to make sense
> semantically, but ditch seems to be fine for smaller canals used for
> drainage and irrigation, at least according to the definitions by
> Wikipedia[^1] and the Cambridge Dictionary[^2].

The meaning of tags in OSM however does not necessarily have anything to 
do with the meaning of the English language words used for key and 
value.

waterway=drain and waterway=ditch were both from the beginning intended 
and used primarily for structures transporting away undesirable water.

> I'm not sure if a (smaller) mill race can also be called a ditch, but
> i think it makes sense to have at least two tags for man-made
> channels of different width.

Because that works so great for natural waterways?

I would say if you want to document the width then tag width=*.

> In my opinion the material of the lining/confine should better be
> tagged separately, perhaps confine:material=concrete/wood/
>
> As for natural waterways, I could imagine the following
> classification based on width:
>
> * waterway=stream/brook - possible to jump across (< 1 m wide?)
> * waterway=creek - small to medium-sized natural stream (1-3 m wide)
> * waterway=river (3-10 m wide)
> * waterway=broad_river (> 10 m wide)

Somewhat off topic here - but as said you cannot change the meaning of a 
tag that is used hundreds of thousands of times.

> Of course we could just use width=*, but it's not always easily
> possible to measure the width (e.g. in a forest) and sometimes it
> changes often.

I would translate this into "i want a subjective non-verifiable 
classification system but i hide this by defining pro forma verifiable 
criteria for the classes".

If you want to map the river width tag width=*, if you don't want to map 
the width then don't create classes based on width thresholds.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Paul Allen]

On Fri, Aug 17, 2018 at 12:16 PM, SelfishSeahorse  wrote:

>
> Actually, it's only drain that doesn't seem to make sense
> semantically, but ditch seems to be fine for smaller canals used for
> drainage and irrigation, at least according to the definitions by
> Wikipedia[^1] and the Cambridge Dictionary[^2].
>

As used by Ordnance Survey in the UK, "drain" is what most of us (even in
the UK) think of as a ditch.

I'm not sure if a (smaller) mill race can also be called a ditch,


I wouldn't think of a mill race as a ditch.  To my mind a ditch is not
about transporting water from A to B and
especially not for generating mechanical power.  I wouldn't call a mill
race a canal, either, but it seems to be
closer in definition than ditch is.


> but i think it makes sense to have at least two tags for man-made channels
> of different width.
>

Not if width=* can do the job.  Not if it's possible to draw an outline of
the bank.  Both are possible.  It's just that
current OSM carto doesn't honour them in this situation.

* waterway=creek - small to medium-sized natural stream (1-3 m wide)
>

That is not UK usage.  See https://en.wiktionary.org/wiki/creek

The only word I can find meaning large stream is "beck."  Or you could have
"brook" for small stream.  Giving
either river/beck/stream or river/stream/brook.  Or we just stick with
river/stream.  Especially as we don't have a
meaningful definition of the difference (how far can you jump).

All this is complicated by the fact that what starts out as a brook often
turns into a stream which can turn into a river, and is
named "River X" or "Afon X" (if you're Welsh") all the way from the mouth
to the dripping tap in somebody's back yard (spot
the old TV comedy reference).

-- 
Paul
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Friday 17 August 2018, Javier Sánchez Portero wrote:
> This kind of man-made structure for water transportation is very
> frequent in Azores, Madeira, Canary Islands and Mediterranean
> countries. Please see https://en.wikipedia.org/wiki/Levada

Yes, levadas are a good example for the kind of structures that were not 
on the radar of those who originally invented the waterway tags.

> Now most of them are tagged with waterway=ditch.

You probably agree that this is somewhat unfortunate because there is no 
way to distinguish them from a classic ditch for the data user.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[SelfishSeahorse]

On Thu, 16 Aug 2018 at 18:20, Christoph Hormann  wrote:
>
> On Thursday 16 August 2018, Martin Koppenhoefer wrote:
> >
> > The canal definition was changed in March 2018, before it said to use
> > canal only for „the largest waterways created for irrigation
> > purposes“
>
> Yes, that was the obvious attempt to expand the narrow scheme to other
> parts of the world in a superficial way oriented at the standard style
> rendering but not at the actual semantics.

Actually, it's only drain that doesn't seem to make sense
semantically, but ditch seems to be fine for smaller canals used for
drainage and irrigation, at least according to the definitions by
Wikipedia[^1] and the Cambridge Dictionary[^2].

[^1]: 
[^2]: 

I'm not sure if a (smaller) mill race can also be called a ditch, but
i think it makes sense to have at least two tags for man-made channels
of different width.

In my opinion the material of the lining/confine should better be
tagged separately, perhaps confine:material=concrete/wood/

As for natural waterways, I could imagine the following classification
based on width:

* waterway=stream/brook - possible to jump across (< 1 m wide?)
* waterway=creek - small to medium-sized natural stream (1-3 m wide)
* waterway=river (3-10 m wide)
* waterway=broad_river (> 10 m wide)

Of course we could just use width=*, but it's not always easily
possible to measure the width (e.g. in a forest) and sometimes it
changes often.

However, if new waterway=* values were introduced, it would be
necessary that OSM Carto would render them from the day they were
approved, otherwise people would not use them.

Cheers

Markus

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Javier Sánchez Portero]

This kind of man-made structure for water transportation is very frequent
in Azores, Madeira, Canary Islands and Mediterranean countries. Please see
https://en.wikipedia.org/wiki/Levada

Now most of them are tagged with waterway=ditch.

As long as I now, they could be used not only for irrigation but also for
drinking. They could be open to air or closed, located on ground, over
(aqueduct) or inside (tunnel). Usual width from 0.2 to 1.0 meters.

El jue., 16 ago. 2018 a las 15:15, Christoph Hormann ()
escribió:

> On Thursday 16 August 2018, SelfishSeahorse wrote:
> > Hello
> >
> > What is the usual (or sensible) way to tag small canals like mill
> > races (example: [^1]) or small irrigation channels (example: [^2]),
> > i.e. the small equivalent of waterway=canal?
>
> A bit of information on current meaning of artificial waterway tags:
>
> waterway=ditch and waterway=drain are largely used without a well
> defined difference between them.  In terms of documented meaning
> waterway=ditch is more for features collecting water while
> waterway=drain is more for features transporting water.  But this is
> not a difference you can find consistently being made in actual
> mapping.  Both tags were invented and are primarily used for waterways
> removing undesirable water.
>
> There is also some minor use of waterway=ditch and waterway=drain for
> smaller natural waterways because it is rendered slightly thinner than
> waterway=stream in the standard style but this is generally accepted to
> be abuse of the tags.
>
> waterway=canal is essentially for all open artificial waterways that are
> not primarily for transporting away undesirable water (which would be
> waterway=ditch or waterway=drain).  However since the standard style
> renders it in a fairly prominent form with a thick line smaller canals
> (like for irrigation purposes) are often tagged differently
> (waterway=ditch, waterway=drain or waterway=stream) despite not
> qualifying as such.  Still waterway=canal is the corrent tagging here,
> nothing except the standard style suggests a lower size limit for
> waterway=canal.
>
> All of this together has its origin in the fact that in the UK and other
> early OSM countries large artificial waterways are almost always for
> navigation and small artificial waterways are almost always for
> transporting away undesirable water.
>
> Long story short: My recommendation would be tagging waterway=canal and
> specifying usage=* and width=*.  This might not look ideal on the map
> but will allow all data users to correctly interpret the data.
>
> --
> Christoph Hormann
> http://www.imagico.de/
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Thursday 16 August 2018, Dave Swarthout wrote:
> I started using the service=irrigation tag a
> couple of years ago when I discovered it in the Wiki somewhere. I
> also noticed today that there is a "usage=irrigation" that applies to
> waterway=canal. It's all a bit confusing.

Yes, it is.  usage=* is the better documented variant for waterway=canal 
while service=* is more widely used.  In documented values service=* is 
more focussed on the overall function of the canal as a whole while 
usage=* is more about the technical design and function of the 
particular part of the waterway.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Thursday 16 August 2018, Martin Koppenhoefer wrote:
> >
> > All of this together has its origin in the fact that in the UK and
> > other early OSM countries large artificial waterways are almost
> > always for navigation and small artificial waterways are almost
> > always for transporting away undesirable water.
>
> +1
> That’s my analysis as well. What do you think, shouldn’t we fix the
> wiki to make it more universally applicable to all kinds of
> waterways?

As usual i don't think you can change the definition of tags that are 
already used hundreds of thousands of times except by extending their 
scope which would make the tags more vague than they already are.

What would make sense is to explicitly mention that waterway=canal does 
not have a lower size.

Inventing a separate tagging scheme for irrigation systems might be an 
option too but irrigation is performed in very different ways in 
different parts of the world so it might not be too easy to create an 
universal tagging system for that.

> The canal definition was changed in March 2018, before it said to use
> canal only for „the largest waterways created for irrigation
> purposes“

Yes, that was the obvious attempt to expand the narrow scheme to other 
parts of the world in a superficial way oriented at the standard style 
rendering but not at the actual semantics.

Going this route would essentially mean turning all of ditch/drain/canal 
into one big uniform catch-all.  Removing the above is the attempt to 
salvage some of the semantic value in the data.  Not sure how 
successful that is without support from the standard style though.

Unfortunately waterway=canal has only 13k combinations with width=* and 
only 1.8k combinations with usage=* at the moment.  Extending that 
would be the best way to move forward IMO.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Dave Swarthout]

There are thousands of kilometers of irrigation waterways scattered around
Thailand. Many but not all of these are concrete lined and may be from 1 to
5 meters across.  The Thai word used in names for the larger ones is "klong",
which means "canal" in English.

The tagging practice of mappers working in Thailand has been to use
waterway=canal, sometimes with boat=no, and sometimes with
service=irrigation. I started using the service=irrigation tag a couple of
years ago when I discovered it in the Wiki somewhere. I also noticed today
that there is a "usage=irrigation" that applies to waterway=canal. It's all
a bit confusing.

Regards,

Dave

On Thu, Aug 16, 2018 at 7:01 AM Martin Koppenhoefer 
wrote:

>
>
> sent from a phone
>
> > On 16. Aug 2018, at 16:14, Christoph Hormann  wrote:
> >
> > All of this together has its origin in the fact that in the UK and other
> > early OSM countries large artificial waterways are almost always for
> > navigation and small artificial waterways are almost always for
> > transporting away undesirable water.
>
>
> +1
> That’s my analysis as well. What do you think, shouldn’t we fix the wiki
> to make it more universally applicable to all kinds of waterways? Or would
> you suggest we invent new waterway types based on different functional
> purposes to fill the gaps that current definitions leave behind?
> Do we really need construction details as parts of basic artificial
> waterway definitions?
>
>
> Cheers,
> Martin
>
>
>
> ___
> Tagging mailing list
> Tagging@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/tagging
>


-- 
Dave Swarthout
Homer, Alaska
Chiang Mai, Thailand
Travel Blog at http://dswarthout.blogspot.com
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Martin Koppenhoefer]

sent from a phone

> On 16. Aug 2018, at 16:14, Christoph Hormann  wrote:
> 
> All of this together has its origin in the fact that in the UK and other 
> early OSM countries large artificial waterways are almost always for 
> navigation and small artificial waterways are almost always for 
> transporting away undesirable water.


+1
That’s my analysis as well. What do you think, shouldn’t we fix the wiki to 
make it more universally applicable to all kinds of waterways? Or would you 
suggest we invent new waterway types based on different functional purposes to 
fill the gaps that current definitions leave behind? 
Do we really need construction details as parts of basic artificial waterway 
definitions?


Cheers,
Martin 



___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Paul Allen]

On Thu, Aug 16, 2018 at 3:02 PM, Andy Townsend  wrote:

> On 16/08/2018 08:49, SelfishSeahorse wrote:
>


> It seems to me that waterway=ditch +
>> usage=headrace/tailrace/irrigation fits best, but the wiki defines
>> waterway=ditch as 'a small man-made draining waterway, often found
>> along roads'.
>>
>
> Personally I'd probably use "drain" for those two, and "ditch" maybe for
> some smaller ones.
>

I used waterway=canal + usage=headrace for the flow to a water mill.  I'm
not entirely sure that is correct because
the wiki seems to be defining headrace in terms of water flowing through a
tunnel (to electrical turbines or similar)
without a thought for other usage.

There's a problem with it.  It renders ridiculously wide (on the standard
map) at low zooms, ending up as large as the
mill pond.  I tried limiting it with the width tag, which appeared to be
ignored on the standard map.  I then tried using
natural=water + water=canal on an area to limit the width that way, but to
no effect.  At high zooms it's correct, at lower
zooms it is rendered far wider than it should be.

See https://www.openstreetmap.org/#map=19/52.08104/-4.67894  And yes, "Mill
Race" and "Mill Pond" look like
descriptions.  They also happen to be the names used to refer to those
features.  The mill itself is called "Y Felin."
Which is Welsh for "The Mill."  Often the name of a thing is also a
description because, over time, the description
becomes the name.

And, yes, I have been unable to determine the route of the water after the
tailrace.  Maybe there's a teleporter
below the wheel.

-- 
Paul
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Christoph Hormann]

On Thursday 16 August 2018, SelfishSeahorse wrote:
> Hello
>
> What is the usual (or sensible) way to tag small canals like mill
> races (example: [^1]) or small irrigation channels (example: [^2]),
> i.e. the small equivalent of waterway=canal?

A bit of information on current meaning of artificial waterway tags:

waterway=ditch and waterway=drain are largely used without a well 
defined difference between them.  In terms of documented meaning 
waterway=ditch is more for features collecting water while 
waterway=drain is more for features transporting water.  But this is 
not a difference you can find consistently being made in actual 
mapping.  Both tags were invented and are primarily used for waterways 
removing undesirable water.

There is also some minor use of waterway=ditch and waterway=drain for 
smaller natural waterways because it is rendered slightly thinner than 
waterway=stream in the standard style but this is generally accepted to 
be abuse of the tags.

waterway=canal is essentially for all open artificial waterways that are 
not primarily for transporting away undesirable water (which would be 
waterway=ditch or waterway=drain).  However since the standard style 
renders it in a fairly prominent form with a thick line smaller canals 
(like for irrigation purposes) are often tagged differently 
(waterway=ditch, waterway=drain or waterway=stream) despite not 
qualifying as such.  Still waterway=canal is the corrent tagging here, 
nothing except the standard style suggests a lower size limit for 
waterway=canal.

All of this together has its origin in the fact that in the UK and other 
early OSM countries large artificial waterways are almost always for 
navigation and small artificial waterways are almost always for 
transporting away undesirable water.

Long story short: My recommendation would be tagging waterway=canal and 
specifying usage=* and width=*.  This might not look ideal on the map 
but will allow all data users to correctly interpret the data.

-- 
Christoph Hormann
http://www.imagico.de/

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Martin Koppenhoefer]

sent from a phone

> On 16. Aug 2018, at 16:02, Andy Townsend  wrote:
> 
> Personally I'd probably use "drain" for those two, and "ditch" maybe for some 
> smaller ones.


me too, on a pragmatic level and because there are no alternatives, just that 
it doesn’t make sense on a semantic level and it is contradicted by the wiki 
definitions.
 
If I don’t recall wrong, last time someone wanted to change the wiki 
definitions, there was a vocal majority for keeping „concrete lining“ and 
draining function as requirements for drains.


cheers,
Martin 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Martin Koppenhoefer]

sent from a phone

> On 16. Aug 2018, at 15:36, Philip Barnes  wrote:
> 
> In British English usage we have regional names for the category between a 
> stream and a river. Where I come from they are called Brooks, but I am aware 
> of Burns and Becks being used in the North.


there are a lot of different categories and types of watercourses in natural 
language, e.g. there are runnel, trickle, runlet, rill, rivulet, torrent, creek 
(AE?), stream, anabranch, waterway, drainage channel, irrigation ditch, ...

We won’t need all of these as tags, but as soon as we can decide/find out what 
the waterway classification is about, we can see how to organize it ;-). 
Clearly there is a reason natural language has developed more than one or two 
words for kinds of watercourses, it makes conversation easier.

As I said, irrigation isn’t covered at all (or at least the wiki is 
self-contradicting with respect to irrigation). As a German, I would also 
appreciate having a category for major rivers (in German: „Strom“, apparently 
English doesn’t have a proper word for it).

Cheers,
Martin 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Andy Townsend]

On 16/08/2018 08:49, SelfishSeahorse wrote:

What is the usual (or sensible) way to tag small canals like mill
races (example: [^1]) or small irrigation channels (example: [^2]),
i.e. the small equivalent of waterway=canal?

[^1]: 
https://commons.wikimedia.org/wiki/File:15-19-077,_mingus_creek_-_panoramio.jpg
[^2]: https://commons.wikimedia.org/wiki/File:2003-Suone.jpg

It seems to me that waterway=ditch +
usage=headrace/tailrace/irrigation fits best, but the wiki defines
waterway=ditch as 'a small man-made draining waterway, often found
along roads'.


Personally I'd probably use "drain" for those two, and "ditch" maybe for 
some smaller ones.


Best Regards,

Andy.


___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Philip Barnes]

On 16 August 2018 14:15:23 BST, Martin Koppenhoefer  
wrote:
>
>There isn’t a universal definition of river, but locally people usually
>know whether a waterway is considered a river or not (in my home area,
>a 3-5 m waterway is usually not considered a river, but you cannot jump
>over it either). There’s a huge difference between a stream of 0,5-1m
>and one of 5m.

In British English usage we have regional names for the category between a 
stream and a river. Where I come from they are called Brooks, but I am aware of 
Burns and Becks being used in the North. 

Phil (trigpoint) 
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] How to tag small canals?

[Martin Koppenhoefer]

sent from a phone

> On 16. Aug 2018, at 09:49, SelfishSeahorse  wrote:
> 
> It seems to me that waterway=ditch +
> usage=headrace/tailrace/irrigation fits best, but the wiki defines
> waterway=ditch as 'a small man-made draining waterway, often found
> along roads'.


yes, waterway tagging is not very clear in general, some refining would be 
seriously appreciated.

Based on the established waterway tags, there are currently only 2 classes for 
natural waterways:
stream (sufficiently narrow that an adult can jump over it) and river (bigger), 
formerly established „wadi“ is now discouraged AFAIK,
and 3 for artificial waterways:
ditch, drain, canal

plus additional attributes (width, intermittent, seasonal, usage, ...).

Apparently (that’s what the wiki says at the moment) the artificial waterways 
are to be classified according to their function and type of construction 
(canal for navigable waterways and big drains, drain for draining waterways 
lined with concrete or similar (storm water and grey water is explicitly 
mentioned) and ditch for „simple“ waterways not lined with concrete.

I’m probably not the only one who thinks this is not a good system. Why would 
we require „concrete“, this is clearly context specific, you can have 
artificial waterways without any concrete. At least once we could mention 
„steel“.

As there is already a „usage“ property, the usage shouldn’t matter for the 
class (let’s not intermingle orthogonal properties, this will work in some 
setting and fail in others).

IMHO we should see the waterway tags as a network hierarchy, similar to roads 
(and I assumed we did this, but looking at how the wiki evolved it is 
apparently not a thought shared with everybody).
If you have to drain vast areas you will build small ditches which discharge 
into bigger artificial waterways, which again might be collected prior to flow 
into a natural waterway. And all that without any „storm water“ or „industrial 
discharge“.

There isn’t a universal definition of river, but locally people usually know 
whether a waterway is considered a river or not (in my home area, a 3-5 m 
waterway is usually not considered a river, but you cannot jump over it 
either). There’s a huge difference between a stream of 0,5-1m and one of 5m.

Following the wiki by the word, we have no waterway type for irrigation, 
because all waterways that are mentioned and defined are for „draining“ or 
navigating. Likely the result of having central and north europeans writing the 
definitions, where abundance of water is the standard.

Cheers,
Martin 
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

[Tagging] How to tag small canals?

[SelfishSeahorse]

Hello

What is the usual (or sensible) way to tag small canals like mill
races (example: [^1]) or small irrigation channels (example: [^2]),
i.e. the small equivalent of waterway=canal?

[^1]: 
https://commons.wikimedia.org/wiki/File:15-19-077,_mingus_creek_-_panoramio.jpg
[^2]: https://commons.wikimedia.org/wiki/File:2003-Suone.jpg

It seems to me that waterway=ditch +
usage=headrace/tailrace/irrigation fits best, but the wiki defines
waterway=ditch as 'a small man-made draining waterway, often found
along roads'.

Regards

Markus

___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging