Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Tomas Straupis 
>>  Water proposal tried to change the tagging:
>>  landuse=reservoir => natural=water|water=reservoir
>>  And in general all water landuse=x => natural=water|water=x (basin, pond 
>> etc.)
>>  waterway=riverbank => natural=water|water=riverbank
> No.
> The water proposal didn't change or deprecate anything, and that is explicitly
> stated on both the proposal and water=* wiki pages.

  Either you're lying, or... it is ok to change landuse=reservoir
status from debated to VALID then? The part in your proposal about
deprecating landuse/riverbank parts was... mistake... fun... local
"referendum"? :-)

> 1. It reassures that it's okay to tag any body of water visible
> 

  So YOU thought that it suits YOUR understanding better and so screw
those thousands of hundreds of objects already marked. Yes? Less than
one year of participation in OSM for you was enough to get the whole
idea?

> Data consumers will have to live with two tagging schemes,
> as that is the way OSM works.

  It does not have to be that terrible way (it was not before your
proposal). Data consumers deserve better data. Which we do have if
selfish people do not screw it up.

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Ilya Zverev 
Tomas Straupis wrote:

>  Water proposal tried to change the tagging:
>  landuse=reservoir => natural=water|water=reservoir
>  And in general all water landuse=x => natural=water|water=x (basin, pond 
> etc.)
>  waterway=riverbank => natural=water|water=riverbank

No.

The water proposal didn't change or deprecate anything, and that is explicitly 
stated on both the proposal and water=* wiki pages.

It does two things:

1. It reassures that it's okay to tag any body of water visible on a satellite 
imagery with natural=water. You don't have to search for clues for the type of 
the body of water, you can just tag it with natural=water, and it would be 
perfectly fine. When another mapper with local knowledge finds that, they can 
add a detailing water=* tag.

2. It allows for discerning lakes, oxbow lakes, and coves; ponds and reflecting 
pools; reservoirs and wastewater reservoirs. The major reason for starting the 
proposal was to invent a way to tag oxbow lakes, btw.

Data consumers will have to live with two tagging schemes, as that is the way 
OSM works. It's not the first time and not the last.

IZ
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Max 
On 2016년 06월 22일 18:36, Martin Koppenhoefer wrote:
> 
> 2016-06-22 10:49 GMT+02:00 Max  >:
> 
> I did not see any
> voting.
> 
how?


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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Martin Koppenhoefer 
2016-06-22 13:18 GMT+02:00 Tomas Straupis :

> > I am generally against such harsh measures, if a new way to tag
> > has advantages, it has them even if only 20% of applicable objects
> > are tagged with it. And 20% endorsement isn't actually a fail IMHO.
>
>   So after a new scheme to tag X is introduced we have two schemes
> valid at the same time and if the new one only gets 20% in two years
> you suggest to continue with TWO ways to map THE SAME thing? How would
> you explain this to data consumers?
>


the question is not "mapping the same thing", but conveying the same
semantics, which is a whole lot different, and can rarely - if ever so far
- be found.



>
> >> P.S. This only influences proposals which are CHANGING tagging.
> > so it would not apply to the water tag, because it doesn't change
> > tagging but is an amendment?
>
>   Water proposal tried to change the tagging:
>   landuse=reservoir => natural=water|water=reservoir
>


that's really different, one is an attribute about the usage of land, the
other is a feature for where there is actually water.



>   And in general all water landuse=x => natural=water|water=x (basin, pond
> etc.)
>   waterway=riverbank => natural=water|water=riverbank
>


yes, the way riverbank was used in OSM is replaced by natural=water,
water=riverbank seems a bad tag indeed, if still used for the actual river
area and not for the riverbank alone.


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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Tomas Straupis 
> I am generally against such harsh measures, if a new way to tag
> has advantages, it has them even if only 20% of applicable objects
> are tagged with it. And 20% endorsement isn't actually a fail IMHO.

  So after a new scheme to tag X is introduced we have two schemes
valid at the same time and if the new one only gets 20% in two years
you suggest to continue with TWO ways to map THE SAME thing? How would
you explain this to data consumers?

>> P.S. This only influences proposals which are CHANGING tagging.
> so it would not apply to the water tag, because it doesn't change
> tagging but is an amendment?

  Water proposal tried to change the tagging:
  landuse=reservoir => natural=water|water=reservoir
  And in general all water landuse=x => natural=water|water=x (basin, pond etc.)
  waterway=riverbank => natural=water|water=riverbank

  So after successful change we would have had no landuse=reservoir
and waterway=riverbank. But we do and they are used much more often
than the new way even after FIVE years => thus proposition "failed".

> there are hardly any proposals that change tagging for changes sake.

  There will always be ideas on how tagging of X could be "improved"
by ones mind (think about renaming "highway" to "way"?). All changing
proposals are fixing something "here" and damaging something "there".
So my idea is to stop changing tags when number of tagged objects is
too large (as it was the case with reservoirs, riverbanks etc.)

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Martin Koppenhoefer 
2016-06-22 10:49 GMT+02:00 Max :

> I did not see any
> voting.
>



maybe you should have started it then ;-)

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Lester Caine 
On 21/06/16 22:44, Greg Troxel wrote:
> I don't see any reason why this can't be
> 
>   landuse=reservoir [entire parcel that has the reservoir on it, or the
>   region that has "water supply area - no trespassing" signs, etc.]]
> 
>   water=reservoir [the water part]

That makes a lot more sense than ALSO adding natural=water to the water
area. If the area of water is covered we have man_made=reservoir_covered
and no 'water' element.
water treatment plant tagging is similarly erratic with waste having an
assortment of tags and clean water not. water=reservoir may well part of
a clean water supply and treatment site.

The problem is not that water has failed to gain traction, but rather
that the whole area is still a mess that can not easily be defined.
Nothing actually flows well?

Adding ...
landuse=wastewater_treatment
or
landuse=water_supply

wrapping

man_made=reservoir
man_made=reservoir_covered

But then we have currently landuse=basin and landuse=pond so dropping
the natural=water and rendering all water= as blue then the use of water
on it's own does make more sense?

JUST use
water=reservoir
water=basin
water=pond
man_made=reservoir_covered -> because water is not visible so may be
landcover=grass over the top?

But where there is navigation passing through the reservoir
waterway=lock and waterway=canal makes the water version of those
obsolete? There is no way 'water' will replace any of the waterway data
structure given all of the secondary tagging to
http://wiki.openstreetmap.org/wiki/Tag:waterway%3Driver and the like.

But where lakes - natural and man-made - have navigation routes through
many of the waterway tags also apply but this has been another area of
discussion to map the actual routes across the water areas.

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Max 
the whole process is too opaque. Also there are different protagonists
posting on wiki and email list. The gallery tag clarification stalled,
because what looked like a consensus on the email list had one loud
opponent on the wiki and not many other voices. I did not see any
voting. IMHO the process is inherently flawed.


On 2016년 06월 22일 15:17, Tomas Straupis wrote:
> My question/proposal was about what to do with failed proposals in
> general. That is:
> 1. How to identify a "failed" proposal
> 2. What to do with it
> 
> My proposal for point 1 is:
> If after say two years new schema does not get at least equal tagging
> count as the old schema - proposal failed.
> 
> My proposal for point 2 is:
> Mark that proposal as filed" in wiki. Mark all wiki pages which were
> marked as "deprecated" because of the failed proposal as "valid".
> 
> P.S. This only influences proposals which are CHANGING tagging.
> P.P.S. There should also be guards against such proposals in the first
> place, but lets park it for now.
> 


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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Warin 

On 6/22/2016 4:17 PM, Tomas Straupis wrote:

My question/proposal was about what to do with failed proposals in
general. That is:
1. How to identify a "failed" proposal
2. What to do with it

My proposal for point 1 is:
If after say two years new schema does not get at least equal tagging
count as the old schema - proposal failed.

My proposal for point 2 is:
Mark that proposal as "failed" in wiki. Mark all wiki pages which were
marked as "deprecated" because of the failed proposal as "valid".

P.S. This only influences proposals which are CHANGING tagging.
P.P.S. There should also be guards against such proposals in the first
place, but lets park it for now.


'Proposals'  these should raised on the tagging list;

usually as firstly status=draft, then (usually very quickly) to 
status='Request For Comment, then is 2 weeks or so o status='Voting' 
then depending on the vote status='rejected' or 'approved'.


Tags that don't go through this process may have there status unset, set 
to 'defacto', 'inuse' ... but should not have there status to 'approved'.


Where I make a tag that might conflict with some other .. I usually 
mention it on the wiki page I make ...


e.g. 
http://wiki.openstreetmap.org/wiki/Tag:sport%3Dshot-put#Alternate_spelling.2Ftagging_schemes










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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Martin Koppenhoefer 


sent from a phone

> Il giorno 22 giu 2016, alle ore 08:17, Tomas Straupis 
>  ha scritto:
> 
> My question/proposal was about what to do with failed proposals in
> general. That is:
> 1. How to identify a "failed" proposal
> 2. What to do with it
> 
> My proposal for point 1 is:
> If after say two years new schema does not get at least equal tagging
> count as the old schema - proposal failed.


I am generally against such harsh measures, if a new way to tag has advantages, 
it has them even if only 20% of applicable objects are tagged with it. And 20% 
endorsement isn't actually a fail IMHO. FWIW, if you want to establish a hard 
time frame, it should be as long as the former tagging was in use.


> 
> P.S. This only influences proposals which are CHANGING tagging.


so it would not apply to the water tag, because it doesn't change tagging but 
is an amendment?



> P.P.S. There should also be guards against such proposals in the first
> place, but lets park it for now.


there are hardly any proposals that change tagging for changes sake.  Even the 
well established but strongly criticized highway=path proposal introduced a new 
feature class that didn't exist before (narrow ways that have no 
dedicated/preferred means of transport, and ways that have two or more equally 
important means of transport/shared dedication).

Usually these proposals that change established tagging are set up because 
there are issues with the current way of doing things, e.g. in the beginning 
power=station was used for substations and power=generator for power stations. 
While you can adopt to oddities like these, they surely raise the bar for new 
mappers, introduce unnecessary complexity and tend to lead to tagging errors by 
people less aware of wiki definitions.

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-22 Thread Tomas Straupis 
My question/proposal was about what to do with failed proposals in
general. That is:
1. How to identify a "failed" proposal
2. What to do with it

My proposal for point 1 is:
If after say two years new schema does not get at least equal tagging
count as the old schema - proposal failed.

My proposal for point 2 is:
Mark that proposal as "failed" in wiki. Mark all wiki pages which were
marked as "deprecated" because of the failed proposal as "valid".

P.S. This only influences proposals which are CHANGING tagging.
P.P.S. There should also be guards against such proposals in the first
place, but lets park it for now.

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-21 Thread Greg Troxel 

Martin Koppenhoefer  writes:

> sent from a phone
>
>> Il giorno 20 giu 2016, alle ore 12:04, Tomas Straupis 
>>  ha scritto:
>> 
>>  My main point is to get back to reservoir/basin being tagged as "landuse"
>
>
> why would that be desirable? Basically landuse is a property of land,
> and generally it's not very clear how to apply (it depends on the
> scale, and our db doesn't have a scale). As opposed to this, mapping a
> reservoir or a basin as a feature is much clearer, you don't have to
> worry whether you include auxiliary stuff like the service road
> leading to the reservoir, or the non-water-storage but legally
> associated areas around it to the feature (you won't).

I don't really see this as a conflict.   We need to have two features
marked in the db.

One is the area of land in use for the reservoir, including the water,
the roads/buildings, and the associated protection area (where most
activities are prohibited, except possiby hiking).  This should be a
landuse= tag.  It does not denote water and should not render blue :-)

Then, there should be some kind of water tag that denotes the area that
contains water or normally contains water.

I don't see any reason why this can't be

  landuse=reservoir [entire parcel that has the reservoir on it, or the
  region that has "water supply area - no trespassing" signs, etc.]]

  water=reservoir [the water part]



signature.asc
Description: PGP signature
___
talk mailing list
talk@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
>>  My main point is to get back to reservoir/basin being tagged as "landuse"
> why would that be desirable?

  There will always be more than one opinion on which naming of tags
is "better" because there is no "universal best way" (unless it's
"42").
  What I'm striving for is STABILITY for tagging.

  Landuse reservoir/basin/etc was there before the water proposal with
hundreds of thousands objects tagged at the time of proposal and it
still is used on more objects than water proposal. That is the reason
why I suggest to roll back to landuse version and make rules/guards so
that such "water" proposals would not go through without a VERY good
reason in the future.

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 
2016-06-20 17:30 GMT+02:00 Lester Caine :

> The simple fact is that there is not a consistent structure for
> identifying 'landcover' on OSM and even natural=wood and landuse=forest
> make it difficult to decide what is naturally occurring and what is man
> made.
>
landuse=reservoir is a lot more practical where at times of the
> year the majority of the surface area is exposed. That is a totally man
> made situation for which 'natural' does not apply.
>


generally, reading "natural=*" as "made by mother nature" is a bogus
interpretation in my view. It is just a kind of feature-group (geographical
/ landscape feature) for things.




> And when moving onto
> areas like marinas which take several forms including basins on the
> waterway system, including land elements as 'retail' or 'residential'
> and water elements as waterway tags as part of the Relation:waterway.
> But the overall area's landuse is marina even if we currently tag it as
> leisure=marina without any agreement as to just what area that should
> cover.
>


the tag should cover the whole marina, is this difficult to apply?



>
> It's the insistence that water only applies to natural elements which
> just does not fit properly, and man_made=reservoir while much more
> accurate does not fit in with a consistent landcover/landuse overlay?
>


man_made=reservoir would be an option for reservoirs (nothing I would
introduce, just another tag for what already has 2 alternative mapping
methods).
What do you mean by "consistent landcover/landuse overlay"? Those are
orthogonal concepts, with often different boundaries, displaying both (all
of it) at the same time will most likely lead to un unreadable map.

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Lester Caine 
On 20/06/16 15:38, Martin Koppenhoefer wrote:
>> > Il giorno 20 giu 2016, alle ore 12:04, Tomas Straupis 
>> >  ha scritto:
>> > 
>> >  My main point is to get back to reservoir/basin being tagged as "landuse"
> 
> why would that be desirable? Basically landuse is a property of land, and 
> generally it's not very clear how to apply (it depends on the scale, and our 
> db doesn't have a scale). As opposed to this, mapping a reservoir or a basin 
> as a feature is much clearer, you don't have to worry whether you include 
> auxiliary stuff like the service road leading to the reservoir, or the 
> non-water-storage but legally associated areas around it to the feature (you 
> won't).

The simple fact is that there is not a consistent structure for
identifying 'landcover' on OSM and even natural=wood and landuse=forest
make it difficult to decide what is naturally occurring and what is man
made. landuse=reservoir is a lot more practical where at times of the
year the majority of the surface area is exposed. That is a totally man
made situation for which 'natural' does not apply. And when moving onto
areas like marinas which take several forms including basins on the
waterway system, including land elements as 'retail' or 'residential'
and water elements as waterway tags as part of the Relation:waterway.
But the overall area's landuse is marina even if we currently tag it as
leisure=marina without any agreement as to just what area that should cover.

It's the insistence that water only applies to natural elements which
just does not fit properly, and man_made=reservoir while much more
accurate does not fit in with a consistent landcover/landuse overlay?

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 


sent from a phone

> Il giorno 20 giu 2016, alle ore 12:04, Tomas Straupis 
>  ha scritto:
> 
>  My main point is to get back to reservoir/basin being tagged as "landuse"


why would that be desirable? Basically landuse is a property of land, and 
generally it's not very clear how to apply (it depends on the scale, and our db 
doesn't have a scale). As opposed to this, mapping a reservoir or a basin as a 
feature is much clearer, you don't have to worry whether you include auxiliary 
stuff like the service road leading to the reservoir, or the non-water-storage 
but legally associated areas around it to the feature (you won't).

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
> You need to decide if you want to abolish the water=* or if you just
> prefer using waterway=riverbank instead of natural=water +
> water=river - which does not in any way conflict with the water=* tag.

  Once again: I do not want to abolish water=*.

  My main point is to get back to reservoir/basin being tagged as "landuse".

  So basically getting back to the state things were before the water
proposal (and adding water=* for those who want to tag more details
about natural water bodies, because adding water=* does not break
anything while deprecating widespread use of
landuse=reservoir/basin/etc is breaking a lot of stuff).

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Christoph Hormann 
On Monday 20 June 2016, Tomas Straupis wrote:
> >
> > If you want to eliminate use of water=* from OSM you'd need to
> > convince the community of this.  A formal proposal can be used but
> > without convincing arguments on the matter this stands little
> > chance in being approved.
>
>   I do not want to „eliminate“ water=*
>   I want to go back to the situation before the water proposal - with
> landuse=reservoir, waterway=riverbank, landuse=basin, etc. etc. As it
> used to be, and as it was and is still being mapped.

You need to decide if you want to abolish the water=* or if you just 
prefer using waterway=riverbank instead of natural=water + 
water=river - which does not in any way conflict with the water=* tag.

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

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Christoph Hormann 
On Monday 20 June 2016, you wrote:
>
> I'd like to add to this that on a semantic / natural language level,
> waterway=riverbank (deliberately ignoring long standing, widespread
> use and acceptance) would seem to indicate a riverbank, i.e. the bank
> of a river, or in other words, the area along a river, which will
> occassionally but not always be flooded.

Indeed - not separating actual water mapping from mapping geomorphology 
is one of the primary disadvantages of the waterway=riverbank tag (for 
which as said there are advantages too) - and is partly responsible for 
quite a few cases where water mapping covers the whole floodplain of a 
braided river like here:

http://www.openstreetmap.org/#map=12/46.0764/12.8456

Similar situation by the way with landuse=reservoir/landuse=basin - 
acutal presence of water vs. dedication of an area for certain use.

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

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 
2016-06-20 11:29 GMT+02:00 Tomas Straupis :

>   My main point is that existing tagging (especially widely used one)
> should not be changed unless it gives some ontological benefit (new
> features/properties being added, features split etc.).
>


actually you do not have to change any existing tagging when you implement
the "new" water=* tags, both can coexist without problems. I do agree that
changing (i.e. removing of established tags) is not the right approach to
add more detail.

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 
2016-06-20 11:34 GMT+02:00 Tomas Straupis :

> > actually the way it was before HAD big issues, you could not even state
> if
> > something was a lake or just the basin of a fountain (most kind of water
> > areas just mapped as natural=water).
>
>   Everything what can be mapped with new water schema can be (and is)
> mapped with old schema.
>


no, or at least not with the same semantic detail (one example is in the
text you have cited above)




>
>   The problem with newbies adding everything as natural=water did not
> go away. Now most iD edits create natural=water|water=pond even if it
> is a reservoir or a lake. So once again - no ontological difference
> here.



Now you are mixing up tag semantics (what can be expressed with the tags)
with people not applying them correctly (e.g. supposedly for editing
software weaknesses), and then draw the conclusion that the tags are bad.

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
> actually the way it was before HAD big issues, you could not even state if
> something was a lake or just the basin of a fountain (most kind of water
> areas just mapped as natural=water).

  Everything what can be mapped with new water schema can be (and is)
mapped with old schema.

  The problem with newbies adding everything as natural=water did not
go away. Now most iD edits create natural=water|water=pond even if it
is a reservoir or a lake. So once again - no ontological difference
here.

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 
2016-06-20 11:14 GMT+02:00 Tomas Straupis :

>   I do not want to „eliminate“ water=*
>   I want to go back to the situation before the water proposal - with
> landuse=reservoir, waterway=riverbank, landuse=basin, etc. etc. As it
> used to be, and as it was and is still being mapped.
>


actually the way it was before HAD big issues, you could not even state if
something was a lake or just the basin of a fountain (most kind of water
areas just mapped as natural=water).

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
> I'd like to add to this that on a semantic / natural language level,
> waterway=riverbank (deliberately ignoring long standing, widespread use and
> acceptance) would seem to indicate a riverbank, i.e. the bank of a river, or
> in other words, the area along a river, which will occassionally but not
> always be flooded.
>
> I am at this point not proposing to remap all of these (unless there would
> be compelling agreement by many mappers), as this is a longstanding, very
> widespread tag (293.000 occurences, 31.700 of them relations) with
> supposedly uniform usage, but it should be noted that there are issues with
> it on a logical level.

  My main point is that existing tagging (especially widely used one)
should not be changed unless it gives some ontological benefit (new
features/properties being added, features split etc.). I think that
data consumers want stability more than they want tag names to be
defined with "more correct" words (especially then lots of consumers
are not from English speaking countries).

  This is why water proposal was wrong in the first place. I agree
that it somehow slipped (I accept my guilt as well because I didn't do
anything at the time of proposal). But if after five years mappers did
not accept it by mapping, maybe it is not too late to revert the
proposal (not tags by automated edits).

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
> You are either deliberately or due to misinformation distorting things
> here.  The water=* is widely used and accepted, there are >700k uses in
> line with the proposal (an additional 255k for the deprecated
> water=intermittent).
>
> The waterway=riverbank tag is considered equivalent to natural=water +
> water=river, mappers may use either depending on what they prefer.
> There are good arguments for either of these options.
>
> If you want to eliminate use of water=* from OSM you'd need to convince
> the community of this.  A formal proposal can be used but without
> convincing arguments on the matter this stands little chance in being
> approved.

  I do not want to „eliminate“ water=*
  I want to go back to the situation before the water proposal - with
landuse=reservoir, waterway=riverbank, landuse=basin, etc. etc. As it
used to be, and as it was and is still being mapped.

> You however must not retag features with those tags to something else
> just for its own sake (i.e. outside normal mapping) - this would not be
> acceptable and would ultimately lead to reversal of such changes and
> possibly bans from editing.

  At least in Lithuania we have an agreement for YEARS as how to map
water bodies and we ARE updating mapping in Lithuania according to
those agreements, because this is what data consumers are expecting.

> If you want to do something productive you could clean up the frequent
> occurences of duplicate and sometimes contradicting tags on member ways
> and multipolygon relations for river mapping.  One of the problems of
> the waterway=riverbank tag is that it was originally meant and is
> widely understood to be a way tag while today it should normally be
> applied to the multipolygons relation.  Cleaning up such ambiguities -
> not mechanically as it has been suggested in the past but with
> individual verification - using either waterway=riverbank or
> natural=water + water=river would be a very good deed.

  If in the final GIS database we get a POLYGON with
waterway=riverbank I see no difference as to how it was mapped - as a
way, or as a relation.

  If you are referring to the old problem where tags are placed on
outer way in relation rather than a relation itself then yes, that is
a problem but it has nothing to do with discussion about water
tagging.

-- 
Tomas

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Martin Koppenhoefer 
2016-06-20 10:48 GMT+02:00 Christoph Hormann :

> If you want to do something productive you could clean up the frequent
> occurences of duplicate and sometimes contradicting tags on member ways
> and multipolygon relations for river mapping.  One of the problems of
> the waterway=riverbank tag is that it was originally meant and is
> widely understood to be a way tag while today it should normally be
> applied to the multipolygons relation.  Cleaning up such ambiguities -
> not mechanically as it has been suggested in the past but with
> individual verification - using either waterway=riverbank or
> natural=water + water=river would be a very good deed.
>


I'd like to add to this that on a semantic / natural language level,
waterway=riverbank (deliberately ignoring long standing, widespread use and
acceptance) would seem to indicate a riverbank, i.e. the bank of a river,
or in other words, the area along a river, which will occassionally but not
always be flooded.

I am at this point not proposing to remap all of these (unless there would
be compelling agreement by many mappers), as this is a longstanding, very
widespread tag (293.000 occurences, 31.700 of them relations) with
supposedly uniform usage, but it should be noted that there are issues with
it on a logical level.

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Christoph Hormann 
On Monday 20 June 2016, Tomas Straupis wrote:
> 2016-06-19 23:35 GMT+03:00 Ilya Zverev:
> > <...> the proposal about water=* was
> > accepted by 16 mappers, and if you have a problem with that, then I
> > agree that we should change our proposal process, but in all these
> > years nobody has even started.
>
>   Accepted by 16(!) wiki editors but ignored by thousands of mappers.
>   5 years after acceptance according to tagwatch:
>   "new" tag usage
>   water=reservoir 79937
>   water=riverbank 1085
>
>   "old" tag usage
>   landuse=reservoir 387793
>   waterway=riverbank 293319

You are either deliberately or due to misinformation distorting things 
here.  The water=* is widely used and accepted, there are >700k uses in 
line with the proposal (an additional 255k for the deprecated 
water=intermittent).

The waterway=riverbank tag is considered equivalent to natural=water + 
water=river, mappers may use either depending on what they prefer.  
There are good arguments for either of these options.

If you want to eliminate use of water=* from OSM you'd need to convince 
the community of this.  A formal proposal can be used but without 
convincing arguments on the matter this stands little chance in being 
approved.

You however must not retag features with those tags to something else 
just for its own sake (i.e. outside normal mapping) - this would not be 
acceptable and would ultimately lead to reversal of such changes and 
possibly bans from editing.

If you want to do something productive you could clean up the frequent 
occurences of duplicate and sometimes contradicting tags on member ways 
and multipolygon relations for river mapping.  One of the problems of 
the waterway=riverbank tag is that it was originally meant and is 
widely understood to be a way tag while today it should normally be 
applied to the multipolygons relation.  Cleaning up such ambiguities - 
not mechanically as it has been suggested in the past but with 
individual verification - using either waterway=riverbank or 
natural=water + water=river would be a very good deed.

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

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

Re: [OSM-talk] Failed water proposal reversal

2016-06-20 Thread Lester Caine 
On 20/06/16 08:54, Tomas Straupis wrote:
>   Accepted by 16(!) wiki editors but ignored by thousands of mappers.
>   5 years after acceptance according to tagwatch:
>   "new" tag usage
>   water=reservoir 79937
>   water=riverbank 1085
> 
>   "old" tag usage
>   landuse=reservoir 387793
>   waterway=riverbank 293319

The very first check of a 'reservoir' may well fail. One does not tag an
artificially created body of water as 'natural' and many of the 'water'
extensions to natural=water directly describe man made features. And the
Idea that natural=water should wrap what are quite clearly man made
waterway features just reinforces that and while the effects are small,
many parts of the world even 'coastline' is now a man made construction.

But the 'tags' list is probably the place to open a discussion on this
... not that I bother following it ;)

-- 
Lester Caine - G8HFL
-
Contact - http://lsces.co.uk/wiki/?page=contact
L.S.Caine Electronic Services - http://lsces.co.uk
EnquirySolve - http://enquirysolve.com/
Model Engineers Digital Workshop - http://medw.co.uk
Rainbow Digital Media - http://rainbowdigitalmedia.co.uk

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

[OSM-talk] Failed water proposal reversal

2016-06-20 Thread Tomas Straupis 
2016-06-19 23:35 GMT+03:00 Ilya Zverev:
> <...> the proposal about water=* was
> accepted by 16 mappers, and if you have a problem with that, then I agree
> that we should change our proposal process, but in all these years nobody
> has even started.

  Accepted by 16(!) wiki editors but ignored by thousands of mappers.
  5 years after acceptance according to tagwatch:
  "new" tag usage
  water=reservoir 79937
  water=riverbank 1085

  "old" tag usage
  landuse=reservoir 387793
  waterway=riverbank 293319

  If data consumers would read wiki to learn how things are already
mapped they would get wrong information.

  What would be the process of reconsidering water proposal?
  1. Wiki proposal for "unproposing" :-)
  2. De facto reversal (based on actual mapping)

  I personally do not see wiki as a good place to do that because wiki
does have a more or less separate group of people. I would assume most
mappers are not watching wiki for proposals, so basically we would get
back to the same 16 wiki editors.

P.S. Sorry for "tank accusation", must have mixed with somebody else.

-- 
Tomas

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