Feature proposal - RFC - Utility markers

classic Classic list List threaded Threaded
14 messages Options
Reply | Threaded
Open this post in threaded view
|

Feature proposal - RFC - Utility markers

François Lacombe-2
Hi all,

Here is another proposal we were two working on it.
It regards several kinds of utility markers usually warning about buried infrastructure beneath them.

Markers are currently described with keys like pipeline=* and power=* although they're not directly involved in infrastructure running processes (like a valve can be on a pipeline for instance).
Then it can be useful to define a new key marker=* to gather more categories on OSM (pipeline is for now the most mapped here) and prevent pipeline, power and telecom keys be cluttered with not directly related features.

Note that markers mapping is important on OSM as location signs and relevant data to verify presence of not visible infrastructures.

Feel free to raise concerns here and on talk page.

All the best

François

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

Joseph Eisenberg
It looks like the main effect of this proposal would be to replace pipeline=marker (used 29k times) with marker=pipeline, though the new key marker= could also be used for power cables and telecommunications cables.

Is it really necessary to change pipeline=marker?

-Joseph

On Sun, Jul 14, 2019 at 11:10 PM François Lacombe <[hidden email]> wrote:
Hi all,

Here is another proposal we were two working on it.
It regards several kinds of utility markers usually warning about buried infrastructure beneath them.

Markers are currently described with keys like pipeline=* and power=* although they're not directly involved in infrastructure running processes (like a valve can be on a pipeline for instance).
Then it can be useful to define a new key marker=* to gather more categories on OSM (pipeline is for now the most mapped here) and prevent pipeline, power and telecom keys be cluttered with not directly related features.

Note that markers mapping is important on OSM as location signs and relevant data to verify presence of not visible infrastructures.

Feel free to raise concerns here and on talk page.

All the best

François
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

François Lacombe-2
Hi Jospeh

This proposal is an attempt to bring consistency in markers mapping, in two ways :
- Provide a common concept to tag them all.
- Free pipeline=* from some features unrelated directly to pipeline operation.

Second point should encourage a mapping good practice I didn't have in mind in previous pipeline mapping evolutions : the marker shouldn't be part of the pipeline way directly as it warns about the presence of pipelines in a given range or distances.
Just like road signs should get their own node beside the road instead of be part the highway way.
To me yes, we should encourage to use marker=pipeline instead of pipeline=marker prior to the last gets *really* used.
29k features is less than the whole amount of pipeline markers we have to find in France (which is a small area).

All the best

François

Le jeu. 18 juil. 2019 à 06:07, Joseph Eisenberg <[hidden email]> a écrit :
It looks like the main effect of this proposal would be to replace pipeline=marker (used 29k times) with marker=pipeline, though the new key marker= could also be used for power cables and telecommunications cables.

Is it really necessary to change pipeline=marker?

-Joseph

On Sun, Jul 14, 2019 at 11:10 PM François Lacombe <[hidden email]> wrote:
Hi all,

Here is another proposal we were two working on it.
It regards several kinds of utility markers usually warning about buried infrastructure beneath them.

Markers are currently described with keys like pipeline=* and power=* although they're not directly involved in infrastructure running processes (like a valve can be on a pipeline for instance).
Then it can be useful to define a new key marker=* to gather more categories on OSM (pipeline is for now the most mapped here) and prevent pipeline, power and telecom keys be cluttered with not directly related features.

Note that markers mapping is important on OSM as location signs and relevant data to verify presence of not visible infrastructures.

Feel free to raise concerns here and on talk page.

All the best

François
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

François Lacombe-2
Hi all

The proposal to introduce marker=* key for all kind of utility markers is about to be voted.

All previous comments have been solved and any new one will be welcome.

Currently, more than 6k object are described with undocumented marker=stone which conflicts a bit with proposed marker classification.
As those are mainly (to be determined on each situation) highway milestones or private ground limits, they're not covered by this proposal
A suggestion would to define marker=milestone or marker=land_limit + support=pedestal + material=stone.

All the best

François

Le ven. 19 juil. 2019 à 21:22, François Lacombe <[hidden email]> a écrit :
Hi Jospeh

This proposal is an attempt to bring consistency in markers mapping, in two ways :
- Provide a common concept to tag them all.
- Free pipeline=* from some features unrelated directly to pipeline operation.

Second point should encourage a mapping good practice I didn't have in mind in previous pipeline mapping evolutions : the marker shouldn't be part of the pipeline way directly as it warns about the presence of pipelines in a given range or distances.
Just like road signs should get their own node beside the road instead of be part the highway way.
To me yes, we should encourage to use marker=pipeline instead of pipeline=marker prior to the last gets *really* used.
29k features is less than the whole amount of pipeline markers we have to find in France (which is a small area).

All the best

François

Le jeu. 18 juil. 2019 à 06:07, Joseph Eisenberg <[hidden email]> a écrit :
It looks like the main effect of this proposal would be to replace pipeline=marker (used 29k times) with marker=pipeline, though the new key marker= could also be used for power cables and telecommunications cables.

Is it really necessary to change pipeline=marker?

-Joseph

On Sun, Jul 14, 2019 at 11:10 PM François Lacombe <[hidden email]> wrote:
Hi all,

Here is another proposal we were two working on it.
It regards several kinds of utility markers usually warning about buried infrastructure beneath them.

Markers are currently described with keys like pipeline=* and power=* although they're not directly involved in infrastructure running processes (like a valve can be on a pipeline for instance).
Then it can be useful to define a new key marker=* to gather more categories on OSM (pipeline is for now the most mapped here) and prevent pipeline, power and telecom keys be cluttered with not directly related features.

Note that markers mapping is important on OSM as location signs and relevant data to verify presence of not visible infrastructures.

Feel free to raise concerns here and on talk page.

All the best

François
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

Joseph Eisenberg
I'm still opposed to this proposal:

This proposal is quite long and complicated-looking. I believe it
would be better to clarify exactly what tags are new: for example,
"support", "material" etc are existing tags, not new tags. Please
update the "Proposal" section at the top to clearly state the tags
that would be added, and the tags that would be deprecated.

I believe there are 2 tags that are being deprecated: pipeline=marker
and marker=stone. I don't see the benefit in moving the first from the
pipeline=* key, where it's really clear that "this is a marker for a
pipeline" to a new marker key, where the values will be mixed between
power, communications, pipeline and fire hydrand features (and
possibly others in the future).

I also think that it's not reasonable to deprecate marker=stone
without clearly discussing what tag is supposed to replace it.
According to taginfo, almost all uses of marker=stone are combined
with boundary=marker, so these are boundary marker stones, "a robust
physical marker that identifies the start of a land boundary or the
change in a boundary, especially a change in direction of a boundary."

- Joseph Eisenberg

On 9/6/19, François Lacombe <[hidden email]> wrote:

> Hi all
>
> The proposal to introduce marker=* key for all kind of utility markers is
> about to be voted.
> https://wiki.openstreetmap.org/wiki/Proposed_features/Utility_markers_proposal
>
> All previous comments have been solved and any new one will be welcome.
>
> Currently, more than 6k object are described with undocumented marker=stone
> which conflicts a bit with proposed marker classification.
> As those are mainly (to be determined on each situation) highway milestones
> or private ground limits, they're not covered by this proposal
> A suggestion would to define marker=milestone or marker=land_limit +
> support=pedestal + material=stone.
>
> All the best
>
> François
>
> Le ven. 19 juil. 2019 à 21:22, François Lacombe <[hidden email]>
> a écrit :
>
>> Hi Jospeh
>>
>> This proposal is an attempt to bring consistency in markers mapping, in
>> two ways :
>> - Provide a common concept to tag them all.
>> - Free pipeline=* from some features unrelated directly to pipeline
>> operation.
>>
>> Second point should encourage a mapping good practice I didn't have in
>> mind in previous pipeline mapping evolutions : the marker shouldn't be
>> part
>> of the pipeline way directly as it warns about the presence of pipelines
>> in
>> a given range or distances.
>> Just like road signs should get their own node beside the road instead of
>> be part the highway way.
>> To me yes, we should encourage to use marker=pipeline instead of
>> pipeline=marker prior to the last gets *really* used.
>> 29k features is less than the whole amount of pipeline markers we have to
>> find in France (which is a small area).
>>
>> All the best
>>
>> François
>>
>> Le jeu. 18 juil. 2019 à 06:07, Joseph Eisenberg <
>> [hidden email]> a écrit :
>>
>>> It looks like the main effect of this proposal would be to replace
>>> pipeline=marker (used 29k times) with marker=pipeline, though the new
>>> key
>>> marker= could also be used for power cables and telecommunications
>>> cables.
>>>
>>> Is it really necessary to change pipeline=marker?
>>>
>>> -Joseph
>>>
>>> On Sun, Jul 14, 2019 at 11:10 PM François Lacombe <
>>> [hidden email]> wrote:
>>>
>>>> Hi all,
>>>>
>>>> Here is another proposal we were two working on it.
>>>> It regards several kinds of utility markers usually warning about
>>>> buried
>>>> infrastructure beneath them.
>>>>
>>>> https://wiki.openstreetmap.org/wiki/Proposed_features/Utility_markers_proposal
>>>>
>>>> Markers are currently described with keys like pipeline=* and power=*
>>>> although they're not directly involved in infrastructure running
>>>> processes
>>>> (like a valve can be on a pipeline for instance).
>>>> Then it can be useful to define a new key marker=* to gather more
>>>> categories on OSM (pipeline is for now the most mapped here) and
>>>> prevent
>>>> pipeline, power and telecom keys be cluttered with not directly related
>>>> features.
>>>>
>>>> Note that markers mapping is important on OSM as location signs and
>>>> relevant data to verify presence of not visible infrastructures.
>>>>
>>>> Feel free to raise concerns here and on talk page.
>>>>
>>>> All the best
>>>>
>>>> François
>>>> _______________________________________________
>>>> Tagging mailing list
>>>> [hidden email]
>>>> https://lists.openstreetmap.org/listinfo/tagging
>>>>
>>> _______________________________________________
>>> Tagging mailing list
>>> [hidden email]
>>> https://lists.openstreetmap.org/listinfo/tagging
>>>
>>
>

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

Jez Nicholson
Arriving fresh to a proposal, my first action would be to look at what is currently in OSM. There are 6,043 "marker"="stone", which is 81.5% of the usage of "marker" in OSM. I would expect the proposal to support current usage.

I would then look at "power":"marker" and be very concerned to see 35,288 tags. That's a very strong existing usage. You might be lucky that power markers aren't as useful to render as power lines, etc. https://openinframap.org/#12.2/49.49246/0.21175

On Fri, Sep 6, 2019 at 8:13 AM Joseph Eisenberg <[hidden email]> wrote:
I'm still opposed to this proposal:

This proposal is quite long and complicated-looking. I believe it
would be better to clarify exactly what tags are new: for example,
"support", "material" etc are existing tags, not new tags. Please
update the "Proposal" section at the top to clearly state the tags
that would be added, and the tags that would be deprecated.

I believe there are 2 tags that are being deprecated: pipeline=marker
and marker=stone. I don't see the benefit in moving the first from the
pipeline=* key, where it's really clear that "this is a marker for a
pipeline" to a new marker key, where the values will be mixed between
power, communications, pipeline and fire hydrand features (and
possibly others in the future).

I also think that it's not reasonable to deprecate marker=stone
without clearly discussing what tag is supposed to replace it.
According to taginfo, almost all uses of marker=stone are combined
with boundary=marker, so these are boundary marker stones, "a robust
physical marker that identifies the start of a land boundary or the
change in a boundary, especially a change in direction of a boundary."

- Joseph Eisenberg

On 9/6/19, François Lacombe <[hidden email]> wrote:
> Hi all
>
> The proposal to introduce marker=* key for all kind of utility markers is
> about to be voted.
> https://wiki.openstreetmap.org/wiki/Proposed_features/Utility_markers_proposal
>
> All previous comments have been solved and any new one will be welcome.
>
> Currently, more than 6k object are described with undocumented marker=stone
> which conflicts a bit with proposed marker classification.
> As those are mainly (to be determined on each situation) highway milestones
> or private ground limits, they're not covered by this proposal
> A suggestion would to define marker=milestone or marker=land_limit +
> support=pedestal + material=stone.
>
> All the best
>
> François
>
> Le ven. 19 juil. 2019 à 21:22, François Lacombe <[hidden email]>
> a écrit :
>
>> Hi Jospeh
>>
>> This proposal is an attempt to bring consistency in markers mapping, in
>> two ways :
>> - Provide a common concept to tag them all.
>> - Free pipeline=* from some features unrelated directly to pipeline
>> operation.
>>
>> Second point should encourage a mapping good practice I didn't have in
>> mind in previous pipeline mapping evolutions : the marker shouldn't be
>> part
>> of the pipeline way directly as it warns about the presence of pipelines
>> in
>> a given range or distances.
>> Just like road signs should get their own node beside the road instead of
>> be part the highway way.
>> To me yes, we should encourage to use marker=pipeline instead of
>> pipeline=marker prior to the last gets *really* used.
>> 29k features is less than the whole amount of pipeline markers we have to
>> find in France (which is a small area).
>>
>> All the best
>>
>> François
>>
>> Le jeu. 18 juil. 2019 à 06:07, Joseph Eisenberg <
>> [hidden email]> a écrit :
>>
>>> It looks like the main effect of this proposal would be to replace
>>> pipeline=marker (used 29k times) with marker=pipeline, though the new
>>> key
>>> marker= could also be used for power cables and telecommunications
>>> cables.
>>>
>>> Is it really necessary to change pipeline=marker?
>>>
>>> -Joseph
>>>
>>> On Sun, Jul 14, 2019 at 11:10 PM François Lacombe <
>>> [hidden email]> wrote:
>>>
>>>> Hi all,
>>>>
>>>> Here is another proposal we were two working on it.
>>>> It regards several kinds of utility markers usually warning about
>>>> buried
>>>> infrastructure beneath them.
>>>>
>>>> https://wiki.openstreetmap.org/wiki/Proposed_features/Utility_markers_proposal
>>>>
>>>> Markers are currently described with keys like pipeline=* and power=*
>>>> although they're not directly involved in infrastructure running
>>>> processes
>>>> (like a valve can be on a pipeline for instance).
>>>> Then it can be useful to define a new key marker=* to gather more
>>>> categories on OSM (pipeline is for now the most mapped here) and
>>>> prevent
>>>> pipeline, power and telecom keys be cluttered with not directly related
>>>> features.
>>>>
>>>> Note that markers mapping is important on OSM as location signs and
>>>> relevant data to verify presence of not visible infrastructures.
>>>>
>>>> Feel free to raise concerns here and on talk page.
>>>>
>>>> All the best
>>>>
>>>> François
>>>> _______________________________________________
>>>> Tagging mailing list
>>>> [hidden email]
>>>> https://lists.openstreetmap.org/listinfo/tagging
>>>>
>>> _______________________________________________
>>> Tagging mailing list
>>> [hidden email]
>>> https://lists.openstreetmap.org/listinfo/tagging
>>>
>>
>

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

dieterdreist
I agree with your definition for marker:
marker=<marker nature>     The node corresponds to a marker seen on ground

but then the actual values that you propose do not (or only implicitly) refer to the marker but to the "thing" the marker refers to:

"pipeline" - Indicate that a pipeline is buried next to the marker, possibly underground and not mandatorly righly below

"power_cable" - A power cable is installed next to the marker, possibly underground and not mandatorly righly below

"telecom_cable" - A telecommunication cable is installed next to the marker, possibly underground and not mandatorly righly below

"fire_hydrant" - A fire hydrant is available next to the marker, possibily underground and not mandatorly rightly below



I would rather have expected a generic description of the marker, like marker=
post
cone
sign
...
aerial_marker (maybe this should be a property, not a type? This seems to be a quite interesting property for our context)

properties which might be useful for markers:
dome_marker=yes/no (or: marker_top=flat/dome/...)
placement(?)=soil / street


If this is for a _utility_ marker only, you should consider using a more specific key name like "utility_marker". I could imagine defining a generic "marker" tag for utility markers but including also other markers, which may be similar by their physical appearance, although for many of them, e.g. survey points, milestones and boundary markers, we already have established different tagging and will likely not change this.

Cheers,
Martin

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

marc marc
Le 06.09.19 à 12:17, Martin Koppenhoefer a écrit :
> placement(?)=soil / street

location seems better and already exist
https://wiki.openstreetmap.org/wiki/Key%3Alocation
_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

dieterdreist
Actually I just found out, placement also exists, but for a different purpose:

Cheers,
Martin

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

François Lacombe-2
In reply to this post by Joseph Eisenberg
Hi all,

Thank you for yout contributions

Le ven. 6 sept. 2019 à 09:13, Joseph Eisenberg <[hidden email]> a écrit :
I'm still opposed to this proposal:

Answers provided at

Le ven. 6 sept. 2019 à 09:18, Jez Nicholson <[hidden email]> a écrit :
Arriving fresh to a proposal, my first action would be to look at what is currently in OSM. There are 6,043 "marker"="stone", which is 81.5% of the usage of "marker" in OSM. I would expect the proposal to support current usage.
I respectably disagree on that point.
Biggest problem is that current usage isn't documented, and may not have been reviewed like we are doing right now.

This proposal aims to define values for marker=* to describe utility markers.
Despite marker=stone may be unconsistent with what is proposed, it doesn't make it incompatible and this proposal only notices this fact without thinking of deprecating it.

I would then look at "power":"marker" and be very concerned to see 35,288 tags. That's a very strong existing usage. You might be lucky that power markers aren't as useful to render as power lines, etc. https://openinframap.org/#12.2/49.49246/0.21175
I don't get the right term here. I see only 222 items for power=marker and nothing for power:marker=*
Which one are you refering to with 35k uses please?

One of the goals is precisely to get a comprehensive render with marker=* for many kind of markers, not only pipeline ones.

Le ven. 6 sept. 2019 à 12:20, Martin Koppenhoefer <[hidden email]> a écrit :
I would rather have expected a generic description of the marker, like marker=
post
cone
sign
...
aerial_marker (maybe this should be a property, not a type? This seems to be a quite interesting property for our context)

Ok this one is really interesting.
Why not using marker=* to give its nature and another key utility=* with values "gas", "power", "telecom", "water"... ?

marker=* + utility=* give a "utility marker", right?

All the best

François

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

Joseph Eisenberg
> I don't get the right term here. I see only 222 items for power=marker and nothing for power:marker=*. Which one are you refering to with 35k uses please?

Probably pipeline=marker, used 35k times:
https://taginfo.openstreetmap.org/tags/?key=pipeline&value=marker

> Why not using marker=* to give its nature and another key utility=* with values...

Because most mappers only add 1 tag to each new object. (Folks like
you and me are an exception - and a year ago, when I was new at this,
I only usually added 1 tag per feature). If an object can be described
with one tag, it's better to do this and create several tags, (e.g.
pipeline=marker, power=marker) rather than requiring each object to be
tagged with 2 or 3 or 4 tags. This saves mapper time and makes sure
that each object is fully described.

- Joseph

On 9/7/19, François Lacombe <[hidden email]> wrote:

> Hi all,
>
> Thank you for yout contributions
>
> Le ven. 6 sept. 2019 à 09:13, Joseph Eisenberg <[hidden email]>
> a écrit :
>
>> I'm still opposed to this proposal:
>>
>
> Answers provided at
> https://wiki.openstreetmap.org/wiki/Talk:Proposed_features/Utility_markers_proposal#Oppose_deprecating_pipeline.3Dmarker_and_marker.3Dstone
>
> Le ven. 6 sept. 2019 à 09:18, Jez Nicholson <[hidden email]> a
> écrit :
>
>> Arriving fresh to a proposal, my first action would be to look at what is
>> currently in OSM. There are 6,043 "marker"="stone", which is 81.5% of the
>> usage of "marker" in OSM. I would expect the proposal to support current
>> usage.
>>
> I respectably disagree on that point.
> Biggest problem is that current usage isn't documented, and may not have
> been reviewed like we are doing right now.
>
> This proposal aims to define values for marker=* to describe utility
> markers.
> Despite marker=stone may be unconsistent with what is proposed, it doesn't
> make it incompatible and this proposal only notices this fact without
> thinking of deprecating it.
>
> I would then look at "power":"marker" and be very concerned to see 35,288
>> tags. That's a very strong existing usage. You might be lucky that power
>> markers aren't as useful to render as power lines, etc.
>> https://openinframap.org/#12.2/49.49246/0.21175
>>
> I don't get the right term here. I see only 222 items for power=marker and
> nothing for power:marker=*
> Which one are you refering to with 35k uses please?
>
> One of the goals is precisely to get a comprehensive render with marker=*
> for many kind of markers, not only pipeline ones.
>
> Le ven. 6 sept. 2019 à 12:20, Martin Koppenhoefer <[hidden email]>
> a écrit :
>
>> I would rather have expected a generic description of the marker, like
>> marker=
>> post
>> cone
>> sign
>> ...
>> aerial_marker (maybe this should be a property, not a type? This seems to
>> be a quite interesting property for our context)
>>
>
> Ok this one is really interesting.
> Why not using marker=* to give its nature and another key utility=* with
> values "gas", "power", "telecom", "water"... ?
> Seems it is already used:
> https://taginfo.openstreetmap.org/keys/utility#values
>
> marker=* + utility=* give a "utility marker", right?
>
> All the best
>
> François
>

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

Mateusz Konieczny-3
In reply to this post by François Lacombe-2



7 Sep 2019, 01:36 by [hidden email]:
Ok this one is really interesting.
Why not using marker=* to give its nature and another key utility=* with values "gas", "power", "telecom", "water"... ?

Is it typical to map "it is marker of an
unknown kind" like splitting shop
and opening hours makes sense?
As many want to map shop and are
not interested in mapping opening
hours.

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

dieterdreist
In reply to this post by Joseph Eisenberg
Am Sa., 7. Sept. 2019 um 02:06 Uhr schrieb Joseph Eisenberg <[hidden email]>:
> Why not using marker=* to give its nature and another key utility=* with values...

Because most mappers only add 1 tag to each new object. (Folks like
you and me are an exception - and a year ago, when I was new at this,
I only usually added 1 tag per feature). If an object can be described
with one tag, it's better to do this and create several tags, (e.g.
pipeline=marker, power=marker) rather than requiring each object to be
tagged with 2 or 3 or 4 tags. This saves mapper time and makes sure
that each object is fully described.


I would expect utility marker mapping to be a special interest. Jane Mapper will not map these, or will be so excited about her discovery on the ground that she will be willing to look it up on a wiki page ;-)
Preset focused editors like iD could set two tags by selecting one preset, but it would require more specific identifiers/search terms for the presets. It could also be done in several phases (guided multi step process). Current working of presets I have seen, tend to reduce tagging possibilities and information depth for the sake of simplicity.

Cheers,
Martin


_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging
Reply | Threaded
Open this post in threaded view
|

Re: Feature proposal - RFC - Utility markers

François Lacombe-2
In reply to this post by Joseph Eisenberg
Hi everyone

Le sam. 7 sept. 2019 à 02:06, Joseph Eisenberg <[hidden email]> a écrit :
Because most mappers only add 1 tag to each new object. (Folks like
you and me are an exception - and a year ago, when I was new at this,
I only usually added 1 tag per feature). If an object can be described
with one tag, it's better to do this and create several tags, (e.g.
pipeline=marker, power=marker) rather than requiring each object to be
tagged with 2 or 3 or 4 tags. This saves mapper time and makes sure
that each object is fully described.

I understand the need of simplicity in chosen terms. Neverthess I can't imagine OSM with single-key objects as a principle.
+1 with Martin : if occasional mappers want to reduce their typing time, they will use presets in convenient editors like iD or JOSM.
This argument comes on many discussions and oppose kind of simplicity to semantic consistency. Tagging with several consistent tags could be more easily handled and versatile than one single key mixing concepts for reasons that are not necessarily shared by the whole community.

However, this proposal will be reworked to take care of this batch of comment, it's appreciable we can discuss these points here.

Le sam. 7 sept. 2019 à 07:17, Mateusz Konieczny <[hidden email]> a écrit :
Is it typical to map "it is marker of an
unknown kind" like splitting shop
and opening hours makes sense?

It makes sense to map "here is a marker what it looks like" without explicitly attaching it to a utility.
One mapper will describe what she/he sees, and a second will complete with her/his own knowledge.

Le sam. 7 sept. 2019 à 16:11, Martin Koppenhoefer <[hidden email]> a écrit :
I would expect utility marker mapping to be a special interest. Jane Mapper will not map these, or will be so excited about her discovery on the ground that she will be willing to look it up on a wiki page ;-)
I definitely have to meet Jane Mapper, Martin :d


All the best

François

_______________________________________________
Tagging mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/tagging