Re: [Tagging] Feature Proposal - RFC - (Phone)

[Philip Barnes]

We also map the phone number of phoneboxes using phone=.

We do not generally contact phoneboxes.

Phil (trigpoint)

On Tuesday, 8 October 2019, Martin Koppenhoefer wrote:
> 
> 
> sent from a phone
> 
> > On 8. Oct 2019, at 15:40, Colin Smale via Tagging 
> >  wrote:
> > 
> > In that case it makes perfect sense to consolidate onto one or the other. 
> > But if there are any perceived semantic differences, however subtle, then 
> > either we find some way to represent that using other tagging, or we accept 
> > that a certain nuance will be lost.
> > 
> 
> there could be phone numbers with automatic announcements, so “phone” will 
> still be valid, but contact:phone would not suit well. To give an example. It 
> cannot be seen from the “phone”-key that this is the case though. I’m happy 
> with loosing the subtle differences that may make  “contact:”-prefixed tags 
> slightly more specific, in exchange for more universally usable 
> “almost-equal” more generic tags without the prefix.
> 
> Cheers Martin

-- 
Sent from my Sailfish device
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Warin]

On 07/10/19 21:00, Valor Naram wrote:

> One problem with enforcing a single tag by mappers or preprocessing
> data before putting it in the database is that if there are subtle 
distinctions they are

> forever lost.

Sven gave us a list of tags which have exact meaning. So no 
distinctions. No differences, just another names.


e.g. `phone` is the same like `contact:phone`. You could just find one 
difference: The name. And to have just the names different is a bad 
thing. In such cases we can make life a lot easier by just using one 
tag AND NOT two tags for the same purpose!




Unfortunately there will always be rebels and errors.
Making the wiki only have one tag for it is fine, that should reduce the 
number of preprocessing replacements but not eliminate it for quite some 
time.


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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Martin Koppenhoefer]

sent from a phone

> On 8. Oct 2019, at 22:08, Colin Smale  wrote:
> 
> So the subtlety you are referring to, is that some phone numbers routinely 
> connect to a recording instead of a human.
>  
> How about phone:recorded_message=* which would leave room for phone=* for a 
> manned line, or recorded_message=yes, which would not?


if we wanted to tag specifically that a phone number is connected to a 
recording then a specific tag would make sense, it was just an example how 
phone can cover more cases than contact:phone, by being more generic in 
meaning. I didn’t imply we should make this distinction, or that it is 
significant (in terms of number of instances), it was an example to demonstrate 
that both tags aren’t completely identical, still I am ok with treating them as 
if they were (of course you can still add your phone:recorded_message tag or 
„answerphone” or whatever, and I would support it, as I see a significant 
enough difference between a number you call and someone real replying vs. a 
number that always gets answered automatically, to consider whether that merits 
a different tag).


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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Paul Allen]

On Tue, 8 Oct 2019 at 20:49, Martin Koppenhoefer 
wrote:

>
> yes , of course, sorry for stepping on your toes, I was being sarcastic to
> better make the point, but I am aware that there is some use for this (even
> around here it may occur that a house has a name but not a number).
>

Not normally a problem in iD anyway, because it has a set of fields for
entering address details.
Except when it doesn't, because the authors decided that particular type of
object shouldn't
have an address.  And even that isn't a great problem, because entering
addr: into the bare
tag area and pressing tab on the first suggestion gets the set of address
fields to appear.
Except when it doesn't because the authors decided that particular type of
object really
shouldn't have an address under any circumstances whatsoever, then you have
to enter
everything via bare tags (if you can remember the key values).

Something I found out a few hours ago when I added a watermill.  A working
watermill.  With
an address.  It seems that iD thinks watermills shouldn't have addresses
under any
circumstances whatsoever.

>
> Statistically, there are 95 million housenumbers and 690.000 housenames in
> the db as of now.
>

I'd guess a lot of those 690,000 are in Wales, going by my experience.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Colin Smale via Tagging]

On 2019-10-08 21:51, Martin Koppenhoefer wrote:

> On 8. Oct 2019, at 15:40, Colin Smale via Tagging  
> wrote: 
> 
>> In that case it makes perfect sense to consolidate onto one or the other. 
>> But if there are any perceived semantic differences, however subtle, then 
>> either we find some way to represent that using other tagging, or we accept 
>> that a certain nuance will be lost.
> 
> there could be phone numbers with automatic announcements, so "phone" will 
> still be valid, but contact:phone would not suit well. To give an example. It 
> cannot be seen from the "phone"-key that this is the case though. I'm happy 
> with loosing the subtle differences that may make  "contact:"-prefixed tags 
> slightly more specific, in exchange for more universally usable 
> "almost-equal" more generic tags without the prefix.

So the subtlety you are referring to, is that some phone numbers
routinely connect to a recording instead of a human. 

How about phone:recorded_message=* which would leave room for phone=*
for a manned line, or recorded_message=yes, which would not?___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

`contact:phone` and `phone` are exactly for the same purpose, no differences just different names.See https://wiki.osm.org/Key:contact~ Sören Reinecke alias Valor Naram Original Message Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Martin Koppenhoefer To: "Tag discussion, strategy and related tools" CC: sent from a phoneOn 8. Oct 2019, at 15:40, Colin Smale via Tagging <tagging@openstreetmap.org> wrote:In that case it makes perfect sense to consolidate onto one or the other. But if there are any perceived semantic differences, however subtle, then either we find some way to represent that using other tagging, or we accept that a certain nuance will be lost.
there could be phone numbers with automatic announcements, so “phone” will still be valid, but contact:phone would not suit well. To give an example. It cannot be seen from the “phone”-key that this is the case though. I’m happy with loosing the subtle differences that may make  “contact:”-prefixed tags slightly more specific, in exchange for more universally usable “almost-equal” more generic tags without the prefix.Cheers Martin ___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Martin Koppenhoefer]

sent from a phone

> On 8. Oct 2019, at 15:40, Colin Smale via Tagging  
> wrote:
> 
> In that case it makes perfect sense to consolidate onto one or the other. But 
> if there are any perceived semantic differences, however subtle, then either 
> we find some way to represent that using other tagging, or we accept that a 
> certain nuance will be lost.
> 

there could be phone numbers with automatic announcements, so “phone” will 
still be valid, but contact:phone would not suit well. To give an example. It 
cannot be seen from the “phone”-key that this is the case though. I’m happy 
with loosing the subtle differences that may make  “contact:”-prefixed tags 
slightly more specific, in exchange for more universally usable “almost-equal” 
more generic tags without the prefix.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Martin Koppenhoefer]

sent from a phone

> On 8. Oct 2019, at 15:14, Paul Allen  wrote:
> 
> Housename may be useless where you live but for some of us it is essential.


yes , of course, sorry for stepping on your toes, I was being sarcastic to 
better make the point, but I am aware that there is some use for this (even 
around here it may occur that a house has a name but not a number). 

Statistically, there are 95 million housenumbers and 690.000 housenames in the 
db as of now.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Markus]

On Tue, 8 Oct 2019, 11:29 marc marc,  wrote:

> Le 07.10.19 à 23:07, Martin Koppenhoefer a écrit :
> > let’s bury the contact: - prefix
>
> in this case, be logical and also propose to bury the prefix addr:
>

There's a difference between the two prefixes: several addr: tags together
form a complete address (a single addr: tag only gives part of an address),
while a single contact: tag already is a complete contact information. Thus
i think the addr: prefix makes more sense that the contact: prefix.

Regards
Markus

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Mateusz Konieczny]

8 Oct 2019, 11:25 by marc_marc_...@hotmail.com:

> Le 07.10.19 à 23:07, Martin Koppenhoefer a écrit :
>
>> let’s bury the contact: - prefix
>>
>
> in this case, be logical and also propose to bury the prefix addr:
>
Main difference is that contact: prefix has
an alternative that is and always was more popular.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Colin Smale via Tagging]

On 2019-10-08 13:25, Valor Naram wrote:

> A short summary of what we have so far:
> - Deprecation of `contact:phone` has some advantages: Key `phone` is used far 
> more often, Key `phone` is shorter to write and better to find in word 
> completion functions of editors like iD, Data users don't have to support two 
> methods of tagging phone numbers.
> - Deprecation of `contact:phone` has one disadvantage: It's not grouped 
> anymore and we have to solve this by creating a new wikipage which lists all 
> keys that can be used for contacting purposes (e.g. throw a contact tab like 
> the Babykarte has).

Has it therefore been determined that phone=* and contact:phone=* are
100% synonymous, and no subtle semantic differences exist between the
two? In that case it makes perfect sense to consolidate onto one or the
other. But if there are any perceived semantic differences, however
subtle, then either we find some way to represent that using other
tagging, or we accept that a certain nuance will be lost.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Paul Allen]

On Mon, 7 Oct 2019 at 22:08, Martin Koppenhoefer 
wrote:

not to speak about addr:housenumber and addr:housename (the latter is
> mostly useless
>

In some areas of the UK I've lived, all houses have numbers and a few also
have names
(which are viewed by many as an affectation).  In rural districts it is
very rare for a house
to have a number, instead it has a name.  In the town in the UK where I
currently live,
maybe 30% or more of houses have ONLY a name.  Perhaps some of those 30% had
numbers once, but if they did nobody uses them or knows them now.  I'm
pretty certain
that one a street around the corner from me, half of the houses could never
have had
numbers: the numbering sequence and the age of various houses shows this
https://www.openstreetmap.org/#map=19/52.08491/-4.65782
8/Caerhuan and 9/Homeleigh on Napier Street are much more recent than any
of the other
houses on Napier Street (with the exception of Y Gorlan, which probably
replaced some
earlier building).

Housename may be useless where you live but for some of us it is essential.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

A short summary of what we have so far:- Deprecation of `contact:phone` has some advantages: Key `phone` is used far more often, Key `phone` is shorter to write and better to find in word completion functions of editors like iD, Data users don't have to support two methods of tagging phone numbers.- Deprecation of `contact:phone` has one disadvantage: It's not grouped anymore and we have to solve this by creating a new wikipage which lists all keys that can be used for contacting purposes (e.g. throw a contact tab like the Babykarte has).~ Sören Reinecke alias Valor Naram Original Message Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Joseph Eisenberg To: "Tag discussion, strategy and related tools" CC: Re: “if you want to display a tab with all the means of contact, you just have to look for contact:”That doesn’t work, because “phone=“ is much more popular and probably always will be, and  the “contact:” prefix is not used terrible consistently (is a website actually a way to contact a feature? Usually not)Keys should be designed for the convenience and ease of use by mappers, because the time of individual mappers is by far the greatest value input into Openstreetmap. Database users and developers (myself included) need to do a little more work sometimes, to give thousands of mappers a little less work and a little more fun.I support deprecating “contact:phone” and just using “phone=“
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Joseph Eisenberg]

Re: “if you want to display a tab with all the means of contact, you just
have to look for contact:”

That doesn’t work, because “phone=“ is much more popular and probably
always will be, and  the “contact:” prefix is not used terrible
consistently (is a website actually a way to contact a feature? Usually not)

Keys should be designed for the convenience and ease of use by mappers,
because the time of individual mappers is by far the greatest value input
into Openstreetmap. Database users and developers (myself included) need to
do a little more work sometimes, to give thousands of mappers a little less
work and a little more fun.

I support deprecating “contact:phone” and just using “phone=“
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[marc marc]

Le 07.10.19 à 23:07, Martin Koppenhoefer a écrit :
> let’s bury the contact: - prefix

in this case, be logical and also propose to bury the prefix addr:

on the contrary, I believe that having a namespace to group the keys 
concerning the addr and another namespace to group the keys concerning 
the means of contact is a practical way to use the data (if you want to 
display a tab with all the means of contact, you just have to look for 
contact: without having to hard code them and having to code again the 
existence of a new social network for example)

the number of characters to write is a bad argument : focus on quality 
and not a:hn because it would be shorter to write than addr:housenumber
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

> let’s us all save a lot of typing and let’s bury the contact: - prefix.The time will come at least I will try to accomplish this.~ Sören Reinecke alias Valor Naram Original Message Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Martin Koppenhoefer To: "Tag discussion, strategy and related tools" CC: sent from a phone> On 7. Oct 2019, at 22:40, Kevin Kenny  wrote:> > I think that's a claim that needs to be demonstrated. Certainly, the> complexity of the contact:* schema and the variety of both editors and> data consumers has proven to be a barrier to widespread acceptance.frankly I am an old fashioned contributor and do not like presets. When I am going to add tags, I type them, the beginning, and usually tag completion will have the correct tag after 1-4 key strokes. I am sure there are more people like me, maybe we‘re a minority, but if we are I bet it’s significant. Now for this kind of workflow, namespaces are problematic. I already dislike farmyard and farmland for the 5 characters required, not to speak about addr:housenumber and addr:housename (the latter is mostly useless but as a is before u, this often forces me to type 12 characters before I can hit return). I have made peace with these tags as they are around for a long time, but they clearly aren’t the direction I would want the project to go more to. let’s us all save a lot of typing and let’s bury the contact: - prefix.Cheers Martin ___Tagging mailing listTagging@openstreetmap.orghttps://lists.openstreetmap.org/listinfo/tagging___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Martin Koppenhoefer]

sent from a phone

> On 7. Oct 2019, at 22:40, Kevin Kenny  wrote:
> 
> I think that's a claim that needs to be demonstrated. Certainly, the
> complexity of the contact:* schema and the variety of both editors and
> data consumers has proven to be a barrier to widespread acceptance.


frankly I am an old fashioned contributor and do not like presets. When I am 
going to add tags, I type them, the beginning, and usually tag completion will 
have the correct tag after 1-4 key strokes. I am sure there are more people 
like me, maybe we‘re a minority, but if we are I bet it’s significant. Now for 
this kind of workflow, namespaces are problematic. I already dislike farmyard 
and farmland for the 5 characters required, not to speak about addr:housenumber 
and addr:housename (the latter is mostly useless but as a is before u, this 
often forces me to type 12 characters before I can hit return). I have made 
peace with these tags as they are around for a long time, but they clearly 
aren’t the direction I would want the project to go more to. let’s us all save 
a lot of typing and let’s bury the contact: - prefix.

Cheers Martin 


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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Kevin Kenny]

On Mon, Oct 7, 2019 at 3:13 PM Michael Patrick  wrote:

> rarely are doers and users exposed to the full complexity, just the simple 
> subset of what is needed for a particular use case.

I think that's a claim that needs to be demonstrated. Certainly, the
complexity of the contact:* schema and the variety of both editors and
data consumers has proven to be a barrier to widespread acceptance.

> '... many on both sides who insist that their way is the one true way' is 
> essentially a manifestation of of the 'Blind men and an elephant' ( 
> https://en.wikipedia.org/wiki/Blind_men_and_an_elephant#The_parable ) 
> situation. While all the blind men will never agree, if the blind men with 
> the ear, the trunk, and the tusk ask questions of someone that can see the 
> whole elephant, they can come up with a functional meaning that covers the at 
> least the head. There's probably no 'tagging' situation in OSM that wasn't 
> solved by the resolution of Electronic data interchange (EDI) issues in the 
> period between 1970 to 1998.

Reconciling an EDI standard with a 'folksonomy' rather boggles the
mind. It would perhaps have been better had OSM started with more
structure than a bucket of 'keyword=value' tags, but it didn't.
(Moreover, since the space of feature types being mapped is still
expanding, the space of available tags needs to expand with it; a
fixed schema isn't quite going to work.)

Of course, pursuing tagging uniformity within reason is a worthy goal,
and I'm convinced you're on the side of the angels, but it's likely to
be a long uphill road to get from where we are to where you want to
be.

Also, obligatory XKCD: https://xkcd.com/927/

-- 
73 de ke9tv/2, Kevin

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

[Tagging] Feature Proposal - RFC - (Phone)

[Michael Patrick]

>  There are too many on both sides who insist that their way is the one
true way.

Actually, there is something pretty close to 'one TRUE way'. There are some
things in the Dataverse that are used by everybody, and everybody uses
everybody else's stuff. And when the Internet arrived they set up common
meanings for the very most common concepts. Like phone numbers.

Schema.org  is a collaborative, community activity
with a mission to create, maintain, and promote schemas for structured data
on the Internet, on web pages, in email messages, and beyond Founded by
Google, Microsoft, Yahoo and Yandex, Schema.org vocabularies are developed
by an open community  process,
using the public-schema...@w3.org
 mailing list and
through GitHub .

See click the ' JSON-LD ' tab on Example 2 at
https://schema.org/ContactPoint#HearingImpairedSupported-gen-208 ( on
https://schema.org/ContactPoint ) which shows how phone numbers play out
for even small businesses in the world.

Yea, yeah, it's complicated, or at least the specification is, but rarely
are doers and users exposed to the full complexity, just the simple subset
of what is needed for a particular use case. Something like a phone number
looks like a discrete 'thing', but it's not. It's a relationship between
other things.

'... many on both sides who insist that their way is the one true way' is
essentially a manifestation of of the 'Blind men and an elephant' (
https://en.wikipedia.org/wiki/Blind_men_and_an_elephant#The_parable )
situation. While all the blind men will never agree, if the blind men with
the ear, the trunk, and the tusk ask questions of someone that can see the
whole elephant, they can come up with a functional meaning that covers the
at least the head. There's probably no 'tagging' situation in OSM that
wasn't solved by the resolution of *Electronic data interchange* (*EDI*)
issues in the period between 1970 to 1998.

Michael Patrick
Data Ferret
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

> One problem with enforcing a single tag by mappers or preprocessing> data before putting it in the database is that if there are subtle distinctions they are> forever lost.Sven gave us a list of tags which have exact meaning. So no distinctions. No differences, just another names.e.g. `phone` is the same like `contact:phone`. You could just find one difference: The name. And to have just the names different is a bad thing. In such cases we can make life a lot easier by just using one tag AND NOT two tags for the same purpose!~ Sören Reinecke alias Valor Naram Original Message Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Paul Allen To: "Tag discussion, strategy and related tools" CC: On Sun, 6 Oct 2019 at 20:35, Sven Geggus <li...@fuchsschwanzdomain.de> wrote:
I fully agree with this.  In opencampingmap POI database I currently do a
replacement of the following tags during database import:

booking -> reservation
contact:phone -> phone
contact:fax -> fax
contact:website -> website
contact:email -> email 

Would be nice to get rid of stuff like this.Maybe, maybe not (I'm on the fence).  I doubt you'll manage to.  There are too manyon both sides who insist that their way is the one true way.  It's possible that yourkluge is actually the best way of handling things like this.  Data consumers can translatecontact:phone to phone or phone to contact:phone or leave both untouched as theychoose.  One problem with enforcing a single tag by mappers or preprocessingdata before putting it in the database is that if there are subtle distinctions they areforever lost.  You can always scramble an egg but you can't unscramble one.-- Paul
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

Are you a datauser Paul Allen? Because I hate doing replacements, they ugly and they are work I hate. Replacements make Queries more ugly and longer.Replacements also make the database less reliable, that's literally the opposite of what OSM mappers actually want.ANDEXAMPLE SCENARIO: I look for a way to list all pois with phones. I find the wikipage for the `contact:phone` key. I start to use it in my query. I do not get all data and I think OSM has not good coverage for phone numbers as I thought. In this situation OSM is less reliable to me.My example scenario shows a datauser (in this example myself) who missed out the wikipage "phone" and just took the first one "contact:phone". That means some data is hidden from me. And datausers don't have the idea to search for a second wikipage which describes another way of tagging phone numbers because it's logial to have one aggreement on storing data and not one. Two or more ways of storing literally the same thing is what I call inconsistent and bad for database usage.~ Sören Reinecke alias Valor Naram Original Message ----Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Paul Allen To: "Tag discussion, strategy and related tools" CC: On Sun, 6 Oct 2019 at 20:35, Sven Geggus <li...@fuchsschwanzdomain.de> wrote:
I fully agree with this.  In opencampingmap POI database I currently do a
replacement of the following tags during database import:

booking -> reservation
contact:phone -> phone
contact:fax -> fax
contact:website -> website
contact:email -> email 

Would be nice to get rid of stuff like this.Maybe, maybe not (I'm on the fence).  I doubt you'll manage to.  There are too manyon both sides who insist that their way is the one true way.  It's possible that yourkluge is actually the best way of handling things like this.  Data consumers can translatecontact:phone to phone or phone to contact:phone or leave both untouched as theychoose.  One problem with enforcing a single tag by mappers or preprocessingdata before putting it in the database is that if there are subtle distinctions they areforever lost.  You can always scramble an egg but you can't unscramble one.-- Paul
___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Paul Allen]

On Sun, 6 Oct 2019 at 20:35, Sven Geggus 
wrote:

>
> I fully agree with this.  In opencampingmap POI database I currently do a
> replacement of the following tags during database import:
>
> booking -> reservation
> contact:phone -> phone
> contact:fax -> fax
> contact:website -> website
> contact:email -> email
>
> Would be nice to get rid of stuff like this.
>

Maybe, maybe not (I'm on the fence).  I doubt you'll manage to.  There are
too many
on both sides who insist that their way is the one true way.  It's possible
that your
kluge is actually the best way of handling things like this.  Data
consumers can translate
contact:phone to phone or phone to contact:phone or leave both untouched as
they
choose.  One problem with enforcing a single tag by mappers or preprocessing
data before putting it in the database is that if there are subtle
distinctions they are
forever lost.  You can always scramble an egg but you can't unscramble one.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Sven Geggus]

Valor Naram  wrote:

> It's awful that we have two tags for the same puropose in our database and
> that makes it more difficult for developers and researchers to work with
> our data.

I fully agree with this.  In opencampingmap POI database I currently do a
replacement of the following tags during database import:

booking -> reservation
contact:phone -> phone
contact:fax -> fax
contact:website -> website
contact:email -> email
url -> website

Would be nice to get rid of stuff like this.

Sven

-- 
"In the land of the brave and the free, we defend our freedom
with the GNU GPL" (Richard M. Stallman on www.gnu.org)

/me is giggls@ircnet, http://sven.gegg.us/ on the Web

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

I moved the content of my proposal to
https://wiki.openstreetmap.org/wiki/Discussions/tagging/contact:phone_or_phone
because I don't work with categories and therefore I did not realize
the problem of two proposal in one page. As you can see I did not
stayed into the "Proposed_features" Schema because I do not propose a
feature. I propose the deprecation of a feature. This is a difference!

The content there tends to document the de-facto described at
https://wiki.openstreetmap.org/wiki/Key:phone but shorter and in my
opinion a bit more clear and faster to read except the last points I
simply copied from the source wiki page. I hope I documented the de-
faction well enough for you. If not, feel free to chance and improve it
but do not mark your changes as minor edit. I want to keep track on the
changes made to the page I created so I can use these changes to handle
the upcoming discussion here. As a reminder: The pre-discussion took
place in thread
https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html
and the outcome was to deprecate "contact:phone"

MY OWN OPINION: I personally think that "contact:phone" is
schematically better but I also agree that "phone" is better because
it's a shorter key, better to find in the wiki and used widely among
the community. So I propose deprecating "contact:phone" for the reason
why I think "phone" will win (or has already won according to its
sightly more usage).

Cheerio

Sören Reinecke alias Valor Naram

On Sat, 2019-09-28 at 10:31 +0200, Valor Naram wrote:
> Hey,
>
> now I'm ready to open a new proposal which you can see here
> https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29
>
> I use the old proposal page for that but seperated content into
> section
> to keep the history intact. The content is based on the discussion at
> https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html
>
> . It tends to deprecate `contact:phone` in favor of the more used de-
> facto `phone` tag. It's awful that we have two tags for the same
> puropose in our database and that makes it more difficult for
> developers and researchers to work with our data.
>
> Cheers
>
> Sören Reinecke alias Valor Naram
>
>
> ___
> 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] Feature Proposal - RFC - (Phone)

[Tobias Knerr]

On 28.09.19 10:31, Valor Naram wrote:
> now I'm ready to open a new proposal which you can see here
> https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29

I agree with the basic goal of ending the co-existence of phone and
contact:phone. I don't care that much about which one "wins", but having
multiple competing tagging styles around makes it just that little bit
harder for new mappers to learn the ropes (and for editor developers to
optimize usability). This particular co-existence has gone on for a
decade, and we should finally come to a decision.

That being said, I find the proposal a bit confusing as currently
written. It isn't clear to me if you are proposing any changes besides
deprecation of contact:phone, because there's a lot of text on that page
about country codes, the value syntax and such. Is this just
describing/affirming the status quo?

What adds to the confusion is that there are now multiple proposals on
the same wiki page, which is unexpected and messes with links and
categories. For example, that page is now in both the "rejected" and
"proposed" categories at the same time. Unless there's a strong reason
I'm missing, I suggest taking the conventional route of creating a new
page for your proposal.

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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Martin Koppenhoefer]

sent from a phone

> On 28. Sep 2019, at 11:48, Chris Hill  wrote:
> 
> Removing seemingly similar tags and so homogenising the OSM database is a 
> very risky path to take. We risk removing subtlety and obscuring mappers' 
> real intent. The world we live in and try to represent with map data is a 
> muddled, mixed-up, jumble of human-made stuff that includes many 
> contradictions and minutely different things. One great strength of OSM 
> tagging is that mappers can find ways to represent this. If we march down the 
> homogenisation highway much of that strength will be lost.


+1, I wholeheartedly agree with this



> 
> I oppose deprecating contact:phone=*



still, I support discouraging the use of contact:phone (as you write, the world 
is full of contradictions ;-) ). And of contact:website by the way.


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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Valor Naram]

You forget that `phone` is identical to `contact:phone` except the name. It's just not similar, it's the same by 99,99%.For example:Researchers will wonder why are there not so many data for tag `contact:phone`. Researchers might not know that there are two ways of tagging phone numbers. So they don't get the most data because they don't notice the `phone` key which is used more widely. Do you really want that? There is no logical reason to have two tags for the same purpose.Another thing: You speak about "similiar purpose" and I speak about "same purpose". This is a big difference.~ Sören Reinecke alias Valor Naram Original Message ----Subject: Re: [Tagging] Feature Proposal - RFC - (Phone)From: Chris Hill To: tagging@openstreetmap.orgCC: I disagree with this idea that we must remove similar tags for the sake of it.Anyone who actually uses OSM data (rather than people who just imagine using it) know that there are many steps and choices to make to achieve the end result. Often this involves combining data with various tags that fit the requirements of the analysis, render, routing or whatever, so combining data from similar tags is normal, not hard to do and once done is repeatable over and over. It is not awful to have two tags for a /similar/ purpose at all.Removing seemingly similar tags and so homogenising the OSM database is a very risky path to take. We risk removing subtlety and obscuring mappers' real intent. The world we live in and try to represent with map data is a muddled, mixed-up, jumble of human-made stuff that includes many contradictions and minutely different things. One great strength of OSM tagging is that mappers can find ways to represent this. If we march down the homogenisation highway much of that strength will be lost.I oppose deprecating contact:phone=*-- cheersChris Hill (chillly)On 28/09/2019 09:31, Valor Naram wrote:> Hey,>> now I'm ready to open a new proposal which you can see here> https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29> I use the old proposal page for that but seperated content into section> to keep the history intact. The content is based on the discussion at> https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html> . It tends to deprecate `contact:phone` in favor of the more used de-> facto `phone` tag. It's awful that we have two tags for the same> puropose in our database and that makes it more difficult for> developers and researchers to work with our data.>>___Tagging mailing listTagging@openstreetmap.orghttps://lists.openstreetmap.org/listinfo/tagging___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Mateusz Konieczny]

28 Sep 2019, 11:48 by o...@raggedred.net:

> I disagree with this idea that we must remove similar tags for the sake of it.
>
Similar - I agree.

Exact duplicates - removal is a great idea!
> Anyone who actually uses OSM data 
>
Note that it is also about mappers
that waste time on discovering that
different tags turn out to be exact duplicates.
> Removing seemingly similar tags
>
This argument is fine, but why it is
supposed to apply to this case?
This tags are exact duplicates from
what I know.
> I oppose deprecating contact:phone=*
>
I support this idea.___
Tagging mailing list
Tagging@openstreetmap.org
https://lists.openstreetmap.org/listinfo/tagging

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Chris Hill]

I disagree with this idea that we must remove similar tags for the sake 
of it.


Anyone who actually uses OSM data (rather than people who just imagine 
using it) know that there are many steps and choices to make to achieve 
the end result. Often this involves combining data with various tags 
that fit the requirements of the analysis, render, routing or whatever, 
so combining data from similar tags is normal, not hard to do and once 
done is repeatable over and over. It is not awful to have two tags for a 
/similar/ purpose at all.


Removing seemingly similar tags and so homogenising the OSM database is 
a very risky path to take. We risk removing subtlety and obscuring 
mappers' real intent. The world we live in and try to represent with map 
data is a muddled, mixed-up, jumble of human-made stuff that includes 
many contradictions and minutely different things. One great strength of 
OSM tagging is that mappers can find ways to represent this. If we march 
down the homogenisation highway much of that strength will be lost.


I oppose deprecating contact:phone=*

--
cheers
Chris Hill (chillly)

On 28/09/2019 09:31, Valor Naram wrote:

Hey,

now I'm ready to open a new proposal which you can see here
https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29
I use the old proposal page for that but seperated content into section
to keep the history intact. The content is based on the discussion at
https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html
. It tends to deprecate `contact:phone` in favor of the more used de-
facto `phone` tag. It's awful that we have two tags for the same
puropose in our database and that makes it more difficult for
developers and researchers to work with our data.




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

Re: [Tagging] Feature Proposal - RFC - (Phone)

[Andrew Hain]

I oppose selectively deprecating contact:phone because without explaining whhy 
it diifers from other contact tags (populist appeals to tag counts don’t count) 
it makes tagging less orthogonal.

--
Andrew

From: Valor Naram 
Sent: 28 September 2019 09:31
To: Tagging@openstreetmap.org 
Subject: [Tagging] Feature Proposal - RFC - (Phone)

Hey,

now I'm ready to open a new proposal which you can see here
https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29
I use the old proposal page for that but seperated content into section
to keep the history intact. The content is based on the discussion at
https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html
. It tends to deprecate `contact:phone` in favor of the more used de-
facto `phone` tag. It's awful that we have two tags for the same
puropose in our database and that makes it more difficult for
developers and researchers to work with our data.

Cheers

Sören Reinecke alias Valor Naram


___
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] Feature Proposal - RFC - (Phone)

[Valor Naram]

Hey,

now I'm ready to open a new proposal which you can see here
https://wiki.openstreetmap.org/wiki/Proposed_features/Phone#Second_proposal_.28pending.29
I use the old proposal page for that but seperated content into section
to keep the history intact. The content is based on the discussion at
https://lists.openstreetmap.org/pipermail/tagging/2019-September/048339.html
. It tends to deprecate `contact:phone` in favor of the more used de-
facto `phone` tag. It's awful that we have two tags for the same
puropose in our database and that makes it more difficult for
developers and researchers to work with our data.

Cheers

Sören Reinecke alias Valor Naram


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