Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Mike N

On 9/7/2016 2:38 PM, Brian Stromberg wrote:

My point was only that applying an automated process nationwide without
any consideration for on-the-ground verification seems likely to make a
less accurate map.


 For the ITT case, there is presumably a reference on the web site that 
lists all schools.  These can be cross-checked against the entity 
address or the nearby street if there was no address.


 There are no formal rules about which types of edits can be made 
without on-the-ground verification.  At best, we want a chance to review 
all automated edits and make the call on a case by case basis.



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


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Brian Stromberg
Thank you! I am trying to get more engaged with some aspects of OSM and
this list has been very interesting to read.

I agree with Elliott that it is wrong-headed to leave mislabeled POIs in
place in order to encourage participation. My point was only that applying
an automated process nationwide without any consideration for on-the-ground
verification seems likely to make a less accurate map. I'm fairly new to
the scene, so maybe there is a policy in place that deals with automated
batch relabeling/deleting that I'm not aware of? If not, then this might be
a good opportunity to start that discussion.

--
Brian

On Wed, Sep 7, 2016 at 2:19 PM, Mike N  wrote:

> On 9/7/2016 10:13 AM, Brian Stromberg wrote:
>
>> Automated nationwide mapping seems like it would introduce more problems
>> than it would solve. If maps are intended to represent the truth on the
>> ground, then the only way to create a useful map is by reporting what is
>> actually there rather than making assumptions. A map that is inaccurate
>> because it is outdated is better than a map that is inaccurate because
>> of a flawed process.
>>
>
>   Welcome to the list!   I've been surprised by both what OSM data is
> applied to, as well as what it is not.  Where it was used, having the
> maximum amount of current information made it the most useful.
>
> You are correct that using a flawed process will lead to inaccurate
> results - there was the famous worldwide edit "Fix and unify all
> MacDonald's restaurant names" which led to wrong results.   The ITT closing
> case is both small enough and well defined that I expect a very high
> quality result.
>
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Mike N

On 9/7/2016 10:13 AM, Brian Stromberg wrote:

Automated nationwide mapping seems like it would introduce more problems
than it would solve. If maps are intended to represent the truth on the
ground, then the only way to create a useful map is by reporting what is
actually there rather than making assumptions. A map that is inaccurate
because it is outdated is better than a map that is inaccurate because
of a flawed process.


  Welcome to the list!   I've been surprised by both what OSM data is 
applied to, as well as what it is not.  Where it was used, having the 
maximum amount of current information made it the most useful.


You are correct that using a flawed process will lead to inaccurate 
results - there was the famous worldwide edit "Fix and unify all 
MacDonald's restaurant names" which led to wrong results.   The ITT 
closing case is both small enough and well defined that I expect a very 
high quality result.



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


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Elliott Plack
I ran some queries on Overpass Turbo. There are only 15 POIs that match
some variant of ITT Tech in the USA. Here's the full list:
https://gist.github.com/anonymous/9a8770998bacbb7bc7ba81f93eca618d . Here
is the query: http://overpass-turbo.eu/s/igB

While it is important to be critical of automated edits, I don't agree that
the community should prefer an outdated map to inspire new mappers.
Following that logic, I should just quit OSM altogether so that the local
map will go stale and others will pick it up, right? No, of course not. We
should be editing what we know or can reasonably assume is there based on
information, and then others can fill in the gaps as they find them.


On Wed, Sep 7, 2016 at 10:41 AM Brian Stromberg 
wrote:

> Automated nationwide mapping seems like it would introduce more problems
> than it would solve. If maps are intended to represent the truth on the
> ground, then the only way to create a useful map is by reporting what is
> actually there rather than making assumptions. A map that is inaccurate
> because it is outdated is better than a map that is inaccurate because of a
> flawed process.
>
> Also, first time participating in these conversations, so I am pressing
> "send" with great trepidation...
>
> --
> Brian
>
> On Wed, Sep 7, 2016 at 10:07 AM, Greg Morgan 
> wrote:
>
>>
>>
>> On Tue, Sep 6, 2016 at 2:36 PM, Frederik Ramm 
>> wrote:
>>
>>> Hi,
>>>
>>> On 09/06/2016 11:01 PM, Elliott Plack wrote:
>>> > Should we launch an automated edit, or some kind of batch process on
>>> OSM
>>> > to clear the database `name=ITT Tech` (or similar) worldwide?
>>>
>>
>> For one I had to go looking for the story.
>> http://www.fox10phoenix.com/news/us-world-news/202562451-story
>> The for-profit college chain ITT Technical Institute is shutting down all
>> 130 of its U.S. campuses, saying Tuesday it can't survive recent sanctions
>> by the U.S. Department of Education...
>>
>>
>>>
>>> This is a discussion that has happened in the past when Domino's Pizza
>>> has rebranded, or when the "Schlecker" drug store chain closed in
>>> Germany.
>>>
>>> I think automated edits are not a good solution mainly for two reasons:
>>>
>>> 1. In many cases, the world doesn't change instantly at the behest of
>>> some guy in marketing or legal. Individual locations might retain their
>>> signage for various reasons and we map what's on the ground,
>>>
>>
>> Not around here.  They pop out the plastic and replace it with the new
>> company's name.  If it takes awhile for a new company to replace the old
>> company, then they flip the plastic over to save costs.  This isn't like
>> the days when bespoke signs were created for every business.
>> http://mapillary.com/map/im/LCodN7YJMEPKKlRnf2eLxw
>>
>>
>>>
>>> 2. If a chain is renamed or closed country-wide, and this change is not
>>> reflected on OSM in one area, then this can be a valuable sign for lack
>>> of mapper attention. A sign that has the best user interface of all:
>>> Because for any map user, dealing with an outdated map is normal, and
>>> the way you identify just *how* outdated something is is exactly by
>>> looking at such things: "Ah, this map seems to be from a time then
>>> Domino's was still called Domino's Pizza!" - Leaving these valuable
>>> markers of outdated-ness in place tells the map user that this area
>>> hasn't been touched for a while and that the other POIs in the vicinity
>>> are likely also a bit aged. When a local mapper touches up the area they
>>> will likely also update other things than just the closed-down shop, and
>>> then the map will be current again. Automatically editing away something
>>> country-wide hides the fact that the map lacks attention in an area.
>>
>>
>> Frederik you are thinking about this from a dense mapper perspective.
>> Germany has 89 million people in the same are as Montana that only has one
>> million people.  You are talking about one way to map if you have that
>> kind-of population that you can create a large mapper population from.  I
>> already know much of my area needs updating.  The problem is that I cannot
>> get to it and the transient nature of the area doesn't mean that I can go
>> out and build an OSM community here.  Automated edits like this by another
>> mapper would be a great addition to the work I already do in an area.  As
>> another perspective, I typically don't put in POIs like this unless I visit
>> them.  My area was one of the ground zeros for the sub-prime rate debacle.
>> I'll put in an address on a building before I will put in a POI.  Business
>> just don't stay around like they used to. I laugh snort when I read that
>> wiki page about armchair mapping.  I am in the top 50 worldwide Mapillary
>> submissions.  You think I can go out there and survey every node before I
>> put it in OSM. That's jsut not going to happen. You think that I even know
>> about changes like the ITT story.
>> https://www.mapillary.com/app/?lat=33.263809986326706&lng=-111.81125590160123&z=

[Talk-us] SoTM - Local Chapters Congress

2016-09-07 Thread Clifford Snow
I'm forwarding a message from Joost Schouppe inviting local organizers to
attend and be a part of the Local Chapters Congress at SoTM Brussels.

-- Forwarded message --
From: joost schouppe 
Date: 2016-09-06 9:48 GMT+02:00
Subject: State of the Local Map at SotM 2016
To:


Hi,

A few of you are invited (or will be) to the Local Chapters Congress.
Others have sent a proposal for a State of the Map (SotM) talk which had to
do with local community building. Some of you got on the program, others'
didn't. I hope all of you can make it to SotM!

The team at SotM do consider local community building extremely important,
but with only so many time slots, choices had to be made. As a kind of
compensation for all the times we had to say "no" to a local community
talk, there will be a session with (potential) local chapters [1] and a
panel discussion on building local communities [2]. I'm writing you about
the panel discussion.

We're not quite sure who exactly should be on the panel, but it could be
you. The first thing we want to do, is to define what we will talk about.
We'd like some input from you about that. If you know someone else from
your local community who would be equally or better placed to answer these
questions, feel free to forward this mail. If you feel like there are more
pertinent questions to answer, please feel free to add those here.

A) What do you consider the most important lesson learnt in your own
efforts to build a local mapping community? What worked for you or did not
work when trying to expand your community.
B) Are there dilemma's or challenges you face about which you would like to
get other people's opinions?
C) How could the global OSM community help you? For example in helping with
the challenges you face. But also making sure the lessons you learned don't
have to be re-learned over and over again. Or in scaling up the solutions
you found (e.g. make a global tool for something you use locally).

Some questions about you:
- what country/region could you represent?
- will you be at SotM?
- would you like to be on the panel?

Thanks for any help in getting this session together!

[1] http://2016.stateofthemap.org/2016/local-chapters-congress/
[2]
http://2016.stateofthemap.org/2016/openstreetmap-awards-and-state-of-the-local-map/


-- 
@osm_seattle
osm_seattle.snowandsnow.us
OpenStreetMap: Maps with a human touch
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Brian Stromberg
Automated nationwide mapping seems like it would introduce more problems
than it would solve. If maps are intended to represent the truth on the
ground, then the only way to create a useful map is by reporting what is
actually there rather than making assumptions. A map that is inaccurate
because it is outdated is better than a map that is inaccurate because of a
flawed process.

Also, first time participating in these conversations, so I am pressing
"send" with great trepidation...

--
Brian

On Wed, Sep 7, 2016 at 10:07 AM, Greg Morgan  wrote:

>
>
> On Tue, Sep 6, 2016 at 2:36 PM, Frederik Ramm  wrote:
>
>> Hi,
>>
>> On 09/06/2016 11:01 PM, Elliott Plack wrote:
>> > Should we launch an automated edit, or some kind of batch process on OSM
>> > to clear the database `name=ITT Tech` (or similar) worldwide?
>>
>
> For one I had to go looking for the story. http://www.
> fox10phoenix.com/news/us-world-news/202562451-story
> The for-profit college chain ITT Technical Institute is shutting down all
> 130 of its U.S. campuses, saying Tuesday it can't survive recent sanctions
> by the U.S. Department of Education...
>
>
>>
>> This is a discussion that has happened in the past when Domino's Pizza
>> has rebranded, or when the "Schlecker" drug store chain closed in Germany.
>>
>> I think automated edits are not a good solution mainly for two reasons:
>>
>> 1. In many cases, the world doesn't change instantly at the behest of
>> some guy in marketing or legal. Individual locations might retain their
>> signage for various reasons and we map what's on the ground,
>>
>
> Not around here.  They pop out the plastic and replace it with the new
> company's name.  If it takes awhile for a new company to replace the old
> company, then they flip the plastic over to save costs.  This isn't like
> the days when bespoke signs were created for every business.
> http://mapillary.com/map/im/LCodN7YJMEPKKlRnf2eLxw
>
>
>>
>> 2. If a chain is renamed or closed country-wide, and this change is not
>> reflected on OSM in one area, then this can be a valuable sign for lack
>> of mapper attention. A sign that has the best user interface of all:
>> Because for any map user, dealing with an outdated map is normal, and
>> the way you identify just *how* outdated something is is exactly by
>> looking at such things: "Ah, this map seems to be from a time then
>> Domino's was still called Domino's Pizza!" - Leaving these valuable
>> markers of outdated-ness in place tells the map user that this area
>> hasn't been touched for a while and that the other POIs in the vicinity
>> are likely also a bit aged. When a local mapper touches up the area they
>> will likely also update other things than just the closed-down shop, and
>> then the map will be current again. Automatically editing away something
>> country-wide hides the fact that the map lacks attention in an area.
>
>
> Frederik you are thinking about this from a dense mapper perspective.
> Germany has 89 million people in the same are as Montana that only has one
> million people.  You are talking about one way to map if you have that
> kind-of population that you can create a large mapper population from.  I
> already know much of my area needs updating.  The problem is that I cannot
> get to it and the transient nature of the area doesn't mean that I can go
> out and build an OSM community here.  Automated edits like this by another
> mapper would be a great addition to the work I already do in an area.  As
> another perspective, I typically don't put in POIs like this unless I visit
> them.  My area was one of the ground zeros for the sub-prime rate debacle.
> I'll put in an address on a building before I will put in a POI.  Business
> just don't stay around like they used to. I laugh snort when I read that
> wiki page about armchair mapping.  I am in the top 50 worldwide Mapillary
> submissions.  You think I can go out there and survey every node before I
> put it in OSM. That's jsut not going to happen. You think that I even know
> about changes like the ITT story. https://www.mapillary.com/app/
> ?lat=33.263809986326706&lng=-111.81125590160123&z=8.272711628984965
>  Woot! Woot!
>
> Elliot, I'd say go for the change.  You are only going to __potentially__
> remove 130 names from OSM objects.  Better yet, rename the name to a note
> tag, with an explanation that the place was shutdown because of sanctions.
>
> Regards,
> Greg
>
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> https://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Greg Morgan
On Tue, Sep 6, 2016 at 2:36 PM, Frederik Ramm  wrote:

> Hi,
>
> On 09/06/2016 11:01 PM, Elliott Plack wrote:
> > Should we launch an automated edit, or some kind of batch process on OSM
> > to clear the database `name=ITT Tech` (or similar) worldwide?
>

For one I had to go looking for the story.
http://www.fox10phoenix.com/news/us-world-news/202562451-story
The for-profit college chain ITT Technical Institute is shutting down all
130 of its U.S. campuses, saying Tuesday it can't survive recent sanctions
by the U.S. Department of Education...


>
> This is a discussion that has happened in the past when Domino's Pizza
> has rebranded, or when the "Schlecker" drug store chain closed in Germany.
>
> I think automated edits are not a good solution mainly for two reasons:
>
> 1. In many cases, the world doesn't change instantly at the behest of
> some guy in marketing or legal. Individual locations might retain their
> signage for various reasons and we map what's on the ground,
>

Not around here.  They pop out the plastic and replace it with the new
company's name.  If it takes awhile for a new company to replace the old
company, then they flip the plastic over to save costs.  This isn't like
the days when bespoke signs were created for every business.
http://mapillary.com/map/im/LCodN7YJMEPKKlRnf2eLxw


>
> 2. If a chain is renamed or closed country-wide, and this change is not
> reflected on OSM in one area, then this can be a valuable sign for lack
> of mapper attention. A sign that has the best user interface of all:
> Because for any map user, dealing with an outdated map is normal, and
> the way you identify just *how* outdated something is is exactly by
> looking at such things: "Ah, this map seems to be from a time then
> Domino's was still called Domino's Pizza!" - Leaving these valuable
> markers of outdated-ness in place tells the map user that this area
> hasn't been touched for a while and that the other POIs in the vicinity
> are likely also a bit aged. When a local mapper touches up the area they
> will likely also update other things than just the closed-down shop, and
> then the map will be current again. Automatically editing away something
> country-wide hides the fact that the map lacks attention in an area.


Frederik you are thinking about this from a dense mapper perspective.
Germany has 89 million people in the same are as Montana that only has one
million people.  You are talking about one way to map if you have that
kind-of population that you can create a large mapper population from.  I
already know much of my area needs updating.  The problem is that I cannot
get to it and the transient nature of the area doesn't mean that I can go
out and build an OSM community here.  Automated edits like this by another
mapper would be a great addition to the work I already do in an area.  As
another perspective, I typically don't put in POIs like this unless I visit
them.  My area was one of the ground zeros for the sub-prime rate debacle.
I'll put in an address on a building before I will put in a POI.  Business
just don't stay around like they used to. I laugh snort when I read that
wiki page about armchair mapping.  I am in the top 50 worldwide Mapillary
submissions.  You think I can go out there and survey every node before I
put it in OSM. That's jsut not going to happen. You think that I even know
about changes like the ITT story.
https://www.mapillary.com/app/?lat=33.263809986326706&lng=-111.81125590160123&z=8.272711628984965
 Woot! Woot!

Elliot, I'd say go for the change.  You are only going to __potentially__
remove 130 names from OSM objects.  Better yet, rename the name to a note
tag, with an explanation that the place was shutdown because of sanctions.

Regards,
Greg
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Frederik Ramm
Hi,

On 09/07/2016 03:25 PM, Greg Troxel wrote:
> Your comment comes across as bizarre and hostile to the US mapping 
> community, as if you think it's horribly broken and proving that
> point is more important than improving the map.

I think there's a misunderstanding here.

I had the same comment when someone in Germany wanted to automatically
remove all "Schlecker" drug stores.

It is neither bizarre nor hostile. Any map in any country has its weak
spots. If a chain restaurant in the US - or in Germany - changes its
name overnight, some areas will be fixed quickly because they have
local mappers who care, and other areas will take half a year or
longer because they lack local mappers who care.

This is an undeniable fact, and it doesn't have anything to do with
whether this is in the US or in Germany or elsewhere on the planet.

To me, *not* running an automated edit means honestly communicating to
the map user where the weaker areas are (and that the map is likely
more reliable in one part of the country than in another). Running an
automated edit hides the weakness but doesn't fix it.

There's no shame in admitting to a weakness in the map; on the
contrary, admitting it is more likely to attract people who will help
fixing it.

And no, I don't think the mapping community in the US is horribly
broken - it is just developing slower than I had hoped, and I see many
attempts to make up for this slow development in ways that ultimately
slow things down even further instead of helping. But this has
absolutely nothing to do with the topic at hand; as I said, I've made
the very same recommendation for the very same reasons in other countrie
s.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

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


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Paul Johnson
On Wed, Sep 7, 2016 at 8:25 AM, Greg Troxel  wrote:
>
> Your comment comes across as bizarre and hostile to the US mapping
> community, as if you think it's horribly broken and proving that point
> is more important than improving the map.


Meh.  Gives you a good excuse to go survey.  Go hit all the locations in
your area for whichever chain and verify it on the ground, edit your
findings as appropriate.  Odds are you'll find other things you can fix or
add in the process.
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Greg Troxel

Frederik Ramm  writes:

> Hi,
>
> On 09/07/2016 02:48 AM, Mike N wrote:
>> But if one less 
>> thing is wrong or outdated, that makes the data more useful to all clients.
>
> Except those humans who could have used that outdated thing as a marker
> to tell them that the map is dated.
>
> Yes they could look at the last modification date of things or analyze
> how many contributors the area has or myriad other OSM insider things.
> But seeing a "Domino's Pizza" on the map doesn't need an API, or insider
> knowledge, it doesn't even need a web site - it is the universal
> language of map dating.
>
> Automatically fixing that is like a car salesperson fixing a leak with
> bubble gum because it looks better and they can't be bothered to fix it
> properly.

Your comment comes across as bizarre and hostile to the US mapping
community, as if you think it's horribly broken and proving that point
is more important than improving the map.


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


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Mike N

On 9/7/2016 3:45 AM, Frederik Ramm wrote:

Except those humans who could have used that outdated thing as a marker
to tell them that the map is dated.

Yes they could look at the last modification date of things or analyze
how many contributors the area has or myriad other OSM insider things.
But seeing a "Domino's Pizza" on the map doesn't need an API, or insider
knowledge, it doesn't even need a web site - it is the universal
language of map dating.



  I can't visualize a mapper that is so starved for places to resurvey 
that they rely on a list of stale POIs as a source instead of a more 
methodical system of date checking or any of the many OSM QA tools that 
automatically flag stale data such as a dead web site.



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


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Rory McCann
On 06/09/16 23:01, Elliott Plack wrote:
> Should we launch an automated edit, or some kind of batch process on OSM
> to clear the database `name=ITT Tech` (or similar) worldwide?

Be careful about the "worldwide" part. There are many country specific
brands, and it could be possible that the name is used elsewhere in the
world for some other group. You don't want to accidentally "close" lots
of shops/companies in another country which has nothing to do with the
company you're thinking of.

In general, if you were to do it, restrict it to a while list of
countries that you know that brand operates in.

(A quick overpass query tells me there's only one name="ITT Tech" in
OSM, so in this particular case it doesn't matter too much 🙂 )


-- 




signature.asc
Description: OpenPGP digital signature
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us


Re: [Talk-us] Deleting / Closing / Renaming all places in a chain

2016-09-07 Thread Frederik Ramm
Hi,

On 09/07/2016 02:48 AM, Mike N wrote:
> But if one less 
> thing is wrong or outdated, that makes the data more useful to all clients.

Except those humans who could have used that outdated thing as a marker
to tell them that the map is dated.

Yes they could look at the last modification date of things or analyze
how many contributors the area has or myriad other OSM insider things.
But seeing a "Domino's Pizza" on the map doesn't need an API, or insider
knowledge, it doesn't even need a web site - it is the universal
language of map dating.

Automatically fixing that is like a car salesperson fixing a leak with
bubble gum because it looks better and they can't be bothered to fix it
properly.

Bye
Frederik

-- 
Frederik Ramm  ##  eMail frede...@remote.org  ##  N49°00'09" E008°23'33"

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