Chain Store Cleanup

classic Classic list List threaded Threaded
36 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Re: Chain Store Cleanup

Andrew MacKinnon-2
> I wouldn't be so sure here.
>
> As an example, there's a bakery chain in the UK called "Greggs". They're
> mostly tagged "shop=bakery" (with a few Subway-esque "amenity=fast_food" /
> "cuisine=sandwich" as well).  Occasionally shops like this get wrongly
> tagged, sometimes as "amenity=cafe", and there's always a temptation to
> "just fix them".  However, guess what?  Yesterday I accidentally walked past
> a genuine Greggs "amenity=cafe":

There are too many of these errors (things like McDonalds without an
apostrophe and amenity=fast_food incorrectly tagged
amenity=restaurant) to check them all individually. I am only
interested in fixing the big chains here and will leave things I am
not familiar with alone. I live in Canada, and we have all the big
worldwide chains (McDonald's, Subway, KFC, Walmart) and some big local
chains (Tim Hortons, Loblaws, Shoppers Drug Mart, Sobeys, Rexall etc.)
With many of the chains most of the stores look pretty similar from an
air photo and it is easy to identify them that way. Also the
increasing popularity of Mapillary makes it easier to check these in
areas where Mapillary is available.

If you find a weird situation like this I would strongly recommend
putting a note tag on it. If I don't mistakenly correct the POI then
someone else might do it. Usually when I find this sort of situation
it is in a different country from the country where the chain has its
stores, and some independent store has the same name as the chain.
Occasionally there might be some store called Subway that doesn't sell
sandwiches that is legal because the Subway trademark only covers fast
food restaurants, or something like that. Any Subway or McDonald's
store that isn't owned by the big chains in one of the countries where
those chains operate would almost certainly be sued for trademark
infringement and shut down.

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

Re: Chain Store Cleanup

Colin Smale

A bit of a meta-discussion.... I wonder why this topic is not going the same way as the debate on talk-gb last November-December in which it was proposed to tidy up and normalise various spelling variants? There was a lot of vehement opposition to any automated "corrections" as many chains are inconsistent with their own orthography and only on-the-ground mappers would be able to tell whether or not there is an apostrophe present in the signage at this particular branch (etc. etc., you get this idea).

Discussion starts here:

https://lists.openstreetmap.org/pipermail/talk-gb/2014-November/016718.html

//colin

 

On 2015-05-02 18:14, Andrew MacKinnon wrote:

I wouldn't be so sure here. As an example, there's a bakery chain in the UK called "Greggs". They're mostly tagged "shop=bakery" (with a few Subway-esque "amenity=fast_food" / "cuisine=sandwich" as well). Occasionally shops like this get wrongly tagged, sometimes as "amenity=cafe", and there's always a temptation to "just fix them". However, guess what? Yesterday I accidentally walked past a genuine Greggs "amenity=cafe":
There are too many of these errors (things like McDonalds without an
apostrophe and amenity=fast_food incorrectly tagged
amenity=restaurant) to check them all individually. I am only
interested in fixing the big chains here and will leave things I am
not familiar with alone. I live in Canada, and we have all the big
worldwide chains (McDonald's, Subway, KFC, Walmart) and some big local
chains (Tim Hortons, Loblaws, Shoppers Drug Mart, Sobeys, Rexall etc.)
With many of the chains most of the stores look pretty similar from an
air photo and it is easy to identify them that way. Also the
increasing popularity of Mapillary makes it easier to check these in
areas where Mapillary is available.

If you find a weird situation like this I would strongly recommend
putting a note tag on it. If I don't mistakenly correct the POI then
someone else might do it. Usually when I find this sort of situation
it is in a different country from the country where the chain has its
stores, and some independent store has the same name as the chain.
Occasionally there might be some store called Subway that doesn't sell
sandwiches that is legal because the Subway trademark only covers fast
food restaurants, or something like that. Any Subway or McDonald's
store that isn't owned by the big chains in one of the countries where
those chains operate would almost certainly be sued for trademark
infringement and shut down.

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

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

Re: Chain Store Cleanup

Andy Mabbett
In reply to this post by Andrew MacKinnon-2
On 2 May 2015 at 02:18, Andrew MacKinnon <[hidden email]> wrote:

> If a store changes name a mechanical edit does not make
> sense because usually new signs get put up gradually. For instance
> Domino's Pizza changed its name to Domino's and is running TV ads
> promoting this, but there are still old signs that say "Domino's
> Pizza"

I suppose it depends whether we want to map what (sometimes incorrect)
store signs say, or what the stores actually are.

I favour the latter, but if you want the former, I have a list defunct
shops whose signs are still visible, which you can add

Another issue to consider is that either method will incude some
errors. Which will include fewest, and which will inconvenience our
users less? How long will it take for all our entries for Domino's to
be manually updated, even after the signs are changed?

--
Andy Mabbett
@pigsonthewing
http://pigsonthewing.org.uk

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

Re: Chain Store Cleanup

Andrew MacKinnon-2
In reply to this post by Colin Smale
On Sat, May 2, 2015 at 1:01 PM, Colin Smale <[hidden email]> wrote:
> A bit of a meta-discussion.... I wonder why this topic is not going the same
> way as the debate on talk-gb last November-December in which it was proposed
> to tidy up and normalise various spelling variants? There was a lot of
> vehement opposition to any automated "corrections" as many chains are
> inconsistent with their own orthography and only on-the-ground mappers would
> be able to tell whether or not there is an apostrophe present in the signage
> at this particular branch (etc. etc., you get this idea).

There are some chains that are very consistent and there are others
that are inconsistent and haven't bothered to change old signage. At
least in Canada I am aware of which ones these are. I am 99% confident
that Tim Hortons never has an apostrophe and McDonald's always has an
apostrophe, the dozens/hundreds of stores in the Greater Toronto Area
are all consistent. Also amenity=restaurant with Subway, McDonald's,
KFC, etc. is wrong because amenity=restaurant means a sit down
restaurant where you pay after eating. The pages on the wiki I am
working on creating are useful for telling which chain stores are
always consistent and which have several different variations.

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

Re: Chain Store Cleanup

Andrew MacKinnon-2
I found a few McDonald's that are not what you think they are. In OSM
there is a McDonald's confectionery shop in Mexico, a McDonald's
supermarket in Missouri, and a McDonald's kitchen store in New
Hampshire that are not fast food restaurants. Looks like you need to
be careful when fixing these.

This is due to weirdness in trademark laws where sometimes companies
in different categories are allowed to use the same trademark - e.g.
Apple Inc. and Apple Records.

I am pretty confident that any fast food/restaurant that is called
McDonald's is what you think it is though. Most of the time (at least
with suburban locations) you can tell that it is a McDonald's from the
aerial imagery.

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

Re: Chain Store Cleanup

Frederik Ramm
Hi,

On 05/02/2015 10:40 PM, Andrew MacKinnon wrote:
> This is due to weirdness in trademark laws where sometimes companies
> in different categories are allowed to use the same trademark - e.g.
> Apple Inc. and Apple Records.

Also, by far not every chain will have a global trademark (or the clout
to actually police it). If there's a "Starbuck's" cafe in Burma, maybe
it is indeed a blatant rip-off run by a cousin of the prime minister,
and maybe it is really called "Starbuck's".

I'm still convinced that fixing these low hangig fruit is an excellent
task for a beginning mapper in an area (you may not dare to draw a
landuse but you will be able to summon the courage of fixing an obvious
spelling mistake), and in turn it will say something about how well kept
an area is in OSM.

You will never be able to make the map good in an area from your desk
thousands of miles away. Doing a chain store "cleanup" from that
distance is just window dressing - the quality of the map will not be
different, at best you'll make it appear different.

Bye
Frederik

--
Frederik Ramm  ##  eMail [hidden email]  ##  N49°00'09" E008°23'33"

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

Re: Chain Store Cleanup

Frederik Ramm
In reply to this post by Andy Mabbett
Hi,

On 05/02/2015 07:58 PM, Andy Mabbett wrote:
> I suppose it depends whether we want to map what (sometimes incorrect)
> store signs say, or what the stores actually are.

Definitely we want to map what's on the ground. We map mis-spellings in
road signs, too.

> I favour the latter, but if you want the former, I have a list defunct
> shops whose signs are still visible, which you can add

I don't know about the general rules about tagging defunct shops but if
there is a tag for a defunct shop then the name tag should certainly
bear the name on the still visible sign; if only as a landmark.

> Another issue to consider is that either method will incude some
> errors. Which will include fewest, and which will inconvenience our
> users less? How long will it take for all our entries for Domino's to
> be manually updated, even after the signs are changed?

We collect observations. The fact that there is a relationship between
all pizza places with a red-and-blue Domino's sign and what kind of
relationship there is - are they all owned by the same company, are they
a franchise, and if they are a franchise, what contractual freedoms in
naming/signage are given to the franchisee in the contract - is not
easily observable and therefore should not inform OSM mapping. There is
no way for the mapper on the ground to know that the name on the
building "should" be something else.

I know you have a Wikidata background and things may be different in
Wikidata; in Wikidata, you might be interested to research and record
the details of the contractual relationship between a Domino's store and
whatever the Domino HQ is, but in OSM this is very low on our list, if
it is on our list at all. We see a place that sells pizzas and the place
has a sign with their name on it, and that's what we map.

Bye
Frederik

--
Frederik Ramm  ##  eMail [hidden email]  ##  N49°00'09" E008°23'33"

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

Data Quality - was Re: Chain Store Cleanup

Colin Smale
On 2015-05-02 23:28, Frederik Ramm wrote:

> We collect observations.

...

> There is
> no way for the mapper on the ground to know that the name on the
> building "should" be something else.

I think that sounds rather disingenuous. We humans are perfectly capable
of correctly interpreting data which contains errors, and recognising
what the error is. And there are plenty of types of information in OSM
which are not (easily) verifiable on the ground - admin boundaries
spring to mind. The important thing in my mind is that the information
should be independently verifiable from publicly accessible (and
appropriately licensed) sources, thus making the information objective.
Of course the signs on the ground come into that category, but they are
not necessarily superior to other valid sources.

There are plenty of spelling and grammatical mistakes on public signs,
and although we are not the world's signage police, we should not be in
the business of propagating obvious errors either.

You mentioned "quality" in another post; that implies "the extent of
adherence to agreed criteria" it's a problem that we cannot yet measure
the quality of our data because there is no consensus on what is "good"
and what is not. That's why these discussions go round and round and
round for a couple of weeks and then die off. There seems to be little
motivation or drive to reach a clear conclusion. We don't even manage to
work out *how* to determine what is "good". It's time we grew the balls
we need to have the very painful talk about good data vs. bad data,
followed by finding the right balance between quality and quantity.
Quality itself can be subjective. What's fit for my purpose may break
the data's usability for yours. And yet there is only one OSM data set.
What are we going to agree to put in there, to keep the majority of
people "happy"? What is our shared definition of quality?

//colin

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

Re: Chain Store Cleanup

Andy Mabbett
In reply to this post by Frederik Ramm
On 2 May 2015 at 22:28, Frederik Ramm <[hidden email]> wrote:

> I know you have a Wikidata background and things may be different in
> Wikidata

I've been editing OSM longer than Wikidata has existed.

Even had I not, I don't think your attempt to analyse my "background"
has any place on this list.

--
Andy Mabbett
@pigsonthewing
http://pigsonthewing.org.uk

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

Re: Data Quality - was Re: Chain Store Cleanup

Pmailkeey .
In reply to this post by Colin Smale


On 2 May 2015 at 23:05, Colin Smale <[hidden email]> wrote:
On 2015-05-02 23:28, Frederik Ramm wrote:

We collect observations.

...

There is
no way for the mapper on the ground to know that the name on the
building "should" be something else.

I think that sounds rather disingenuous. We humans are perfectly capable of correctly interpreting data which contains errors, and recognising what the error is. And there are plenty of types of information in OSM which are not (easily) verifiable on the ground - admin boundaries spring to mind. The important thing in my mind is that the information should be independently verifiable from publicly accessible (and appropriately licensed) sources, thus making the information objective. Of course the signs on the ground come into that category, but they are not necessarily superior to other valid sources.

There are plenty of spelling and grammatical mistakes on public signs, and although we are not the world's signage police, we should not be in the business of propagating obvious errors either.

You mentioned "quality" in another post; that implies "the extent of adherence to agreed criteria" it's a problem that we cannot yet measure the quality of our data because there is no consensus on what is "good" and what is not. That's why these discussions go round and round and round for a couple of weeks and then die off. There seems to be little motivation or drive to reach a clear conclusion. We don't even manage to work out *how* to determine what is "good". It's time we grew the balls we need to have the very painful talk about good data vs. bad data, followed by finding the right balance between quality and quantity. Quality itself can be subjective. What's fit for my purpose may break the data's usability for yours. And yet there is only one OSM data set. What are we going to agree to put in there, to keep the majority of people "happy"? What is our shared definition of quality?

//colin


HERE HERE.

Having said that, I fear the grey area is almost as large as a popular blue-green planet. I think whatever is decided as being the correct way can only end up being a guide. There's always the possibility the correct way will change in time as we learn. The correct way needs to be easily accessible too - to all - especially newbies to OSM. It seems iD is the preferred newbie editor so that needs to be designed to guide all in using correct ways.

When we come across incorrect signs, how big a deal is it to point them out and at least start the ball rolling to get them corrected ? As a keen data observer, I'm doing it - even correcting Ordnance Survey.

--
Mike.
@millomweb - For all your info on Millom and South Copeland
via the area's premier website - 

currently unavailable due to ongoing harassment of me, my family, property & pets


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

Re: Chain Store Cleanup

Pmailkeey .
In reply to this post by Andy Mabbett


On 2 May 2015 at 23:18, Andy Mabbett <[hidden email]> wrote:
On 2 May 2015 at 22:28, Frederik Ramm <[hidden email]> wrote:

> I know you have a Wikidata background and things may be different in
> Wikidata

I've been editing OSM longer than Wikidata has existed.

Even had I not, I don't think your attempt to analyse my "background"
has any place on this list.



Frederik, Andy started out as a valve in ENIAC. 

--
Mike.
@millomweb - For all your info on Millom and South Copeland
via the area's premier website - 

currently unavailable due to ongoing harassment of me, my family, property & pets


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

Re: Chain Store Cleanup

Jo-2
While you're at it, please consider adding brand:wikidata=Q38076 for McDonald's you verify.

Polyglot

2015-05-03 0:28 GMT+02:00 pmailkeey . <[hidden email]>:


On 2 May 2015 at 23:18, Andy Mabbett <[hidden email]> wrote:
On 2 May 2015 at 22:28, Frederik Ramm <[hidden email]> wrote:

> I know you have a Wikidata background and things may be different in
> Wikidata

I've been editing OSM longer than Wikidata has existed.

Even had I not, I don't think your attempt to analyse my "background"
has any place on this list.



Frederik, Andy started out as a valve in ENIAC. 

--
Mike.
@millomweb - For all your info on Millom and South Copeland
via the area's premier website - 

currently unavailable due to ongoing harassment of me, my family, property & pets


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



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

Re: Chain Store Cleanup

Frederik Ramm
In reply to this post by Andy Mabbett
Andy,

On 05/03/2015 12:18 AM, Andy Mabbett wrote:
> Even had I not, I don't think your attempt to analyse my "background"
> has any place on this list.

I didn't mean to "analyse your background" (not did I intend my Wikidata
reference to be derogatory in an way), and I'm sorry if I sounded
condescending.

Your line of reasoning expressed in the post I replied to was:

"If we are more interested in mapping the actual sign on a shop than the
generic name that the shop should have because of its affiliation, then
we are surely also interested in mapping a bunch of derelict shops."

You said that in a way that implied you weren't thinking very highly of
someone mapping derelict shops (because you had collected information
about them and decided not to add it yourself), or (by logical
consequence) someone who maps the actual name on the sign.

I would like to say that

1. Just because it is more OSM-like to map the actual sign than some
internal/contractual name-that-should-be-on-the-sign-but isn't, doesn't
mean we have to be interested in derelict shops.

2. If someone were interested in mapping derelict shops that are still
recognizable as such (as opposed to a shop long replaced by something
else), then that would be totally fine; the information that a building
houses a derelict shop could for example be used as a landmark or by
someone scouting for a location where to open a new shop.

Bye
Frederik

--
Frederik Ramm  ##  eMail [hidden email]  ##  N49°00'09" E008°23'33"

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

Re: Chain Store Cleanup

Frederik Ramm
In reply to this post by Frederik Ramm
Hi,

On 05/01/2015 05:07 PM, Frederik Ramm wrote:
> I don't think that is something that really advances the quality in OSM,
> and I would encourage you to grab a notepad and venture outside to do
> some mapping. That way you wouldn't be scripting world-wide cleanup
> operations but who knows, you might actually add real value to OSM.

I would still say that to anyone who suggests some sort of world wide
"cleanup" but I recognize that it was a bit silly saying it to you
(Andrew) - given that you are probably the human being with the most
survey-based edits in OSM.

Bye
Frederik

--
Frederik Ramm  ##  eMail [hidden email]  ##  N49°00'09" E008°23'33"

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

Re: Chain Store Cleanup

Shaun McDonald
In reply to this post by Janko Mihelić

On 1 May 2015, at 16:56, Janko Mihelić <[hidden email]> wrote:

If we want to unify a kind of chain store, I think we should leave the name tag, and focus on other tags. Some examples are ref:vatin=* for the vat id of the store, brand:wikidata=* for the wikidata id of the brand owner, website=* for the central website, or we can find a new better tag that should be the same with all certain chain stores. Names should be left to the local mapper.

Janko

When I add the website to an object in OSM, I try to add a link to the specific branch in the branch finder rather than the home page, thus would need to be careful with the matching, as exact matching won’t work. Also need to cater for when one person uses https url, and another uses http for example.

Shaun

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

Re: Chain Store Cleanup

SomeoneElse-2
In reply to this post by Colin Smale


On 02/05/15 18:01, Colin Smale wrote:

>
> A bit of a meta-discussion.... I wonder why this topic is not going
> the same way as the debate on talk-gb last November-December in which
> it was proposed to tidy up and normalise various spelling variants?
> There was a lot of vehement opposition to any automated "corrections"
> as many chains are inconsistent with their own orthography and only
> on-the-ground mappers would be able to tell whether or not there is an
> apostrophe present in the signage at this particular branch (etc.
> etc., you get this idea).
>
>

Possibly because the various points of view were well articulated then
and many of them (chains don't rebrand instantly and consistently,
fixing misspellings gives a false impression of quality - or not) have
already been raised here in some form - everyone piling in with "me
too!" posts doesn't help the list become more navigable.

Cheers,

Andy

... and apologies for a largely content-free meta-post :)


_______________________________________________
talk mailing list
[hidden email]
https://lists.openstreetmap.org/listinfo/talk
12