Re: [Talk-us] Typical maxweight signs in USA? (editor developmnent question)

[Peter Dobratz]

Thanks for trying to standardize on this.  I've seen a few of these maximum
weight signs and was unsure of how to tag.

>From what I've seen in the United States, I've seen maximum weights listed
as both lbs (pounds) and tons (where 1 ton = 2000 pounds).

In Portland, Oregon, I've recently come across the following following text
on signs:

WEIGHT
RESTRICTED
BRIDGE
SINGLE AXLE TRUCKS
50,000 LBS MAX
COMBINATION TRUCKS
80,000 LBS MAX

WEIGHT LIMIT REDUCED
ANY SINGLE AXLE 20,000 LBS
ANY TANDEM AXLE 34,000 LBS
ANY GROSS WEIGHT 105,500 LBS
LEGAL AXLE LOADS ONLY

WEIGHT
LIMIT
6
TONS

Reading this page, I see the potential ambiguity extends deeper than I
realized (short ton, metric ton, long ton)
https://en.wikipedia.org/wiki/Tonne

Peter



On Tue, Jun 25, 2019 at 3:35 AM Mateusz Konieczny 
wrote:

> How often weight limit signs other than plain
> "Weight limit X tons"
> https://wiki.openstreetmap.org/wiki/File:MUTCD_R12-1.svg
>
> and
>
> R12-5 https://wiki.openstreetmap.org/wiki/File:MUTCD_R12-5.svg
> appear?
>
> Some of them are listed at
> https://wiki.openstreetmap.org/wiki/Key:maxweight#United_States
> but I am unsure is it case of something actually popular or rare curiosity?
>
> -
>
> I am asking as I work on extending StreetComplete by adding max weight
> quest for bridges
> and I want to support also USA-style weight limits.
>
> 
>
> PS StreetComplete is a bit different editor, available as an Android
> application - it allows to make
> a very limited set of edits, but all can be done by answering simple
> questions with no OSM specific
> knowledge necessary.
>
> PPS In case that somebody used StreetComplete and noticed some stupid and
> preventable behavior,
> reports at https://github.com/westnordost/StreetComplete/issues are
> welcomed (or
> within this thread if someone is unable/unwilling to make a Github account)
>
>
>
> ___
> 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

[Talk-us] motel vs. hotel

[Peter Dobratz]

How do you distinguish between the tourism=hotel and tourism=motel tags?

The criteria that I was imagining is that a motel is a single story
building where you have the ability to park you car directly outside of
your room. A hotel would be other types of buildings such as multi-story
where most guests cannot park directly outside their room.

There's the curious case of the two Motel 6 facilities directly across the
road from each other.  I had marked these as tourism=hotel based on the
building architecture, but maybe all Motel 6's should be tourism=motel?

https://www.openstreetmap.org/note/1645570

What do you think?

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

Re: [Talk-us] USPS Post Boxes

[Peter Dobratz]

It would be good to standardize on operator tags for the 4 major carriers
as you mentioned.  The logos for DHL, FedEx, and UPS have those exact
letters in them, so if people map what they see, then they will end up with
those exact values.  However, as mentioned, the USPS logo actually contains
the text United States Postal Service.

I could be convinced to switch my tagging from United States Postal Service
to USPS.  Are there any arguments to support the short form beyond it being
easier to type?



Both "USPS" and "United States Postal Service" are used within the text of
this page:
"The History Behind the USPS Logo"
https://uspsblog.com/the-history-behind-the-usps-logo/

Looking at USPS trademarks
https://pe.usps.com/text/imm/immc1_006.htm
and the USPS FAQ
http://faq.usps.com/
you can see that "U.S. Postal Service" is another possible variation on the
same name.

Looking through Taginfo, OSM tag usage reflects the same inconsistencies as
in the USPS FAQ.  The following similar key/vales exist:
operator=USPS
operator=United States Postal Service
operator=US Postal Service
operator=U.S. Postal Service

OSM Tag History can be used to graph total usage of these tags over time:

http://taghistory.raifer.tech/
[image: taghistory_USPS.png]

All 4 lines on the graph are increasing over time with USPS and United
States Postal Service being an order of magnitude higher than the
variations with only the US abbreviated.  The graph doesn't yet reflect
Leif's efforts to standardize tagging (spike in USPS and dip in United
States Postal Service).

Once we come to a consensus, we should document on the OSM wiki pages for
amenity=post_box and amenity=post_office, and also flow to editor presets.

Peter

On Thu, Sep 6, 2018 at 8:18 PM Bryan Housel  wrote:

> Can we please leave it as USPS?  Other shippers UPS, FedEx, and DHL are
> also abbreviated.
> The “expand abbreviations” rule is really for street addresses, not
> everything in OSM.
>
> Taginfo can show you what values are currently stored in the `operator`
> tag:
> https://taginfo.openstreetmap.org/keys/operator#values
>
> If we were to switch the `operator` field in iD over to a taginfo-based
> autocomplete, 2 things would happen:
> 1. It would show exactly the results in the url above, which contains all
> the operators (not just the postal ones).  It would not be a good list to
> pick from.
> 2. The taginfo service would crash.  We’ve been asked by Jochen not to do
> lookups against tags with a lot of distinct values (like “name” and
> “operator”) because queries like this are resource intensive.
>
> Thanks, Bryan
>
>
>
>
> On Sep 6, 2018, at 6:44 PM, Leif Rasmussen <354...@gmail.com> wrote:
>
> Hi Peter,
> I would like to do the operator value conversion this weekend, but would
> also like to make sure to listen to all the reasons for not expanding
> USPS.  These include having the tagging style as consistent as possible,
> the long version being harder to type, a list of semicolon separated
> expanded operators being too long, and more.  These are valid concerns,
> which should be addressed before anything else happens.
> First, for keeping the tagging style as consistent as possible, each post
> box will be given the tag "operator:wikidata"="Q668687".  This way, even if
> the operator=* tags are changed later on, all post boxes will still be
> consistent and easy to querry.
> Second, for making sure that the operator tag is not too hard to type, I
> think that adding a taginfo based autocomplete to the operator=* field in
> iD could solve this.  I would have to open an issue about that on Github,
> though.
> Third, another fear was that a list of expanded operators separated by
> semicolons would be hard to work with.  I have no easy solution for this.
> Perhaps, iD could have something to organize the operator tags into blocks,
> rather than just text like other fields.
> I would like to convert the tags this weekend, so please let me know if
> you have any suggestions for how to make the conversion less problematic.
> Thank you,
> Leif Rasmussen
> ___
> 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
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] USPS Post Boxes

[Peter Dobratz]

Leif,

What's the next step in your process on this?

I prefer the "operator"="United States Postal Service".  Not only does it
match the sticker on the side of many blue mail boxes (
https://wiki.openstreetmap.org/wiki/File:Post_office_drivethrough_lane.jpg
), it follows the general OSM guideline of not putting abbreviations in
names of things (
https://wiki.openstreetmap.org/wiki/Names#Abbreviation_.28don.27t_do_it.29
).  The https://www.usps.com/ website seems to use both USPS and United
States Postal Service.

Thanks,
Peter

On Tue, Aug 28, 2018 at 5:02 PM Jack Burke  wrote:

> Because they are labeled "United States Postal Service" with big stickers,
> that's how I've been tagging them.
>
> -jack
>
> --
> Typos courtesy of fancy auto spell technology
>
> On August 28, 2018 4:02:40 PM EDT, Jmapb  wrote:
>>
>> On 8/28/2018 3:31 PM, Leif Rasmussen wrote:
>>
>> Hi everyone,
>> A couple of days ago, I noticed that different post boxes in the United
>> States had different ways of tagging that they were part of the USPS
>> system.  Roughly 60% had "operator"="USPS", 40% "operator"="United States
>> Postal Service", and about 15 similar to "operator"="United States Post
>> Services" or "United States Post Office".  I wanted to make them all the
>> same, so I asked on the OSMUS Slack group about the tags, and most people
>> supported "USPS".  I then proceeded to upload a changeset manually
>> converting 1500 "United States Postal Service" or similar to "USPS".  I
>> should have waited longer before uploading.
>> https://www.openstreetmap.org/changeset/62020302
>> https://overpass-turbo.eu/s/Boq
>> All post boxes in the USPS system now have "operator"="USPS".
>> After uploading, people expressed concern about the choice of "USPS" over
>> "United States Postal Service".  The wiki states that abbreviations should
>> be expanded, "USPS" could be confused with other agencies
>> , and this
>> counted as an automated edit, which should have had better planning with
>> the OSM community.
>> I would now like to propose converting all post boxes with the tag
>> "operator"="USPS" to "operator"="United States Postal Service".  I would
>> also like to add the tag "operator:wikidata"="Q668687" to the post boxes
>> that currently have "operator"="USPS".  Please reply with any feedback or
>> concerns you have with this proposal.
>>
>>
>> Thanks for looping us in, Leif. Haven't made it to the slack group yet,
>> but when I noticed that change I thought to myself "thank the Lord, it's
>> over!" Every time I've tagged a post box I've managed to convince myself
>> that I did it wrong the previous time. So I rotated between "USPS", "usps"
>> and "United States Postal Service" and figured I'd get around to fixing
>> them when my brain settled down. Seeing someone else take the initiative
>> was a relief, despite the understandable grumbling about automated edits.
>>
>> Now that it's done, I'm in favor of making an exception to the
>> no-abbreviations rule for a fixed set of couriers, so sticking with USPS.
>> It's easy to tag and read, and I don't think it's truly ambiguous... if the
>> United States Power Squadrons start a courier service, *they* should be
>> tagged with the fully expanded name.
>>
>> Another reason is that there are plenty of private company post boxes,
>> and I'd like the tagging style to be consistent as possible, with the
>> commonly-used compact versions of their names. I feel operator=UPS,
>> operator=FedEx, operator=DHL will be easier to write, maintain, and read
>> than operator=United Parcel Service, operator=Federal Express, operator=..
>> well, I still can't figure out what DHL actually stands for.
>>
>> Also -- not an issue for postboxes per se, but I've found occasion to tag
>> some some shops with the shipping companies they offer, and the idea of a
>> semicolon-seperated list of fully expanded shipping company names makes me
>> weary.
>>
>> (No problem with adding the operator:wikidata tag.)
>>
>> Thanks, Jason
>>
>>
>> ___
> 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] Naming numbered roads as "State Route X", "Interstate X", etc.

[Peter Dobratz]

On Sat, Sep 1, 2018 at 5:28 AM Greg Troxel  wrote:

>
>   From: Albert Pundt 
>   Subject: [Talk-us] Naming numbered roads as "State Route X", "Interstate
> X", etc.
>   To: "talk-us@openstreetmap.org Openstreetmap"  >
>   Date: Fri, 31 Aug 2018 22:06:50 -0400 (10 hours, 17 minutes, 15 seconds
> ago)
>   Attachment: [3. text/html]...
>
>   I notice the user SSR_317 has been adding the route numbers of unnamed
>   roads to the name=* tag of roads around Indianapolis. For example,
>   name=Interstate
>   465, name=US 31, name=State Route 37, etc. Isn't this practice frowned
> upon
>   as being redundant and not reflecting the lack of a proper name to the
>   road? This seems to be the case around the country. All route numbers
> were
>   listed in alternate names of the roads in the original TIGER data, but
> the
>   vast majority of these have been removed in favor of route relations and
>   ref=* tags.
>
>   I removed these name tags from the affected roads, but the user has since
>   re-added them.
>
> Seconded or thirded: a route number is not an actual name and does not
> belong in the name tag, or even alt_name.
>
> This has come up (in discussion, not actual edits as far as I know)
> because there are some roads in Massachusetts that have actual names and
> route numbers (example: name="Grand Army of the Republic Highway"
> ref="US 6"), but for reasons unknown there are street addresses like
> "3570 Route 6".  Even though there is zero evidence that "Route 6" is is
> any way a street name.   As far as I know this is limited to on the
> order of a half dozen roads, 1 US highway and about 5 state highways,
> mainly in the southeast and cape.
>
> So beyond agreeing that sticking things in name (presumably on the
> notion that everything should have a name, even though that notion is
> confused as the real world does not have that property), I wanted to
> point out that just because a building near a road has an *address* with
> a name, doesn't mean that the road itself has that name.
>

As a counterpoint to this, consider that street names in OSM include
directional prefix, base name, suffix, and directional suffix.  The post
office breaks down the standard components of delivery address line here:

https://pe.usps.com/text/pub28/28c2_012.htm

In this example of "101 W Main St S Apt 12" we would represent this address
OSM in the following tags on the POI Node or Area:
addr:housenumber=101
addr:street=West Main Street South
addr:unit=12

On the nearby Way for the street, there would be a "name=West Main Street
South" tag.

Values in addr:street do need to include numbered routes when the numbered
route is part of the actual address.  The post office shows some examples
of various variations on standard names for numbered routes here:

https://pe.usps.com/text/pub28/28apf.htm

To cite a specific example of how we might map something, consider the town
of Waldport, Oregon.

https://www.openstreetmap.org/#map=18/44.42718/-124.06667

As you can see, there is a US Route 101 running north-south through town.
Roads north of Northwest Hemlock Street include Northwest as part of their
names and roads south of Northwest Hemlock Street include Southwest as part
of their name.  US Route 101 is currently mapped in OSM with
"name=Northwest Highway 101" for the portions north of Northwest Hemlock
Street and "name="Southwest Highway 101" for the portions south of
Northwest Hemlock Street.  If we drop the name tag from this road in OSM,
then we lose the Northwest and Southwest directional prefix.  I think we
should retain the name tags on roads like this.

Here is an examples of a POI along this route:

https://www.grand-central-pizza.com/
Grand Central Pizza
245 SW Hwy 101
Waldport, OR 97394

USPS standard format of the address:
245 SW HIGHWAY 101
WALDPORT, OR 97394-3036

https://www.openstreetmap.org/way/463377211
addr:housenumber=245
addr:street=Southwest Highway 101
addr:city=Waldport
addr:state=OR
addr:postcode=97394

I believe this is correct in OSM.  The value in addr:street matches the
name tag on the nearby road.  The fact that the road is a member of the
route 101 route relation and ref=US 101 tag on the Way does not capture the
directional prefix on portions of the road.

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

Re: [Talk-us] STOP - All vehicles must register

[Peter Dobratz]

The first thing that comes to mind is a barrier=toll_booth:

https://wiki.openstreetmap.org/wiki/Tag:barrier%3Dtoll_booth

Perhaps also with a fee=no to indicate that you don't have to pay any money.

Peter

On Fri, Apr 20, 2018 at 11:10 AM Kevin Kenny 
wrote:

> The case of non-hard-surface roads brought this to mind.  There are a few
> roads across the Adirondack Park that are open to the public (in summer)
> and have endpoints that look like
>
> http://i65.tinypic.com/2enq9ew.jpg
>
> In this case, I already recognize - tag the two cabins (they are a ranger
> station, not often staffed), the gate, the outhouse, the well, perhaps the
> signage. But what to do about the kiosk/register?
>
> The requirement is that vehicles (or bicycles, pedestrians, horses, etc.)
> must stop, and the party leader fills out an entry in the book that's kept
> in one of the boxes on the kiosk. (There happen to be two books at that
> location; one for vehicles that are merely traversing the road and another
> for hikers, skiers, and riders who are venturing off-road into the
> backcountry) The understood procedure is that it's all right to pass as
> long as the gate is unlocked, and the expectation is that drivers will
> leave the gate as they found it - close it if they opened it.
>
> Is there an OSM tag for "this is where you stop and execute formalities"
> that I've missed? I know that calling it out as an access restriction on
> the road has been controversial, and I don't tag that because any scheme
> gets someone up in arms. For this reason, tagging that registration is a
> requirement to drive on the road is Out Of Scope. I'm simply trying to tag
> the on-the-ground "this is where you stop and register."
>
> By the way, I'd tag the road in the picture as at least
> highway=unclassified and possibly even tertiary. In the seasons when it's
> open (usually late April to early November, depending on snow conditions)
> it's the main route between the villages of Indian Lake and Inlet, and when
> it's closed the trip is many miles longer. The registration requirement is
> largely to make sure that drivers have all the warnings in front of their
> faces - they're going to be spending the next hour driving on an unpaved
> road across some very forbidding country, with no services available and no
> mobile phone coverage nor radio call boxes. (It's also there because Search
> and Rescue isn't going to give quite the same treatment if someone goes
> missing without having signed in.)
>
> There are simpler but similar setups on some footways - where the hiker
> must not only register, but take the carbon copy along as proof of
> registration. (The plan is that sometimes the number of forms will be
> limited as a primitive means of 'first come, first served' capacity
> control. I don't think it's ever actually got that far.)
>
>
> ___
> 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] Walmart Import

[Peter Dobratz]

Hi Ilya,

This copnflation audit software tool is really good to allow multiple OSM
users to review the data before changes are made to OSM data.  I didn't
initially realize the amount of manual review of the data that is being
done prior to importing the data into OSM.

I've been reviewing a bunch of Walmarts using this tool and there will be a
lot of new good data being added to OSM through this process.  It seems
like people are generally preserving the choice of shop=supermarket or
shop=department_store and also name of "Walmart" or "Walmart Supercenter"
as the choice between the two is somewhat arbitrary.  Also, in many cases
the addr:full tag is being omitted if that same information is already
contained in existing addr:housenumber, addr:street, and addr:unit tags.

Peter

On Thu, Dec 28, 2017 at 6:22 AM, Ilya Zverev  wrote:

> Hi Peter,
>
> Thank you for the extended example of opening_hours containing holidays
> exceptions. While validating the Walmart import, I've encountered a couple
> of these, albeit simpler.
>
> My opinion on tags does not have an extra weight in relation to other
> mappers. With the introduction of the "Conflation Audit" website, to which
> I link, the data owner does not dictate which tags to overwrite as well. By
> participating in validation, you choose which tags to keep and which to
> override, and where to place imported points.
>
> For example, when seeing "24/7; Dec 25 off" on a Walmart, I clicked on
> that line, so it doesn't get overwritten on import. The line becomes
> highlighted, which means that's what will be in OSM after the import. See
> for an example this object:
>
> http://audit.osmz.ru/browse/walmart/4404
>
> Sorry I didn't get to answer earlier,
> Ilya
>
> > 22 дек. 2017 г., в 10:46, Peter Dobratz  написал(а):
> >
> > Ilya,
> >
> > I'm trying to wrap my head around how making this a "frictionless series
> of imports" is going to work.  So if a local mapper edits details on a
> Walmart, those details could potentially be swiftly overwritten with your
> data?
> >
> > As you can see, the opening hours for next week are non-standard due to
> the Christmas holiday.  What if someone decides they want to add that level
> of detail to the opening_hours tag:
> > opening_hours=00:00-01:00,05:00-24:00; Dec 24 5:00-18:00; Dec 25 off;
> Dec 26 06:00-24:00
> >
> > How long until you automatically replace this with "05:00-01:00" ?
> >
> > Do you see the problem with doing that?
> >
> > As you say there are differences of opinion in how things are tagged.
> Why does your opinion get to have more weight?
> >
> > Peter
> >
> > On Thu, Dec 21, 2017 at 1:12 AM, Ilya Zverev  wrote:
> > Hi Peter,
> >
> > Thank you for suggestions.
> >
> > First, the highlighted tag value is what goes into OSM. In your case,
> the import will keep the shop=department_store.
> >
> > Regarding updates to opening_hours, you are suggesting I parse each
> opening_hours value and then compare these? It would be quite hard and in
> my opinion excessive. With the two values being equivalent, I don't how the
> data becomes worse. A few times I omitted "Mo-Su", I was being told it's
> better to specify the weekdays, so it is again a matter of opinion.
> >
> > In your example, you override the definition for Monday, so it doesn't
> demonstrate anything besides how complex the opening_hours notation is.
> >
> > The rest I answered in imports@, and some of it goes against what other
> community members suggest, so again I can conclude that is a matter of
> opinion and not important one way or another:
> >
> > * URL is provided by Walmart and is much better than what we have.
> /whats-new can be fixed later, and does not really matter, because it still
> takes to a store page.
> > * addr:full is provided by Walmart and may be used to improve addressing
> where there are no addr:* tags. Sorry the importing script cannot do
> conditional tagging.
> > * operator was recommended by community members, and is a good tag to
> filter all Walmarts.
> > * ref:walmart will be used for updating the data. Ref may refer not only
> to a store, but to a building or another feature.
> >
> > Please understand that this is not a one-off handcrafted import. We are
> working on a process for frictionless series of imports, with regular
> updates later on. I understand you have mapped a Walmart and feel
> protective of it. I felt the same a few years into OSM, because everything
> you add to the map is important. With this import, I believe it does no

Re: [Talk-us] Walmart Import

[Peter Dobratz]

Ilya, I apologize for the tone of my last message.  Ian has done a good job
of restating my question in a reasonable manner.

Overall, I think this effort is great for filling in gaps in OSM data and I
applaud your efforts.  If I had looked more at the data review tool, I
would have realized that there is more of an element to manual review of
the data.   I will spend some time helping review the data.

Peter

On Fri, Dec 22, 2017 at 5:13 AM, Ian Dees  wrote:

> Hi list,
>
> I wanted to remind everyone that we're all trying to make the map better
> here. Please be careful with the words you choose to use so that they
> aren't condescending or attacking.
>
> For example, in this message from Peter it's very easy to read the
> sentences about opening hours as a personal attack on Ilya and not a new
> critique of the proposal. It might have been better to write:
>
> "I'm worried that this import will replace more specific holiday opening
> hours with the generic year round string. Will the import process include a
> check to make sure the object hasn't changed since you ran the conflation
> so that changes like this don't get overwritten?"
>
> There's no need to make the claim that the import gets to have more weight
> (no one said it did) and there's no need to direct the message at Ilya. As
> has been discussed before, he's helping Brandify with tools.
>
> Happy holidays,
> Ian
>
>
> On Dec 22, 2017 01:47, "Peter Dobratz"  wrote:
>
> Ilya,
>
> I'm trying to wrap my head around how making this a "frictionless series
> of imports" is going to work.  So if a local mapper edits details on a
> Walmart, those details could potentially be swiftly overwritten with your
> data?
>
> As you can see, the opening hours for next week are non-standard due to
> the Christmas holiday.  What if someone decides they want to add that level
> of detail to the opening_hours tag:
> opening_hours=00:00-01:00,05:00-24:00; Dec 24 5:00-18:00; Dec 25 off; Dec
> 26 06:00-24:00
>
> How long until you automatically replace this with "05:00-01:00" ?
>
> Do you see the problem with doing that?
>
> As you say there are differences of opinion in how things are tagged.  Why
> does your opinion get to have more weight?
>
> Peter
>
> On Thu, Dec 21, 2017 at 1:12 AM, Ilya Zverev  wrote:
>
>> Hi Peter,
>>
>> Thank you for suggestions.
>>
>> First, the highlighted tag value is what goes into OSM. In your case, the
>> import will keep the shop=department_store.
>>
>> Regarding updates to opening_hours, you are suggesting I parse each
>> opening_hours value and then compare these? It would be quite hard and in
>> my opinion excessive. With the two values being equivalent, I don't how the
>> data becomes worse. A few times I omitted "Mo-Su", I was being told it's
>> better to specify the weekdays, so it is again a matter of opinion.
>>
>> In your example, you override the definition for Monday, so it doesn't
>> demonstrate anything besides how complex the opening_hours notation is.
>>
>> The rest I answered in imports@, and some of it goes against what other
>> community members suggest, so again I can conclude that is a matter of
>> opinion and not important one way or another:
>>
>> * URL is provided by Walmart and is much better than what we have.
>> /whats-new can be fixed later, and does not really matter, because it still
>> takes to a store page.
>> * addr:full is provided by Walmart and may be used to improve addressing
>> where there are no addr:* tags. Sorry the importing script cannot do
>> conditional tagging.
>> * operator was recommended by community members, and is a good tag to
>> filter all Walmarts.
>> * ref:walmart will be used for updating the data. Ref may refer not only
>> to a store, but to a building or another feature.
>>
>> Please understand that this is not a one-off handcrafted import. We are
>> working on a process for frictionless series of imports, with regular
>> updates later on. I understand you have mapped a Walmart and feel
>> protective of it. I felt the same a few years into OSM, because everything
>> you add to the map is important. With this import, I believe it does not
>> make the data worse. More attributes is not bad, even if some of these are
>> redundant. The main thing is, until now there were zero mappers who care
>> about keeping all the Walmart stores in OSM up-to-date, and after, there
>> will be more. To me, that is a good thing.
>>
>> Thanks,
>> Ilya
>>
>>
>> > 21

Re: [Talk-us] Walmart Import

[Peter Dobratz]

Ilya,

I'm trying to wrap my head around how making this a "frictionless series of
imports" is going to work.  So if a local mapper edits details on a
Walmart, those details could potentially be swiftly overwritten with your
data?

As you can see, the opening hours for next week are non-standard due to the
Christmas holiday.  What if someone decides they want to add that level of
detail to the opening_hours tag:
opening_hours=00:00-01:00,05:00-24:00; Dec 24 5:00-18:00; Dec 25 off; Dec
26 06:00-24:00

How long until you automatically replace this with "05:00-01:00" ?

Do you see the problem with doing that?

As you say there are differences of opinion in how things are tagged.  Why
does your opinion get to have more weight?

Peter

On Thu, Dec 21, 2017 at 1:12 AM, Ilya Zverev  wrote:

> Hi Peter,
>
> Thank you for suggestions.
>
> First, the highlighted tag value is what goes into OSM. In your case, the
> import will keep the shop=department_store.
>
> Regarding updates to opening_hours, you are suggesting I parse each
> opening_hours value and then compare these? It would be quite hard and in
> my opinion excessive. With the two values being equivalent, I don't how the
> data becomes worse. A few times I omitted "Mo-Su", I was being told it's
> better to specify the weekdays, so it is again a matter of opinion.
>
> In your example, you override the definition for Monday, so it doesn't
> demonstrate anything besides how complex the opening_hours notation is.
>
> The rest I answered in imports@, and some of it goes against what other
> community members suggest, so again I can conclude that is a matter of
> opinion and not important one way or another:
>
> * URL is provided by Walmart and is much better than what we have.
> /whats-new can be fixed later, and does not really matter, because it still
> takes to a store page.
> * addr:full is provided by Walmart and may be used to improve addressing
> where there are no addr:* tags. Sorry the importing script cannot do
> conditional tagging.
> * operator was recommended by community members, and is a good tag to
> filter all Walmarts.
> * ref:walmart will be used for updating the data. Ref may refer not only
> to a store, but to a building or another feature.
>
> Please understand that this is not a one-off handcrafted import. We are
> working on a process for frictionless series of imports, with regular
> updates later on. I understand you have mapped a Walmart and feel
> protective of it. I felt the same a few years into OSM, because everything
> you add to the map is important. With this import, I believe it does not
> make the data worse. More attributes is not bad, even if some of these are
> redundant. The main thing is, until now there were zero mappers who care
> about keeping all the Walmart stores in OSM up-to-date, and after, there
> will be more. To me, that is a good thing.
>
> Thanks,
> Ilya
>
>
> > 21 дек. 2017 г., в 1:22, Peter Dobratz  написал(а):
> >
> > Ilya,
> >
> > Here's a Walmart that's been built in the last few years I recently
> added to OSM:
> >
> > http://audit.osmz.ru/browse/walmart/5935
> >
> > What's currently in OSM represents my own mapping style, but I think
> it's worth discussing the differences before you change them across the
> whole country.
> >
> >
> > If I read this correctly, you are planning on changing the top-level tag
> from shop=department_store to shop=supermarket.  I have been using
> shop=supermarket only for Walmarts branded as "Walmart Neighborhood Market"
> and using shop=department_store for all other Walmarts.  For me the main
> distinction is that most Walmarts have departments for things like clothing
> and electronics that don't exist in the "Walmart Neighborhood Market"
> stores.
> >
> >
> > You are proposing changing the opening_hours tag from
> "00:00-01:00,05:00-24:00" to "Mo-Su 05:00-01:00".  When I'm adding opening
> hours, I avoid timespans that cross midnight as there is some difference of
> opinion as to what those timespans actually represent.  I think most people
> would say that these two opening hours are equivalent, but if you were to
> have something like "Mo-Sa 05:00-01:00; Su 10:00-22:00" this illustrates
> the problem with different interpretations of times that span midnight:
> > http://openingh.openstreetmap.de/evaluation_tool/?EXP=Mo-Sa%
> 2005%3A00-01%3A00%3B%20Su%2010%3A00-22%3A00
> > http://projets.pavie.info/yohours/?oh=Mo-Sa%2005:00-01:
> 00;%20Su%2010:00-22:00
> > I also omit "Mo-Su" when writing opening_hours as, in my opinion, this
> doesn&#x

Re: [Talk-us] Walmart Import

[Peter Dobratz]

On Wed, Dec 20, 2017 at 3:23 PM, Toby Murray  wrote:

> On Wed, Dec 20, 2017 at 4:22 PM, Peter Dobratz  wrote:
> > Ilya,
> >
> > Here's a Walmart that's been built in the last few years I recently
> added to
> > OSM:
> >
> > http://audit.osmz.ru/browse/walmart/5935
> >
> > What's currently in OSM represents my own mapping style, but I think it's
> > worth discussing the differences before you change them across the whole
> > country.
> >
> >
> > If I read this correctly, you are planning on changing the top-level tag
> > from shop=department_store to shop=supermarket.  I have been using
> > shop=supermarket only for Walmarts branded as "Walmart Neighborhood
> Market"
> > and using shop=department_store for all other Walmarts.  For me the main
> > distinction is that most Walmarts have departments for things like
> clothing
> > and electronics that don't exist in the "Walmart Neighborhood Market"
> > stores.
>
> This was my first thought as well but apparently we are wrong. The
> wiki is actually fairly clear about it:
> https://wiki.openstreetmap.org/wiki/Tag:shop%3Dsupermarket
>
> Under the "Similar Tags" section on that page:
> "A department store that sells a full selection of groceries (e.g.
> Walmart Supercentre) which is often called a "hypermarket" should be
> tagged shop=supermarket"
>
> And over on the shop=department_store page it says that only
> non-supercenter stores should use the department_store tag.
>
> I'm not saying it makes sense to me... but this is apparently the
> convention. Decided upon by... someone :)
>
> Toby
>


OSM user andrewpmk added similar language in May 2015 to both the
shop=supermarket and shop=department_store wiki pages describing that a
shop=department_store becomes a shop=supermarket if there is a section for
selling fresh grocery items.  The wiki change comments don't give any
rationale for the new wording, and to me the idea is counterintuitive.  I
have attempted to communicate with user andrewpmk in the past, but never
received any response.  Is anyone aware of a discussion that took place
prior to introducing these changes to the wiki?

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

Re: [Talk-us] Walmart Import

[Peter Dobratz]

Ilya,

Here's a Walmart that's been built in the last few years I recently added
to OSM:

http://audit.osmz.ru/browse/walmart/5935

What's currently in OSM represents my own mapping style, but I think it's
worth discussing the differences before you change them across the whole
country.


If I read this correctly, you are planning on changing the top-level tag
from shop=department_store to shop=supermarket.  I have been using
shop=supermarket only for Walmarts branded as "Walmart Neighborhood Market"
and using shop=department_store for all other Walmarts.  For me the main
distinction is that most Walmarts have departments for things like clothing
and electronics that don't exist in the "Walmart Neighborhood Market"
stores.


You are proposing changing the opening_hours tag from
"00:00-01:00,05:00-24:00" to "Mo-Su 05:00-01:00".  When I'm adding opening
hours, I avoid timespans that cross midnight as there is some difference of
opinion as to what those timespans actually represent.  I think most people
would say that these two opening hours are equivalent, but if you were to
have something like "Mo-Sa 05:00-01:00; Su 10:00-22:00" this illustrates
the problem with different interpretations of times that span midnight:
http://openingh.openstreetmap.de/evaluation_tool/?EXP=Mo-Sa%
2005%3A00-01%3A00%3B%20Su%2010%3A00-22%3A00
http://projets.pavie.info/yohours/?oh=Mo-Sa%2005:00-01:00;%
20Su%2010:00-22:00
I also omit "Mo-Su" when writing opening_hours as, in my opinion, this
doesn't add value to the data and from a human-readable perspective it
looks very similar to "Mo-Sa" which of course means something different.
Another problem with "Mo-Su" is that it is equivalent to "Su-Sa" or any of
the other 7 possible variations.
One of the problems with the opening_hours tag in general is that there are
many ways of representing the same set of opening hours and I'm not aware
of any efforts to establish a canonical representation for opening_hours.
Can you avoid updating opening_hours if the new value is equivalent to the
old value?


For the website tag, I prefer to use simple URLs:
http://www.walmart.com/
https://www.walmart.com/store/5935/tigard-or/whats-new
The advantage to using the basic URL is that it will still continue to work
in the future even if Walmart decides to reformat their webiste.  The
disadvantage is that it does not go to a page specific to this particular
store.  Reading through the imports list, I see there is already some
discussion about whether to use the "whats-new" or "details" URL for each
store.  What happens in a year when the "whats-new" URL no longer works?


I would omit the addr:full tag at least on objects where addr:street and
addr:housenumber already exist.


I would also omit the operator=Walmart tag.  I only use operator if the
value of the operator is different from the value of name.


The object already has ref=5935.  The ref:walmart=5935 tag is superfluous
and I would omit it.  It's fairly well established practice to use the
plain ref tag for mapping store identifiers.  We don't need a separate
ref:* tag for every store.


Overall, I think you need to put in more effort to determine whether the
data that you are using to overwrite existing OSM data is in fact better.
Since you want to get this done "before the New Year", maybe it would be
better to just leave existing OSM tags alone for now and then manually
check them against your dataset later.

I don't have any specific examples handy, but I've seen cases where the
data on the company website is wrong for things like postal code, house
numbers, or streets.  Sometimes it seems that companies change their
addresses to get their geocoding service of choice to put the placemarker
in the right spot, so I would be especially wary of changing existing OSM
data.

Peter


On Wed, Dec 20, 2017 at 11:45 AM, Ilya Zverev  wrote:

> Hi,
>
> As some of you know, Brandify company wants to import all the Walmart
> locations in the US into OpenStreetMap. They have full permission to do
> that. See the message from their VP Product, Damian, for more detailed
> explanation:
>
> https://lists.openstreetmap.org/pipermail/imports/2017-Decem
> ber/005279.html
>
> After a brief discussion on the imports@ list, they made a few proposed
> changes, and I re-uploaded the result to my imports validation website.
> Could you please look through a couple dozen points, to ensure they will be
> added to correct locations with correct tags? The more the better — I plan
> to to a few hundred myself:
>
> http://audit.osmz.ru/project/walmart
>
> I really hope we can finish this before the New Year.
>
> Thanks,
> Ilya
>
> ___
> 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

[Talk-us] Pokémon Go no officially using OpenStreetMap

[Peter Dobratz]

I'm not sure how many active OSM contributors also play Pokémon Go, but the
game is now officially using OSM for the basemap that players see in the
game (previously was using Google Maps data for the base map).  The in-game
about screen has text in the bottom of the License section correctly
attributing OSM.

As previously mentioned, the points of interest in the game come from a
separate dataset.  And it's also not clear whether OSM data is used to
control where the characters spawn in the game.  Notably, the game does not
spawn any characters inside of school boundaries, but it's not clear
whether those boundaries come from OSM or a separate dataset.

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

[Talk-us] anyone know what software is generating these Q/A Notes?

[Peter Dobratz]

I'm seeing a pattern of OSM Notes being added where a specific question is
being asked about missing information in OSM and I was wondering if anyone
knows what software is being used to generate these notes.  See for example:

http://www.openstreetmap.org/note/1054474

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

Re: [Talk-us] Pittsburgh neighborhood boundaries mapped with admin level 9?

[Peter Dobratz]

(Appologies as I was in the middle of writing my reply when inadvertantly
hitting send.  Here's the whole message)

Boundaries below admin_level=8 are still being discussed.  There was some
discussion on this list as well as the OSM wiki

https://wiki.openstreetmap.org/wiki/Talk:United_States_
admin_level#Nine_state_improvement

Having lived in Pittsburgh, I remember that the neighborhood boundaries are
well defined and many of the street signs have the neighborhood names
printed across the top of them (epecially on more major roads with bigger
signs).

If you were to divide up Pittsburgh into smaller administrative units, how
would you do it?

Pittsburgh resides within Allegheny County.  Allegheny County is divided
into Wards and districts, some of which could be used to divide up
Pittsburgh:
http://apps.alleghenycounty.us/website/MuniPgh.asp

Pittsburgh city council is made up of 9 people who each represent a council
district of the city.  It looks like each council district covers a group
of neighborhoods (that might lend itself to making the council districts
admin_level=9 and the neighborhoods admin_level=10).  For example, council
district 5 contains the neighborhoods, Hazelwood, Glen Hazel, Greenfield,
Hays, Lincoln Place, New Homestead, and Regeant Square
http://pittsburghpa.gov/district5/about

Pittsburgh is also divided up into 32 wards, each being divided further
into a variable number of districts each.  These wards and districts are
separate from the Allegheny County wards and districts.  I'm not sure how
the cities wards relate to the neighborhood boundaries.

Peter



On Thu, Jul 27, 2017 at 10:15 AM, Peter Dobratz  wrote:

> Boundaries below admin_level=8 are still being discussed.  There was some
> discussion on this list as well as the OSM wiki
>
> https://wiki.openstreetmap.org/wiki/Talk:United_States_
> admin_level#Nine_state_improvement
>
> Having lived in Pittsburgh, I remember that the neighborhood boundaries
> are well defined and many of the street signs have the neighborhood names
> printed across the top of them (epecially on more major roads with bigger
> signs).
>
> If you were to divide up Pittsburgh into smaller administrative units, how
> would you do it?
>
> Pittsburgh reside within the Allegheny County
>
>
>
> On Wed, Jul 26, 2017 at 7:36 PM, Albert Pundt 
> wrote:
>
>> I noticed that the neighborhoods in Pittsburgh are mapped as
>> administrative boundaries with admin_level=9. Is this proper? The wiki
>> page <https://wiki.openstreetmap.org/wiki/United_States_admin_level> for
>> U.S. admin levels doesn't list any use for admin level 9 in Pennsylvania,
>> though this seems appropriate if Pittsburgh neighborhoods are true
>> administrative divisions. It just needs to be documented, or perhaps used
>> elsewhere in the state, like with the fairly distinct neighborhoods in
>> Philadelphia.
>>
>> ___
>> 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] Pittsburgh neighborhood boundaries mapped with admin level 9?

[Peter Dobratz]

Boundaries below admin_level=8 are still being discussed.  There was some
discussion on this list as well as the OSM wiki

https://wiki.openstreetmap.org/wiki/Talk:United_States_admin_level#Nine_state_improvement

Having lived in Pittsburgh, I remember that the neighborhood boundaries are
well defined and many of the street signs have the neighborhood names
printed across the top of them (epecially on more major roads with bigger
signs).

If you were to divide up Pittsburgh into smaller administrative units, how
would you do it?

Pittsburgh reside within the Allegheny County



On Wed, Jul 26, 2017 at 7:36 PM, Albert Pundt  wrote:

> I noticed that the neighborhoods in Pittsburgh are mapped as
> administrative boundaries with admin_level=9. Is this proper? The wiki
> page  for
> U.S. admin levels doesn't list any use for admin level 9 in Pennsylvania,
> though this seems appropriate if Pittsburgh neighborhoods are true
> administrative divisions. It just needs to be documented, or perhaps used
> elsewhere in the state, like with the fairly distinct neighborhoods in
> Philadelphia.
>
> ___
> 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] NJ mass road demotions?

[Peter Dobratz]

I've found that even active OSM contributors don't necessarily check the
email address associated with their OSM account.  If that is the case, they
may have not yet seen any changeset comments you have made on their
changesets.

It's good to follow up with an OSM private message as these pop-up when you
log in to www.openstreetmap.org or start an editor like JOSM.

Peter



On Sun, Jun 11, 2017 at 3:36 AM, Richard Fairhurst 
wrote:

> Bryan Housel wrote:
> > What’s an acceptable amount of time to wait for a response before I
> > just start reverting?
>
> I commented on another of granpueblo's changesets on 21st May and have also
> not had a response yet. Given that, you probably only need to wait just a
> couple of days before embarking on a revert.
>
> (More generally, we need to think about how we communicate "be bold in what
> you add, careful in what you change" to new mappers. We see this fairly
> often in the UK too - over-assertive changes from someone who, through no
> fault of their own, doesn't understand OSM conventions; and on occasion the
> response ends up putting the contributor off continuing with OSM.)
>
> Richard
>
>
>
> --
> View this message in context: http://gis.19327.n8.nabble.
> com/NJ-mass-road-demotions-tp5894719p5897753.html
> Sent from the USA mailing list archive at Nabble.com.
>
> ___
> 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

[Talk-us] greenways tagging

[Peter Dobratz]

I wanted to give a wider audience to this changeset discussion:
http://www.openstreetmap.org/changeset/46967476

Some Mapzen folks have been changing highway=residential to
highway=living_street for routes that are designated as "Greenways."
 There's a related blog post too:

https://mapzen.com/blog/bike-poll/

I haven't really seen any roads in the USA that would be similar in nature
to the living street designation found throughout Europe where kids are
allowed to play in the street because the cars are required to go walking
speed.

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

Re: [Talk-us] Relation roles for two-way way segments carrying routes in a single direction

[Peter Dobratz]

There's a bit of confusion around this and it took me quite a while playing
around with the relation editor in JOSM ("Select Next Gap") to understand
how the "forward" and "backward" roles work.

https://i.imgur.com/RGV2XOX.png

First, consider how you would create a road route relation for your example
if the red segment didn't exist.  Also, for the moment ignore the signed
cardinal directions of the route.  In that case, you would arbitrarily
chose to start from the left-most or right-most segment in the picture.  If
you chose to start with the right-most segment:

right-most (role: "")
blue (role: "")
green (role: "")
left-most (role: "")

All of the roles would be empty and you would not need to use the roles
"forward" or "backward".

Now consider that the red segment exists and you need to indicate that the
road route splits in two and then reconverges.  In this case, you would do
the following, again starting at the right side of the picture:

right-most (role: "")
blue (role: "backward")
green (role: "backward")
red (role: "forward")
left-most (role: "")

You traverse the segments where the road splits first following the
direction of the route and then backtracking to the point where the road
splits and then start adding segments with empty roles.  If either blue,
green, or red is reversed in direction, then they would need to have the
role "backward" instead of "forward", but would still retain the same
position in the list (by convention red should not be reversed since it has
oneway=yes).


So now consider that you want to change the route to indicate "east" and
"west" signed cardinal directions.

One idea is to leave the roles as "forward" and "backward" and just add a
direction=west tag on the relation object.

One idea is that "east" and "west" become synonyms for "forward".  First,
reverse the direction of blue and green and change their roles to
"forward".  Then change the roles to "east" and "west":

right-most (role: "")
blue (role: "west")
green (role: "west")
red (role: "east")
left-most (role: "")

In this case, it is critical that blue and green are not reversed again.


Another approach is to create separate route relations for each direction
of the route.  In that case, you would have one relation with
direction=east:
left-most (role: "")
red (role: "")
right-most (role: "")

And a separate relation with direction=west
right-most (role: "")
blue (role: "")
green (role: "")
left-most (role: "")

In this case, it will also help to update the name tag of the relation to
include "(East)" or "(West)" to more easily tell the route relations apart
in the editor.  Also, the relations can be added to a route_master relation
(type=route_master  route_master=road).


Overall, I think it is cleanest to create 2 route relations, 1 for each
signed cardinal direction of the route.  You don't have to arbitrarily
decide whether to start the route at the eastern or western terminus as
each route relation has a natural starting point based on the signed
cardinal direction.  You don't have to deal with roles on member Ways,
which are very confusing.  You don't have to worry about the direction of
each member way either (again in practice many of them will be marked
oneway=yes and have the direction following the flow of traffic).


Note that whatever approach you chose, the route relations will often break
when the member Ways are split or combined.  Not all OSM editors do the
right thing to preserve the contiguousness of the route relations when
performing these operations.  When people decide to add lane or speed limit
information to the roads that make up the route, they may inadvertently
break the route relation.  In the rare case where multiple users are
concurrently splitting Ways that belong to the same route relation, the
relation usually breaks.

Peter

On Mon, Dec 26, 2016 at 9:29 AM, Albert Pundt  wrote:

> So I understand that one-way ways carrying a route (e.g. a one-way pair or
> divided highway) should have relation roles of north/south/east/west, but
> say you have a situation like this . Say
> you have an east-west route that follows the primary roads in that picture.
> The eastbound direction follows the channelized right turn slip ramp,
> marked with a red arrow. The westbound direction follows the blue-arrow
> way, before turning left onto the green-arrow way.
>
> How should relation memberships and roles be assigned here? I would think
> that the slip ramp would be part of the relation, since right-turning
> traffic must follow it. Ideally, that would be given the role "east", but
> what about the green and blue ways? It might seem right to give them the
> role "west", but how then is it differentiated which direction is westbound
> for it? Since all the ways in this picture are arranged "pointing" north or
> east, the green and blue ways would need to be given the role "backward",
> which is the older way of doing things that can't specify cardinal

Re: [Talk-us] Mapping bus stops

[Peter Dobratz]

For mapping bus stops, I put a Node at the location of the bus stop sign
(next to, but not touching the road).  The following three tags are used to
specify that this is a bus stop:

highway=bus_stop
public_transport=platform
bus=yes

For the "bus stop number", use the "ref" tag:

ref=1234

If your public transport agency names their bus stops, then also include
the name tag.

As far as the line numbers, there is no direct way to specify the line
number on the bus stop itself.  Each variation of a bus route should have
its own public transport route relation.  The ref tag on the bus route
relation specifies which bus line number the route refers to.  The bus stop
Node would be a member of the public transport route relation with a role
"platform".  For each bus route relation, you put the list of bus stops
Nodes in order followed by the list of Ways in order that make up that
direction of the bus route.  There is a different bus route relation for
each direction or variation of the route.

Here's an example of a typical bus stop:

http://www.openstreetmap.org/node/819693767

As you can see, this bus stop is served by 8 bus lines.

Note that in addition to specifying the location next to the road where
people wait for the bus (the platform), you can also specify the position
on the road where the bus stops (the stop position).  I haven't bothered to
do that for the bus stops that I've added since it doesn't seem to be
adding useful information (where else is the bus going to stop along the
road besides right next to the bus stop?).

I made a wiki page with more detail about public transport tagging around
Portland, Oregon:

http://wiki.openstreetmap.org/wiki/Portland,_Oregon/Transit_Route_Relations

Peter


On Sat, Aug 6, 2016 at 7:03 PM, Kevin Morgan 
wrote:

> When I add a bus stop under what tags do i add the bus line numbers and
> bus stop number?
>
> --
>   Kevin Morgan
>   morgankev...@fastmail.fm
>
> ___
> 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] shop=chemist as "Drugstore" for Walgreens, CVS, Rite Aid, etc.

[Peter Dobratz]

After arriving at a local drugstore chain with a prescription in hand,
walking past the shampoo aisle, only to find that the pharmacy counter is
closed for the day, I've been updating tagging of drugstores and pharmacies
as described here.

For example, there is a Walgreens:

http://www.openstreetmap.org/way/266495943

The building outline has shop=chemist.

Inside the building, there are 2 Nodes, one for the pharmacy and one for a
clinic:

amenity=pharmacy
dispensing=yes
drive_through=yes
http://www.openstreetmap.org/node/4064469934

amenity=doctors
http://www.openstreetmap.org/node/4064469933

In this case, all of these entities have different opening hours.  Other
contact info like phone and website may be different as well.


For the older style independent pharmacy, I do still use amenity=pharmacy
on the building outline:

http://www.openstreetmap.org/way/218176300

They do sell a few items like bandages without a prescription, but they
don't have the extensive personal hygiene section of typical of the
drugstore chains.  The pharmacist is also on duty for the entire time this
shop is open, so it doesn't feel like there are two separate entities
operating in the space.

Peter


On Tue, Jul 5, 2016 at 8:45 AM, Minh Nguyen 
wrote:

> Recently, iD was changed so that shop=chemist is labeled as "Drugstore"
> for American English users (and continues to be labeled "Chemist" for
> British English users). [1] An American mapping a Walgreens, CVS, or Rite
> Aid who searches for "drugs" will see the following choices, in order:
>
> * Drugstore (shop=chemist, marked with a shopping cart icon)
> * Pharmacy (amenity=pharmacy, marked with a pill bottle)
>
> Meanwhile, searching for "pharmacy" -- a synonym of "drugstore" in
> American English -- produces only the amenity=pharmacy preset.
>
> The rationale is that amenity=pharmacy should be used only for pharmacy
> counters (which can be found at both drugstores and inside supermarkets),
> while shop=chemist should be used for full-service drugstores that *may*
> contain pharmacy counters. Currently, this is at odds with the wiki and
> longstanding practice, which stipulates that a shop=chemist *may not* fill
> prescriptions.
>
> This change to iD came about due to a discussion in the Name Suggestion
> Index project, which is the component in iD that suggests tags when you
> fill in a commonly used name. [2] I happened to notice the change because
> it caused Transifex to prompt me to update iD's Vietnamese localization. To
> my knowledge, there has been no discussion on the mailing lists or formal
> proposal on the wiki, though the iD maintainer intends to edit the wiki to
> match iD's interpretation. iD is the only software that has made this
> change.
>
> On the one hand, I've come around to liking the proposal, because it makes
> it easier for data consumers to distinguish between pharmacy counters and
> full-fledged drugstores. On the other hand, I think it's problematic
> because an American mapping a Walgreens or CVS could potentially tag a
> "drugstore" and be unaware that they'd need to separately map the pharmacy
> counter in order to indicate that prescriptions may be filled on-site.
>
> Currently, amenity=pharmacy is far and away more common than shop=chemist
> in the U.S. as a way to tag drugstores. Certainly anyone retagging
> amenity=pharmacy to shop=chemist would be careful to add an additional
> amenity=pharmacy POI where the pharmacy counter would be. (For a typical
> Walgreens or CVS, it'd be next to the drive-through canopy.) However, I
> have little faith that the average iD user would know to do the same, since
> the word "drugstore", like "pharmacy", implies the sale of prescription
> drugs.
>
> I've hashed out many of these points in [3], but I think the discussion
> needs to involve the wider OSM community now. There's little chance of data
> consumers and other editors updating their logic if the change is only
> discussed in the iD project.
>
> [1] https://github.com/openstreetmap/iD/issues/3201
> [2] https://github.com/osmlab/name-suggestion-index/issues/30
> [3] https://github.com/openstreetmap/iD/issues/3213
>
> --
> m...@nguyen.cincinnati.oh.us
>
>
> ___
> 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] Best practices for dealing with old TIGER tags?

[Peter Dobratz]

JOSM does automatically discard some TIGER:* tags.  There's a list of keys
in the tags.discardable JOSM preference item.  Among the list of tags that
JOSM automatically deletes are:
tiger:source
tiger:separated
tiger:tlid
tiger:upload_uuid

These tags are hidden from the editor so you don't normally see them.  They
are automatically removed from any objects that you modify.  You may notice
them if you are looking at the history of an object in JOSM.

That being said, I delete the other tiger:* from roads as I am editing
them.  Usually, I am verifying addr:* tags of things along the road and
checking that the addr:street matches the name of the road.  I also often
remove the name_1, name_2, etc tags that came from the TIGER import.  Where
appropriate, I retain them in alt_name or old_name.

Peter

On Fri, Jun 3, 2016 at 8:51 AM, Steve Friedl  wrote:

> Ø  Unless something changed, I think both Potlatch and JOSM will remove
> the ‘junk’ tags from TIGER if you delete the reviewed=no
>
>
>
> I’ve deleted thousands of tiger:reviewed tags (after proper review) and
> have never seen JOSM take anything else along for the ride.  JOSM **does**
> remove the yellow glow around ways once you remove tiger:reviewed, but
> that’s all I’ve seen.
>
>
>
> I have very much wanted to dump the tags that have no obvious use for OSM,
> but had no idea if somebody else, somewhere, might use them: probably not,
> but it didn’t feel like it was my call to make. I’d love for there to be a
> consensus on this.
>
>
>
> So the only things I’ve removed are tiger:reviewed, plus spurious
> additional tags that duplicate existing ones (tiger:zip_left_1 when it’s
> the same as tiger:zip_left).
>
>
>
> Steve
>
>
>
> *From:* Russell Deffner [mailto:russdeff...@gmail.com]
> *Sent:* Friday, June 03, 2016 8:45 AM
> *To:* 'Adam Franco' ; talk-us@openstreetmap.org
>
> *Subject:* Re: [Talk-us] Best practices for dealing with old TIGER tags?
>
>
>
> Oops, sorry Adam, replied directly to you versus the list; here’s the
> message:
>
>
>
> My thoughts:
>
>
>
> Unless something changed, I think both Potlatch and JOSM will remove the
> ‘junk’ tags from TIGER if you delete the reviewed=no. Maybe this is not the
> case with iD?
>
>
>
> As far as classification; please note that it is not about whether the
> road is rural or not; it’s the function – there have been people who
> started changing all ‘dirt roads’ to track around me in rural Colorado –
> this is NOT correct. Most of the ‘dirt roads’ around here are 100%
> verifiably “residential”. So please don’t encourage mass changing of
> classification based on anything but function of the roadway.
>
>
>
> =Russ
>
>
>
> *From:* Adam Franco [mailto:adamfra...@gmail.com ]
> *Sent:* Friday, June 03, 2016 9:28 AM
> *To:* talk-us@openstreetmap.org Openstreetmap
> *Subject:* Re: [Talk-us] Best practices for dealing with old TIGER tags?
>
>
>
> Just some more feedback on the idea of a TIGER rural-residential challenge
> based on cleanup I've done throughout much of Vermont:
>
>- Most of the roads in rural areas should have their highway= changed
>to something other than residential. (well known issue).
>- Surface tags would be GREAT! I've added surface tags to most roads
>in Vermont, but have not quite gotten to all of them yet.
>- At least here in Vermont, "private road" means that the ownership
>and maintenance of the road is the responsibility of the resident[s], not
>that "access=private". We have many private roads due to low densities of
>residences and Towns generally won't take over ownership/maintenance unless
>there are at least 3 residences and the proposal passes a public vote. The
>TIGER import mistakenly tagged many private-roads as "access=private". It
>would be great to remove this tag if it hasn't been added by a person.
>
> If there is any way to help out with this effort I'd love to lend a hand.
>
>
>
> On Fri, Jun 3, 2016 at 10:18 AM, James Umbanhowar 
> wrote:
>
> Funny, I just looked at the MapRoulette beta and noticed that you were
> already doing this.
>
>
>
>
> On Fri, 2016-06-03 at 10:00 -0400, James Umbanhowar wrote:
> > Minor suggestion for this MapRoulette challenge:  Could you structure
> > it by state (or other geographic region, county?) and do each region
> > sequentially.  I, personally, think it would be neat to see areas get
> > "done" as far as Tiger clean up.
> >
> > Either way, thanks for these.
> >
> > James
> >
> > On Fri, 2016-06-03 at 10:21 +0200, Martijn van Exel wrote:
> > >
> > > Well said. I have space in my basement also.
> > >
> > > I am eager to launch a MapRoulette challenge for untouched rural
> > > ‘residential’ roads - a challenge which will probably take some
> > > time
> > > to complete. If someone can furnish a good Overpass query for this,
> > > please go ahead and do it.
> > >
> > > Martijn
> > >
> > > >
> > > > On Jun 3, 2016, at 8:55 AM, Richard Fairhurst  > > > et
> > > > >
> > > > > wrote:
> > > > There is a 

[Talk-us] IKEA vs Ikea?

[Peter Dobratz]

When putting locations of the Swedish furniture store in OSM, do you use

name=IKEA

or

name=Ikea

?

Although we expand abbreviations on things like street names in OSM,
company names are typically left as abbreviations (for example, "St. Mary's
Bank" or "Einstein Bros. Bagels").

Often, store logos have the store name in all caps, but we capture the name
in OSM as mixed case (for example "Banana Republic").  However, I generally
do use all caps for store names which are pronounced as a string of letters
(for example "AT&T" or "The UPS Store").

According to http://en.wikipedia.org/wiki/IKEA , this is an acronym that is
pronounced as a word and not as a string of letters.  For OSM, I generally
prefer "Ikea" as we like to use mixed case as opposed to all caps.  That
being said, "IKEA" seems to be more prevalent in OSM at the moment.

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

Re: [Talk-us] 'Honorary' street name conflicts with posted name - how to decide

[Peter Dobratz]

I like to confirm street names by checking for POIs along the street.
There's a Great Clips hair salon near there with the following address (see
http://www.greatclips.com/ ):

906 E 900 S
Salt Lake City, UT 84105

I would put the following addr:* tags on that business:
addr:housenumber=906
addr:street=East 900 South
addr:city=Salt Lake City
addr:state=UT
addr:postcode=84105

I may not fully understand the nuances of Salt Lake City addresses, but
generally I like to make whatever is in addr:street match what goes in name
on the road itself.  In this case, I would use:
name=East 900 South
alt_name=Harvey Milk Boulevard

Looking a little to the west of there, there's a Pier 1 Imports (
http://www.pier1.com/ ):

30 West 900 South Salt Lake City, UT 84101-2930

addr:housenumber=30
addr:street=West 900 South
addr:city=Salt Lake City
addr:state=UT
addr:postcode=84101

By that point the road should change to:
name=West 900 South
alt_name=Harvey Milk Boulevard

I generally avoid using the "_1" suffix whenever possible.

Peter


On Mon, May 23, 2016 at 10:12 AM, Jack Burke  wrote:

> I concur. You could also put Harvey as an alt_name tag.
>
> --
> Typos courtesy of fancy auto spell technology
>
> On May 23, 2016 1:03:15 PM EDT, Martijn van Exel  wrote:
>
>> Salt Lake City just renamed a part of 900 South to ‘Harvey Milk
>> Boulevard’. This is a so-called ‘honorary designation’. But now I see a
>> conflict.
>>
>> 1) The common tagging practice is that posted names rule. The signs
>> changed to show the new honorary name:
>> https://goo.gl/photos/xqAxQCwCmPRUGWkm9 . So that would suggest I change
>> the name to ‘Harvey Milk Boulevard’ and demote ‘900 South’ to name_1 or
>> loc_name.
>> 2) Addressing and geocoding. People will continue to refer to this street
>> as 900 South. Official addresses will not change. That would suggest that I
>> leave the name=900 South in place and add name_1=Harvey Milk Boulevard.
>>
>> I went with option 2, here is a segment —>
>> https://www.openstreetmap.org/way/418190301
>>
>> Opinions?
>>
>> Martijn
>>
>> --
>>
>> 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
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Restoring bus routes in Portland

[Peter Dobratz]

Hi Greg,

I've fixed all of the relations that had their contents replaced wholesale
by restoring to the last known good version and then adjusting as necessary
to match the current data.  The one change to my method was to download all
of the relation members after I load the prior version from a .osm XML
file.  That way if any of the relation members have since been deleted they
will be filtered out by JOSM when resolving the conflict with the existing
relation.

I think the rest of the bus route relations in this area are still intact,
but have gaps and/or incorrectly ordered members here and there.  I can fix
these fairly easily though.

Thanks,
Peter



On Sat, May 21, 2016 at 9:43 PM, Greg Morgan  wrote:

> On Fri, May 20, 2016 at 9:24 AM, Peter Dobratz  wrote:
> > Luckily, I saved a link to all of these bus routes on a wiki page as it's
> > impossible to download a relation with zero members (unless you happen to
> > know the ID):
> >
> http://wiki.openstreetmap.org/wiki/Portland,_Oregon/Transit_Route_Relations
>
> Please let me know if you need additional help with the relationships.
> It is unclear to me if this issue has been resolved or not.
>
> You have done the hard work.  You have a handy list of relations.
> I've been thinking of creating such lists for other highways and what
> not.
> If you don't want to hack and whack on an xml file, then you can also
> do something like this.
> 1. ) Select the relation number that you are interested in and copy it
> from your wiki page, control c.
>  The only downside to your wiki page is that you used a name and
> did not directly expose the relation number.
> 2..) In Josm menu bar>File>Down object...
> 3.) The dialog will show with the relation number already filled in.
> 4.) Pick any of the check box options in the dialog. You will have to
> experiment depending on what you are mapping at the time.
>  a.) Since you need to work with different layers pick
> the"Separate layer" check box.
>  b.) Uncheck the other two check boxes:  "Download referrers
> (parent relations)" and "Download relation members".
> 5.) Next click the Download Object button.
> 6.) Another layer can be used to download the current state with the
> steps above.
>  Inside the relation editor is a "Download incomplete members"
> button.  Use this to download all the ways or nodes in a relationship.
>  Of course this is great for working on extending and existing
> relationship.   I don't know if it will help you repair these
> relationships.
> I thought that I'd throw this out there for others.
>
> These are the rules that I use when I am extending a relationship.
> 1.) You have to be careful because you cannot delete. The way or way
> nodes may be connected to other parts of the map.
> 2.) In some node moves, you will have to download more data before
> moving nodes.  That way you see other ways attached to the nodes.
> 3.) Deleting nodes also requires more data.  You have to see the whole
> context with additional downloads of data.
>
> 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] Restoring bus routes in Portland

[Peter Dobratz]

I just found a decent workflow that I can use to restore previous versions
of these route relations.

https://help.openstreetmap.org/questions/4019/how-do-i-restore-data-that-was-lost-during-editing

1. Use the API to download the last good version of the route relation,
save as a .osm file.
2. Add action="modify" to the relation element in the OSM XML file
3. Load OSM XML file into JOSM
4. Download the relation by ID to get the latest version.
5. Handle conflicts by discarding the current relation and using the
version that I loaded from the file.
6. Use the JOSM relation editor to download the members
7. Use the JOSM relation editor to "find the next gap"
8. Download current OSM data for the area with the gap and add appropriate
members to the relation

Luckily, I saved a link to all of these bus routes on a wiki page as it's
impossible to download a relation with zero members (unless you happen to
know the ID):
http://wiki.openstreetmap.org/wiki/Portland,_Oregon/Transit_Route_Relations

Peter

On Fri, May 20, 2016 at 6:06 AM, Andy Townsend  wrote:

> On 20/05/2016 13:20, Arun Ganesh wrote:
>
>>
>> Reverting the changesets are causing multiple conflicts and there does
>> not seem to be a mechanism to just restore relation membership to a
>> historical version. Any help and ideas appreciated
>>
>> [1] https://github.com/mapbox/mapping/issues/185
>>
>
> If it helps, what I've done when things like this have happened in the
> past is:
>
> 1) Edit failing OSC files to not include relation changes that cause
> problems - i.e. get all the ways and nodes in OSM how you want them.
>
> 2) Find the "desired version" of the relation that you want to update, and
> try and apply that (e.g. via level0).  It'll fail, so find out which ways
> don't exist, take them out of the relation details you're trying to upload
> and try again.
>
> 3) Obviously now you'll have gaps in the relation - either fill them in
> manually if it's obvious, or figure out where the deleted ways that used to
> be in the relation were (by looking at undeleted nodes and the history of
> deleted nodes), and use that information to fill in gaps.
>
> As to "what caused it", there were examples of "relation contents being
> lost in JOSM uploads" a while back, but I've not heard reports of recent
> problems**, though there were reports of API upload slowness a few days ago
> (much discussed elsewhere - and in fact the message at the top of #osm
> still says "Ongoing server issues, sysadmins are on the case").  With any
> change in OSM it makes sense to double check the changes that you've made
> immediately after you've made them - did you really mean to make all
> changes in the resulting changeset?  Are all relations intact? etc.
>
> Cheers,
>
> Andy
>
> PS: and FWIW I guessed that it was the Portland with the homelessness
> problem* and the nice river running through it, not the one with a crab
> fixation or the one that's scared of "underground mutton" :)
>
> * when I last visited, which was a while ago.  They might have got their
> act together now.
>
> ** The DWG often gets cries for help when big relations disappear
>
>
>
> ___
> 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] OSM attribution on website store locator pages

[Peter Dobratz]

I heard back from Mobify.  It seems that they think because they have
purchased an "enterprise subscription" to MapBox, they are exempt from
needing to include OSM attribution on their maps.

My message to Mobify:

> I was looking at the following webpage with your logo in the lower left
> corner of the map: http://locations.pieology.com/ The map tiles you are
> using are derived from OpenStreetMap data. This data is created by
> thousands of worldwide volunteers such as myself. Please update the
> Pieology Store Locator page to include OpenStreetMap attribution as
> described here: http://www.openstreetmap.org/copyright The map tiles you
> are using are generated by MapBox using OpenStreetMap data. The following
> page lists some technical tips for adding proper attribution:
> https://www.mapbox.com/help/attribution/


Melissa Cheng @ Mobify's response:

> Hi Peter,
> Thanks for reaching out and thanks for the work you do!  We love our maps.
> Re: your message below, I wanted to let you know that we have an
> enterprise subscription to mapbox that allows us to whitelabel the maps.
>
> Have a great weekend,
> Melissa



 Peter

On Thu, Apr 21, 2016 at 10:52 PM, Peter Dobratz  wrote:

> I filled out the contact form on the Mobify website as they are the ones
> that are being attributed without also crediting OSM.  I'll let you know if
> I hear anything back.
>
>
> Meanwhile, Cricket Wireless is running a store locator page, which is also
> using the MapBox tiles derived from OSM data, but without attribution:
> https://www.cricketwireless.com/stores
>
> Cricket has an elaborate page about submitting copyright claims:
>
> https://www.cricketwireless.com/legal-info/cricket-wireless-copyright-trademark-claims.html
>
> As a first step, I tried sending an email to copyri...@cricketwireless.com.
> However, that email bounced back with a "#550 5.1.1
> RESOLVER.ADR.RecipNotFound; not found ##"
> I'm not sure about the best way to proceed from here.
>
> Peter
>
> On Thu, Apr 14, 2016 at 9:22 PM, Greg Morgan 
> wrote:
>
>>
>>
>> On Thu, Apr 14, 2016 at 9:39 AM, Matt Simpson  wrote:
>>
>>> All - seaming to understand the intent of this thread and hoping to shed
>>> some light for you.
>>>
>>> Our team (MAGNETIC) was hired to design/develop the pieology website.
>>> Pieology also hired a company to build their store locator, which was
>>> outside our scope of work. The company they hired for store location
>>> services is Momentfeed (https://momentfeed.com/), seemingly a branch or
>>> distributor of Mobify, who looks to be using your service at the core of
>>> the location tool they provide.
>>>
>>> I would suggest contacting Momentfeed and/or Mobify to discuss
>>> attribution. As someone mentioned earlier, Pieology will have very little
>>> input on the matter.
>>>
>>> Hope this helps -
>>>
>>>
>> Matt,
>>
>> Thank you for the response.  www.OpenStreetMap.org is a crowd sourced
>> map of the world. I love it that the Pieology site is using OSM.  There are
>> a number of other notable sites that are using OSM such as Craig's list
>> http://phoenix.craigslist.org/search/fua .  We are just trying to get to
>> the bottom of which firm should put the correct attribution on the Pieology
>> site.  I provided the CL map because it has the proper attribution.
>>
>> Again thank you for the response.
>>
>> Regards,
>> Greg
>>
>>
>>
>>
>>> On Thu, Apr 14, 2016 at 6:40 AM, Greg Morgan 
>>> wrote:
>>>
>>>>
>>>>
>>>> On Wed, Apr 13, 2016 at 10:24 AM, Martijn van Exel  wrote:
>>>>
>>>>> Interesting, a bit troubling, to see an increasing number of layers —
>>>>> abstraction — between the end user and the OSM data. Now it’s three — the
>>>>> web site itself, Mobify (never heard of them), Mapbox —  between the end
>>>>> user and OSM. Makes it easier for all the intermediaries to point to the
>>>>> others for attribution requirements. And harder for the OSM community to
>>>>> enforce this simple but important requirement. OSM can only benefit from
>>>>> having many eyes on the map if we can continue to close this feedback 
>>>>> loop.
>>>>>
>>>>>
>>>>
>>>>
>>>> There is one more layer of abstraction: the design firm
>>>> http://www.magneticcreative.com/.  It seems that Magnetic Creative
>>>> would understand this issue.  They hav

Re: [Talk-us] OSM attribution on website store locator pages

[Peter Dobratz]

I filled out the contact form on the Mobify website as they are the ones
that are being attributed without also crediting OSM.  I'll let you know if
I hear anything back.


Meanwhile, Cricket Wireless is running a store locator page, which is also
using the MapBox tiles derived from OSM data, but without attribution:
https://www.cricketwireless.com/stores

Cricket has an elaborate page about submitting copyright claims:
https://www.cricketwireless.com/legal-info/cricket-wireless-copyright-trademark-claims.html

As a first step, I tried sending an email to copyri...@cricketwireless.com.
However, that email bounced back with a "#550 5.1.1
RESOLVER.ADR.RecipNotFound; not found ##"
I'm not sure about the best way to proceed from here.

Peter

On Thu, Apr 14, 2016 at 9:22 PM, Greg Morgan  wrote:

>
>
> On Thu, Apr 14, 2016 at 9:39 AM, Matt Simpson  wrote:
>
>> All - seaming to understand the intent of this thread and hoping to shed
>> some light for you.
>>
>> Our team (MAGNETIC) was hired to design/develop the pieology website.
>> Pieology also hired a company to build their store locator, which was
>> outside our scope of work. The company they hired for store location
>> services is Momentfeed (https://momentfeed.com/), seemingly a branch or
>> distributor of Mobify, who looks to be using your service at the core of
>> the location tool they provide.
>>
>> I would suggest contacting Momentfeed and/or Mobify to discuss
>> attribution. As someone mentioned earlier, Pieology will have very little
>> input on the matter.
>>
>> Hope this helps -
>>
>>
> Matt,
>
> Thank you for the response.  www.OpenStreetMap.org is a crowd sourced map
> of the world. I love it that the Pieology site is using OSM.  There are a
> number of other notable sites that are using OSM such as Craig's list
> http://phoenix.craigslist.org/search/fua .  We are just trying to get to
> the bottom of which firm should put the correct attribution on the Pieology
> site.  I provided the CL map because it has the proper attribution.
>
> Again thank you for the response.
>
> Regards,
> Greg
>
>
>
>
>> On Thu, Apr 14, 2016 at 6:40 AM, Greg Morgan 
>> wrote:
>>
>>>
>>>
>>> On Wed, Apr 13, 2016 at 10:24 AM, Martijn van Exel  wrote:
>>>
>>>> Interesting, a bit troubling, to see an increasing number of layers —
>>>> abstraction — between the end user and the OSM data. Now it’s three — the
>>>> web site itself, Mobify (never heard of them), Mapbox —  between the end
>>>> user and OSM. Makes it easier for all the intermediaries to point to the
>>>> others for attribution requirements. And harder for the OSM community to
>>>> enforce this simple but important requirement. OSM can only benefit from
>>>> having many eyes on the map if we can continue to close this feedback loop.
>>>>
>>>>
>>>
>>>
>>> There is one more layer of abstraction: the design firm
>>> http://www.magneticcreative.com/.  It seems that Magnetic Creative
>>> would understand this issue.  They have their name in fine print at the
>>> bottom of each page on the site.  Just as they want their name published
>>> for potential clients, OpenStreetMap would like attribution for potential
>>> future mappers.  I added their email to the list.  They would have to join
>>> the list to respond, etc.  However, they should be able to pick up the
>>> drift from this email chain.
>>>
>>> Regards,
>>> Greg
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>> To my mind it’s on Mapbox to enforce attribution, I assume having this
>>>> present on the map is in the terms they present to their customers?
>>>>
>>>> Martijn
>>>>
>>>>
>>>> On Apr 13, 2016, at 7:31 AM, Fred Hillhouse 
>>>> wrote:
>>>>
>>>>
>>>> Their tiles are coming from Mapbox. Who gets the attribution?
>>>>
>>>>
>>>> http://b.tiles.mapbox.com/v4/dondeinc.i1021nhc/17/37433/50038.jpg70?access_token=pk.eyJ1IjoiZG9uZGVpbmMiLCJhIjoiZ200QzN4dyJ9.KIhMqSmPUSn9Kru51GZT4g
>>>>
>>>> Fred
>>>>
>>>> --
>>>> *From:* Harald Kliems [mailto:kli...@gmail.com ]
>>>> *Sent:* Wednesday, April 13, 2016 8:08 AM
>>>> *To:* Hans De Kryger; Peter Dobratz
>>>> *Cc:* talk-us@openstreetmap.org
>>>> *Subject:* Re: [Talk-us] OSM attribution on website store locator page

[Talk-us] OSM attribution on website store locator pages

[Peter Dobratz]

I'm seeing OSM data used more and more for generating the basemap onto
which things like store location data is displayed on store websites.
However, it's not always easy to find links to OSM attribution on such maps.

Has anyone seen this?

http://locations.pieology.com/

There's a link to http://www.mobify.com/ in the lower-left corner, but I
can't seem to find any links to http://www.openstreetmap.org/ anywhere on
the page.

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

Re: [Talk-us] Relations and boundaries

[Peter Dobratz]

On Thu, Mar 3, 2016 at 11:02 AM, Steve Friedl  wrote:

> Hi all,
>
> I’ve been updating all the cities in Orange County California to have fully
> segmented relationalized boundaries, such that cities sharing a common
> border share a single way in each of their relations; this eliminates
> overlapping ways.  It’s been very tedious but it's really getting cleaned
> up.
>
> So a few questions:
>
> First: The individual relations – city, county, national forest, etc. – all
> have full information tags about the entity, but how should the way members
> themselves be tagged?
>

> I’ve seen some OSM notes that say the ways can be tagged too, but I’m not
> sure how. There’s no common name, and there may not even be the same admin
> level (a city boundary on the border of a county would have two admin
> levels
> for the given way.
>
> My experience is that if the relation is fully tagged, and one of the ways
> is tagged with the same info, we see duplicate city names along the border,
> as the renderer takes the name from both the relation and the way.
>
> I am not sure I see any value *other than* tagging it as a boundary, with
> no
> other information. But I’d really like to do this right.
>


The Ways can actually be without tags as the information is fully described
in the Relations.  Depending on the situation, it may make sense to utilize
existing map features such as roads or rivers in the administrative
boundary relations.


> Second: I’m not sure how to handle quasi-enclaves.  Orange county is made
> of
> many cities, and a few cities contain some small *county* regions - think
> of
> them as donut holes. I don't know how to handle this.
>
> The Orange County relation is tagged with
>
> boundary = administrative
> border_type = county
> admin_level = 6
>
> The city of Westminster relation (fully inside Orange County) is tagged
> with
>
> boundary = administrative
> border_type = city
> admin_level = 8
>
> Within Westminster is a "donut hole" , and the Westminster relation has it
> as a role=inner.
>
> Question: should that same donut hole be tagged role=outer in the Orange
> County relation?
>
> It just doesn't feel right to have a role=outer fully within another
> role=outer, but that's the only way I can think of to handle this.
>
> Or do we just assume that because this donut hole has been excluded from
> the
> City of Westminster, it's automatically part of the next outer item (in
> this
> case, Orange County). The renderer doesn't identify any parts of the
> donut-hole boundary.
>
> The hole in question:
> http://www.openstreetmap.org/#map=17/33.73488/-117.98422&layers=N  it's
> just
> to the west of Star View Elementary School  There are many of these in
> Orange County.
>

I think you are misunderstanding the administrative boundary situation.
All areas within the city limits of Westminster are also contained in
Orange County.  In other words, when you enter Westminster, you are not
leaving Orange County.  Similarly, all areas of within Westminster are also
contained within the state of California.  You are not leaving California
when you enter Orange County and you are also not leaving California when
you enter Westminster.

When you leave Westminster, you are entering an area that is in Orange
County.  If that area is not contained within another city or town, then it
is said to be in an unincorporated area of Orange County.  So you would be
traveling from an incorporated area of Orange County to an unincorporated
area of Orange County, but all still within Orange County.

The holes in the Westminster boundary do not pertain to the Orange County
boundary and they shouldn't be part of the Orange County relation.  Does
that make sense?

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

Re: [Talk-us] How to tag Edible Arrangements?

[Peter Dobratz]

In the past, I've used shop=florist for the Edible Arrangements shops that
I have mapped.  I guess I was drawn to the word "Arrangements" in their
business name.  They sell fruit cut and put on sticks to look like bouquets
of flowers.  I just got a yellow pages the other day, and there is an
Edible Arrangements shop listed under florists, along with the flower shops
that you would typically think of when you think of florists.

However, now that I think about it, shop=confectionery may be a better
fit.  They sell plain fruit pieces and also fruit pieces dipped in
chocolate.  Chocolate or not, their wares are sweet and meant to be eaten.

As with a lot of the amenity/shop/craft/office tags, there is often overlap
and you have to use your best judgement to try to pick the best one.

Peter

On Wed, Mar 2, 2016 at 3:15 AM, Tod Fitch  wrote:

> There is an Edible Arrangements [1] shop near by that I tagged with only
> name and and the phone number found on the door. I could not figure out
> what type of shop it ought to be tagged with. Now I am cleaning up issues
> in my area noted by some of the QA tools and that has been flagged as
> “missing object kind”. So it is time to figure out an appropriate shop
> value.
>
> Nothing in the Wiki of course, so I looked at taginfo [2] to see what
> others have done and it seems that others are as confused as I: Of the 81
> shops mapped there are 13 different shop values and 13 with no shop or
> amenity tag at all. Values are florist (21 instances), gift (12 instances),
> yes (12 instances), confectionery (10 instances), bakery (1 amenity and 2
> shops), candy (2 instances), boutique (2 instances), food (2 instances),
> catering (1 instance), edible (1 instance), deli (1 instance),
> confectionary (1 instance) and greengrocer (1 instance).
>
> Suggestions on how I ought to tag my local shop (I probably will not
> “correct” the shops tagged by others)?
>
> Thanks!
>
> [1] https://www.ediblearrangements.com/default.aspx
> [2] http://taginfo.openstreetmap.org/tags/name=Edible%20Arrangements
> ___
> 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] Android Phone App that Uses Garmin Format -- Was Whole-US Garmin Map update - 2015-09-30

[Peter Dobratz]

No, but if you just want offline GPS navigation based on OSM data, get
Maps.me

Peter

On Tue, Oct 6, 2015 at 2:44 PM, Greg Morgan  wrote:

> On Fri, Oct 2, 2015 at 1:03 PM, Dave Hansen  wrote:
> > These are based off of Lambertus's work here:
> >
> > http://garmin.openstreetmap.nl
> >
> > If you have questions or comments about these maps, please feel
> > free to ask.  However, please do not send me private mail.  The
> > odds are, someone else will have the same questions, and by
> > asking on the talk-us@ list, others can benefit.
> >
> > Downloads:
> >
> > http://daveh.dev.openstreetmap.org/garmin/Lambertus/2015-09-30
> >
> > Map to visualize what each file contains:
> >
> >
> http://daveh.dev.openstreetmap.org/garmin/Lambertus/2015-09-30/kml/kml.html
>
>
> I think my Garmin GPS finally died.  Has anyone had experience with or
> can recommend an Android app that uses the Garmin file format?  I seem
> to be in areas where the data roaming does not shine.  It would be
> nice to toss one of these US Garmin files on the SD card and still be
> able to navigate.  The primary concern is navigation so the other
> editing features and such are not as great of a concern.
>
> Thanks,
> 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] Best iOS app for GPS wander, GPX to laptop into OSM?

[Peter Dobratz]

Steve,

For most of my data collection, I find that the built in photos app is
actually all I use.  Each photo is geotagged, which comes in handy if I
forget exactly where I was when I took the photo.  I end up picking up
details when looking at the pictures later that I didn't originally see
(phone number in a shop window, fire hydrant in front of a shop, etc).

If I just want to capture a GPX file, then I use GPXTrack (costs $1.99):

http://bafford.com/software/gpstrack/

Once you've finished your track, then you can email the .gpx file to
yourself.

Another option for collecting GPS tracks is to use RunKeeper (free for
basic mode):

http://runkeeper.com/

This is a program designed to keep track of your workouts and you can
specify what mode of travel you are using before you start (run, walk,
bike, etc).  I believe the program requires you to sign up for a RunKeeper
user account.  Once you are done with your run, you can view your run on
the RunKeeper website and even download a .gpx file yourself.  I believe
they also put everyone's tracks together in aggregate to find holes in OSM
data.

Note: when running GPXTrack or RunKeeper and switching to other apps
(Photos, email, OSM website in web browser, etc), the phone OS can actually
suspend your GPX collecting app and it won't resume data collection until
you switch back to the app.  I've noticed this phenomenon much less when
moving from the iPhone 4S to iPhone 6 Plus.

I've also recently started using Mapillary (free):

http://www.mapillary.com/

Mapillary requires a user account as well.  You take a bunch of photos and
upload them to Mapillary's website for anyone to see.  It also connects the
dots and shows you the path you took from one photo to the next.  You can
either take individual shots or set it to auto mode and just point the
camera in front of you as you are walking around.  They do some
post-processing on images to attempt to blur things like license plates.
When you view the photos on the website, you can ask for portions to be
blurred or unblurred.  There's a JOSM plug-in for Mapillary photos and also
they have their own version of the iD editor that lets you see Mapillary
photos.

Maps.me is a very interesting app that I use a lot.  I love that I can
download my map data for my state and a few surrounding states to have a
wealth of map data directly on my phone.  It's a little bit frustrating for
OSM data contributors because the Maps.me data is 1 or 2 months old.  It's
great if you go to Canada and need a map on your phone and don't want to
pay exorbitant rates for mobile data.

Peter


On Mon, Oct 5, 2015 at 10:53 PM, stevea  wrote:

> Marc Gemis writes:
>
>> I've tried OsmAnd, OSMTracker and KeypadMapper 3.
>>
> (and more).
>
> Thank you, Marc!  I "prefer" (as it's familiar, though maybe old school) a
> dedicated GPS and a USB cable, too.  But then there ARE smartphones enabled
> with GPS, and they CAN and SHOULD do this sort of "capture and squirt" (GPX
> data at OSM) the way I'm discussing.  Just finding the right app (iOS and
> Android) to do it is all.  Wireless (Bluetooth?) would be very neat, and
> GPX seems like the right data format to act as a data format vehicle.  Yes,
> and not too darn many tap-tap-taps on that tiny screen!  The phone is the
> data capture (and squirt it) device, the laptop (and iD in a web browser)
> are the editing environment.  Let's connect these (wirelessly).
>
> SteveA
> California
>
>
> ___
> 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] Should driveways be on OSM?

[Peter Dobratz]

On Fri, Oct 2, 2015 at 7:23 AM, Tom Bloom  wrote:
>
> -Living Streets. This is a distinct entity and not something decided by OSM. 
> They are decided by local administration, and OSM should tag them only after 
> that. Paul suggests that there could be townhouses at the end of rural lanes 
> in rural Oregon. They are farmhouses and the tag is wrong. Tho I do prefer 
> the wrong green tag to the wrong red tag :-).
>
> Almost everywhere that I see people using highway=living_street in the
United States, I think they should be really using highway=service and
service=driveway as you suggest.  I did live in Germany for a while and I
know exactly what is meant by the highway=living_street tag there,
especially because they have a special road sign for them. I have never
come across a living street in the United States.

Access:
Living streets are a type of residential street which are generally public.
Driveways generally have some sort of implied access restrictions.

Ownership:
Living streets are generally owned and maintained by the government.
Driveways are generally owned and maintained by the owner of the adjoining
buildings.

Speed:
Living streets are generally really slow (2-3 mph) because you expect
children to run out at any time.
Driveways are generally faster (10-15 mph), though the speed limit, if any,
is usually set by the property owner.  You expect pedestrians to at least
pause and look before entering the driveway.

Crosswalks:
Living streets don't have crosswalks because the whole area sort of acts
like a crosswalk because vehicles have to always watch for people and yield
to them.
Driveways, especially in larger apartment complexes, can have designated
crosswalks and sidewalks for pedestrians.

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

Re: [Talk-us] North Carolina Outer Banks

[Peter Dobratz]

The OpenStreetMap database covers the whole world.  It's often too resource
intensive to do processing on the whole world's dataset at once, so often
third party applications will use extracts of data.  It sounds like you are
using an extract of data which doesn't fully cover the area you are
interested in.  Where are you getting your map datasets?

Peter

On Sat, Apr 4, 2015 at 1:49 PM, Jeff Klein  wrote:

> If I am correct, probably confused, the map sets only go as far east as
> -76 West, which excludes the Outer Banks of North Carolina. How do I get
> maps that go further east?
> ___
> 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] Facts about the world

[Peter Dobratz]

On Fri, Apr 3, 2015 at 12:11 PM, Frederik Ramm  wrote:

> Darrell,
>
> On 04/03/2015 08:39 PM, Darrell Fuhriman wrote:
> > Ignore the haters, we’re doing fine.
>
> I don't know if that thing about "haters" is just a generic figure of
> speech but if you should indeed believe that I have expressed hate about
> anything, then you are mistaken (and I would feel a bit offended by you
> branding me a "hater" - does that then make you one too?). If I have
> expressed negative feelings in my message then they were pain or
> sadness, not hate.
>

The word hater in modern American speak is often used to describe someone
who is simply expressing a generally cynical or pessimistic opinion.

There are many local events in the US that don't get published to this
talk-us email list, nor do they get added to the event calendar on the OSM
wiki.  All of the OSM in person meetings that I have been able to attend
have been positive experiences for me.

I am in favor of making use of publicly available geodata to enhance the
data that we collect on the ground.

Recently, I've been deliberately incorporating exercising into my OSM
efforts.  I often go running for at least a mile, then walk around and
capture some photos and notes on my phone, and then run home.  I think it's
great that if I use the RunKeeper application on my phone, then I can visit
runkeeper.com to see my route overlaid on top of a map that is generated
with OSM data.

I recently added some trails to OSM through the woods (some of these trails
were just built in summer 2014):
http://www.openstreetmap.org/#map=17/45.49406/-122.69378

Multi-modal routing based on OSM data is also used at our local public
transportation agency: http://ride.trimet.org/  I've seen improvements that
I have made to the road and trail network propagate to this site.

http://maps.me/ is a great mobile app, which utilizes OSM data to display
maps on the go without the need for an internet connection.  I love seeing
improvements that I have made to OSM data show up on my phone through this
app.  I appreciate not having to pay through the nose for mobile data usage
if I want to look at a map on my phone when I travel to Canada.

I also just like mapping shopping centers and viewing the rendered results
on the OSM website:
http://www.openstreetmap.org/#map=18/45.51448/-122.79099
http://www.openstreetmap.org/#map=19/45.53583/-122.86951
http://www.openstreetmap.org/#map=19/45.44305/-122.80189
http://www.openstreetmap.org/#map=17/42.89235/-71.32614

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

Re: [Talk-us] USA Rail: Calling all OSM railfans! (especially in California)

[Peter Dobratz]

Hi SteveA,

I see that you have summarized the a lot of the same information from your
email on the United States Railways wiki page:

http://wiki.openstreetmap.org/wiki/WikiProject_United_States_railways

Looking through Paul's comments and yours, I don't see any specific
information about exactly how one would go about identifying specific
railways in Oregon so that they could be added to relations.

For the railways, Paul may be objecting to the content of the name and ref
tag on the Way objects themselves for the railway.  However, it is not
clear how to find out what the name actually should be.  The wiki page does
indicate that the name tag on the Way objects should match the name tag on
the Relation object with type=route and route=railway tags.  For many rails
around Portland, these Relations (type=route, route=railway) have not yet
been created.

You mention 2 specific examples (type=route; route=railway): Brooklyn
Subdivision (http://www.osm.org/relation/2203588) and Fallbridge
Subdivision (http://www.osm.org/relation/1443651).  Some of the Way objects
in Fallbridge Subdivision are also contained in
http://www.osm.org/relation/4734792.  Both of the relations for Fallbridge
Subdivision have FIXME tags expressing uncertainty about exactly where the
route Relation should begin and end.  How would one determine the exact end
of the Relation for the Fallbridge Subdivision?

Also, looking through the history of the above relations, I can't really
find anything in the changeset tags regarding the source of the data about
the railroads.  Where do the names Brooklyn Subdivision and Fallbridge
Subdivision come from?

Paul mentions that we should be using the name "Banfield Mainline" but
where does that name come from and what exactly does it refer to?

Are there signs on the ground with these things?

Peter



On Wed, Mar 25, 2015 at 5:18 AM, stevea  wrote:

> Hello Peter:
>
> The California/Rail wiki page you describe documents a couple of different
> ways we tag rail.  OpenRailwayMap (ORM) documents a three tier
> (route=tracks, route=railway, route=train) method used in parts of
> Germany.  As that page (as well as the USA Rail WikiProject) explain(s),
> because of the way TIGER entered rail in the USA, (and the way we structure
> and name rail) we often use just two of these, skipping route=tracks
> relations and "jumping" right to putting "named rail" into relations of
> route=railway:  rail "infrastructure."  You might say that two
> ORM/German-style "lower and middle level" relations have been merged into a
> single "middle level" relation here in the USA.  There are also ("higher
> level," and the whole OSM world agrees) passenger rail relations:
> route=train (or route=light_rail, route=subway, route=tram...effectively at
> the same logical "level" as route=train).  That's OSM rail "structure" in a
> nutshell.
>
> In Oregon, there are the Brooklyn Subdivision (
> http://www.osm.org/relation/2203588), the Fallbridge Subdivision (
> http://www.osm.org/relation/1443651)... these are (correctly) the
> middle-level infrastructure relations tagged route=railway.  There are also
> (predictably, also, the higher-level) route=train passenger rail relations
> like Amtrak Cascades (http://www.osm.org/relation/71428) which are often
> made up of a group of Subdivisions (route=railway relations) like Brooklyn
> and parts of Fallbridge.
>
> THIS is what Paul was typing about in those Notes.  Specifically, a
> (higher-level/passenger) route=train relation should not have as its name=*
> tag the name of the system (like MAX, BART, Metro or Amtrak), it should be
> the name of the passenger line (Green Line, Downtown to University...).
> And, the "underlying" (lower-level infrastructure) route=railway relation
> should be correctly "named" as the rail company (or public works
> department, transit district...) names it: often something like XYZ
> Subdivision or ABC Industrial Line.
>
> OSM's Transport Layer is handy to display (rather raw) railway=* and (at
> closer zoom levels) route=bus.
> ORM is handy to display rail infrastructure (with Infrastructure radio
> button selected), especially usage=* tags.
> OpenPublicTransportMap (http://openptmap.org) is handy to display
> passenger rail relations.
>
> The USA is largely under construction for all of these, but we've come a
> long way.
>
> It's all in those wikis.  Makes sense?
>
> Regards,
> SteveA
> California
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] USA Rail: Calling all OSM railfans! (especially in California)

[Peter Dobratz]

On Sun, Mar 22, 2015 at 10:31 AM, Paul Johnson  wrote:

> Portland also needs help.  Seems whenever the map gets fixed, someone goes
> through and stomps the name back to something incorrect like "Metropolitan
> Area Express" or "Portland Streetcar" instead of the subdivision name, and
> pull things like putting the line of the service running on the tracks as
> ref=*.   God help you if you actually try to point it out, Grant Humphries
> or Peter Dobratz will get bitchy about it...
>

For those following along from home, please see the following note in
Portland, Oregon:

http://www.openstreetmap.org/note/335545

As you can see, I closed out this note (twice).  The issues in this note
are too vague and refers to the whole Portland metropolitan area.  Also,
the actual area where this note is located is a new section of track which
is not scheduled to be operational until September 2015.  There are a bunch
of things in this immediate area with construction or proposed tags since
the area is still very much in flux.

There was a similar situation which I resolved a few months ago.  Paul put
some OSM notes in downtown Portland saying he objects to the use of
oneway=yes tags on OSM Ways for Portland area light rail and street car
lines.  I looked into the history and determined that there were a few
cases where Grant Humphries had added the oneway=yes tag and Paul had came
back later and removed the oneway=yes tag, only to have Grant Humphries add
the oneway=yes tag in a subsequent edit.  I sent OSM private messages to
both Grant and Paul in December.  Paul never responded to my OSM private
message on this subject.  However, I did have a productive conversation
with Grant.  Grant was not even aware that he had been undoing some of
Paul's edits. In any case, we came to the agreement that oneway=yes does
not make sense for Portland area railways and I removed the tag as part of
my effort to update the route relations to use the new route_master format
with a separate route relation for each direction of travel.  For what it's
worth, there are actually signs on the ground that tell pedestrians to look
both ways before crossing these train tracks and the new route relations
implicitly include the standard direction of travel along the railway
because the rail segments are added to the relation in the other they are
traversed.

However, in the case of the above note, I can't discern exactly which tags
Paul is objecting to, nor can I find any specific information on the OSM
wiki about exactly what should be in the name tag on railways.  In the
note, Paul says "It's not rocket surgery to create the relations and have
things named like "Banfield Mainline" like it's supposed to be instead of
"Metropolitan Area Express", which is wrong."  I have no idea what Paul is
talking about here.  The phrase "Banfield Mainline" does not occur in the
OSM wiki, and I can't find anything on the internet to indicate exactly
what tracks would be best referred to as the "Bainfield Mainline."  I moved
to Portland about 9 months ago, I often hear these tracks colloquially
referred to as the "MAX", which is an acronym for "Metropolitan Area
Express."  Or maybe "MAX" just refers to the name of the trains that run on
those tracks.

I do not have any objections to updates to Portland area railways to be
more correct/complete and/or consistent with the work SteveA is doing in
California.  For anyone doing these edits, it would be helpful to check the
OSM history on the affected Ways and communicate with anyone who has also
changed the tags that you would be changing.  Maybe even start with an
Oregon equivalent of the http://wiki.osm.org/wiki/California/Railroads wiki
page.  A bunch of the route relations for the Portland area are already
linked from http://wiki.openstreetmap.org/wiki/Portland,_Oregon.

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

Re: [Talk-us] Best practices for outdoor mapping party

[Peter Dobratz]

I've given up the dedicated GPS and/or pen and paper for data collection
and I do everything using the camera and notes application on my smartphone.

All the pictures I take are geotagged.  I transfer the photos to a computer
via the USB port and then load them into JOSM with a marker showing where
they were taken.

There are also smartphone applications to create GPX files, but this seems
superfluous for most things I map since recent high-resolution aerial
photos are available from Bing.

Peter


On Mon, Mar 9, 2015 at 9:09 AM, Harald Kliems  wrote:

> With help from the wonderful folks at Maptime Madison, we're planning on
> hosting the first Madison (Wisc.) mapping party on the Spring Mapathon
> weekend. Nobody involved has ever organized or even attended a mapping
> party, so we wouldn't mind some advice. From reading on the wiki and
> various user diaries, I've come up with the following rough plan:
>
> - Meet at coffee shop, distribute Field Papers maps of the area to be
> surveyed, GPSrs , cameras, calibrate camera clocks. Mention non-obvious
> things that can be mapped, e.g. diet, payment method, collection times,
> opening hours, backrests on benches.
> - Depending on the number of participants, start surveying all together or
> in groups of three to four people. Plan on about one hour of surveying.
> - Group works it way toward the final meeting point at the local public
> library. Have a least two hours to process data and get it into OSM.
> Laptops are available at the library.
>
> Does this sound reasonable? Anything else I should be thinking of?
>
>  Harald.
>
> ___
> 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] Tagging addresses on area's

[Peter Dobratz]

On Thu, Feb 12, 2015 at 10:29 AM, Clifford Snow 
wrote:

>
> On Thu, Feb 12, 2015 at 9:35 AM, Peter Dobratz  wrote:
>
>> I agree with the sentiment in this wiki page that I quoted here.  I think
>> that if there is only 1 POI that occupies the building, then that is best
>> indicated by putting the tags for the POI on the building polygon.  If you
>> just have a Node for the POI, then you are not saying anything about the
>> size of the business, and you are not saying anything about whether there
>> are multiple businesses occupying the same building.
>>
>> Though there are often different entities that own the building and the
>> business in the building, I think for OSM we are much more concerned with
>> information about the business
>>
>> If the business closes or moves, then just delete any tags related to the
>> business (shop, phone, website, etc), but keep the building and addr tags.
>> When a new business comes in, add tags as appropriate for that business.
>>
>
> Generally I agree with adding the business name to the building polygon.
> But what do you do when a single business operates in a "named" building?
> My choice is to preserve the building name on the polygon and add the
> business as a node within the building polygon.
>

Creating the internal node for the business is an option in the case where
you want to also have a name tag on the building that is different from the
name.  Depending on the situation, it may also be more appropriate to put
the name on a land use polygon (landuse=retail or landuse=commercial) that
surrounds the building and grounds.  You could also have more than one name
tag on the building by making use of various name tags (for example
alt_name or old_name).  If there's a name etched in stone that's part of
the building denoting it's original use (for example a former elementary
school), then I capture that with old_name.

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

Re: [Talk-us] For comment: import of amenity=bicycle_repair_stations

[Peter Dobratz]

On Wed, Feb 11, 2015 at 2:55 PM, Bryce Nesbitt  wrote:

> I'd like to start with California and Arizona (70 nodes),
> and add notes where on-site survey is needed to exactly align to the OSM
> base layer.
>
> What's different about this import is that armchair mappers can get only
> so far, verifying press releases and such.  The features are too small to
> spot even on the best available air photos (e.g. better than the Bing
> ones).  Even the street cameras are not generally enough: this really takes
> in person spotting.
>
> Any more comment / objection?
>

The import wiki page for this (
http://wiki.openstreetmap.org/wiki/Import/Dero_Bike_Repair ) references a
python script.  Is this script available to look at?

I'm looking through the mailing lists and I can't find any sample .osm
files for data that you plan to import.  Can you send out a link?

Have you created a separate OSM user account dedicated for this import?

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

Re: [Talk-us] Tagging addresses on area's

[Peter Dobratz]

On Wed, Feb 11, 2015 at 6:04 PM, Bryce Nesbitt  wrote:

>
>> This seems to be "tagging for the renderer" which is not a good idea.  If
>> the only thing occupying the building is a single POI, then put the POI
>> tags on the closed Way for the building outline.  By adding a new object
>> (Node) for the POI, you are also going against the "One feature, one OSM
>> element" idea.  There are renderers in existence that will display both the
>> house number and the name of a restaurant with food icon.
>>
>
>
>> > http://wiki.openstreetmap.org/wiki/Tagging_for_the_renderer
>> > http://wiki.openstreetmap.org/wiki/One_feature,_one_OSM_element
>>
>> > http://wiki.openstreetmap.org/wiki/Good_practice
>>
>>
> There are two features here:  the building, and the business.
> The building and business may well have different owners.
> In some cases a business will sell it's building but lease back the space.
>
> The address is generally an attribute of the building and does not change
> as tenants come and go.
> The business inherits the address for the time they are in the building.
> They may have a website= that keeps working even if they move.
>
> ---
>
> Thus one could argue that the "one feature one osm element" rule would
> want:
> * Attach an address to the building outline.  Tag the owner if that is
> known (e.g. the leasing company).
> * Add POI nodes for one or more businesses inside that building.  These
> may have website/phone tags.
>

What you are suggesting is the opposite of what the "one feature one osm
element" page is suggesting.  In fact, what you are suggesting is the very
first thing in the list of things to avoid:

>From http://wiki.openstreetmap.org/wiki/One_feature,_one_OSM_element:
"Examples of bad situations:

   - An area object representing a single-use building with a point object
   inside it. Move the tags to the area object and delete the point."


I agree with the sentiment in this wiki page that I quoted here.  I think
that if there is only 1 POI that occupies the building, then that is best
indicated by putting the tags for the POI on the building polygon.  If you
just have a Node for the POI, then you are not saying anything about the
size of the business, and you are not saying anything about whether there
are multiple businesses occupying the same building.

Though there are often different entities that own the building and the
business in the building, I think for OSM we are much more concerned with
information about the business

If the business closes or moves, then just delete any tags related to the
business (shop, phone, website, etc), but keep the building and addr tags.
When a new business comes in, add tags as appropriate for that business.

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

Re: [Talk-us] Tagging addresses on area's

[Peter Dobratz]

On Thu, Feb 5, 2015 at 7:58 PM, Greg Morgan  wrote:

> On Thu, Feb 5, 2015 at 4:23 AM, Mike N  wrote:
>
...

> > The addr:housenumber doesn't render anyway if the
> > building has also been tagged as another POI that renders.
>
>
> I've solved this problem by putting the address on the building.  If
> the building also has a POI tag such as amenity=fuel, I move this tag
> to a new POI node.  I place the POI node so that both the
> addr:housenumber render along with the POI information.  Addresses and
> buildings don't change as much as how  busiensses can come and go.
>

This seems to be "tagging for the renderer" which is not a good idea.  If
the only thing occupying the building is a single POI, then put the POI
tags on the closed Way for the building outline.  By adding a new object
(Node) for the POI, you are also going against the "One feature, one OSM
element" idea.  There are renderers in existence that will display both the
house number and the name of a restaurant with food icon.

http://wiki.openstreetmap.org/wiki/Tagging_for_the_renderer
http://wiki.openstreetmap.org/wiki/One_feature,_one_OSM_element
http://wiki.openstreetmap.org/wiki/Good_practice

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

Re: [Talk-us] For comment: import of amenity=bicycle_repair_stations

[Peter Dobratz]

On Sun, Feb 8, 2015 at 7:58 PM, Greg Morgan  wrote:

> ...
>


> In a note tag I added a reference to your import page.  That would be
> useful for other mappers to know why and how.
>

Seems like this note would be better suited for the changeset comment
instead of the OSM Object.


>
> I added service:bicycle:charging=no based on the main page.  There's a
> few electric bicycle shops around downtown.
>
> I added a website tag and pointed to the Dero fix it map.
>

I think it's better to make the simplest URL possible here.  For this case,
use "http://www.dero.com/"; instead of "
http://www.dero.com/fixitmap/fixitmap.html";
Web developers change their mind about how they want their sites laid out
all the time.  Usually the domain name remains the same.

I added source_alt = survey.  I am guessing that source=osmsync:dero
> is part of you conflation process.  If so, then you might want to make
> that clear on you import page
> https://wiki.openstreetmap.org/wiki/Import/Dero_Bike_Repair


Information about the source of data makes a lot more sense as a source tag
on the changeset instead of the OSM Object.

I dutifully added ref= but did not understand the point.  Will
> this be part of the conflation magic?  If so, then you might want to
> make that clear on you import page.
>

I don't think there is any precedence for using the actual value ""
for a ref tag (or any other tag).  If you are trying to indicate that there
is no reference number associated with these, then just omit the ref tag.

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

Re: [Talk-us] Tagging outdoor US shopping centers

[Peter Dobratz]

On Tue, Dec 23, 2014 at 7:17 PM, Tod Fitch  wrote:

> I am more likely to use your option 1: Each shopping center a separate
> landuse=retail, name=* covering the entire area including parking but not
> the land associated with the roadway right of way. As an example there are
> two named shopping centers at this intersection:
> http://www.openstreetmap.org/#map=18/37.36843/-122.03593


Yes, I agree that you don't want to connect the landuse polygons to the
adjacent roads.  Two adjacent landuse polygons can share nodes if there's
nothing between them.  Generally, you have the Way objects for the roads in
the center of the road and the landuse for the shopping center starts once
you get off the road and onto the property for the shopping center.

landuse=retail is for primarily stores and restaurants and
landuse=commercial is primarily for things like doctors' offices.  Using
landuse=retail for the shopping center is a great place to also put the
name tag.  You can have smaller landuse polygons that are inside of larger
"broad brush" ones.  Many of these have websites which you can add using
the website tag as well.  I also use landuse=residential to put a name tag
on a named residential housing division.

I prefer to split up the larger buildings for each store rather than just
drawing a big building polygon and putting nodes in for the stores.  See
the buildings within the following example:
http://www.openstreetmap.org/way/178251857

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

Re: [Talk-us] Retail Example using OSM

[Peter Dobratz]

The company providing the service is Where 2 Get It (
http://www.where2getit.com/ ).  They provide maps for store locator
functions for a number of companies.  For example: Moe's Southwest Grill,
Safeway, Trader Joe's, Jo-Ann Fabric, Michael's.  OSM is used for the base
map with the store location placemarks as an overlay.  You can tell that
they are not utilizing the OSM data for the store locations because if you
look closely you can see that their placemarks don't always line up with
the actual locations of the stores in OSM.

--Peter


On Mon, Dec 22, 2014 at 1:17 AM, Hans De Kryger 
wrote:
>
> ​And it has proper credit for osm. Awesome!​
>
> *Regards,*
>
> *Hans*
>
> On Sun, Dec 21, 2014 at 6:09 PM, Michael Patrick 
> wrote:
>
>> Michaels craft stores
>> http://www.michaels.com/store-locator
>>
>> ___
>> 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
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Misspelled names

[Peter Dobratz]

On Wed, Dec 17, 2014 at 2:13 PM, Mike Henson  wrote:
>
> I am also having an issue with Walmart, should it be a Marketplace,
> Supermarket, Department store??? All 3 are used in OSM in Oklahoma.
>

Generally, Walmart is tagged as shop=department_store.

However, there are a few location in the Portland, Oregon area that are
zoned to only allow grocery type stores.  These are tagged:

name=Walmart Neighborhood Market
shop=supermarket

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

Re: [Talk-us] Coastal Maine needs love

[Peter Dobratz]

On Wed, Oct 15, 2014 at 9:33 AM, Eric Kidd 
wrote:

> Thank you for your response!
>
> I've made two test edits:
>
>- I imported TIGER County Subdivision files for four town boundaries.
>- I modified an existing TIGER Place outline to be admin_level=9,
>because it's a actually a subdivision of the real town. That's the brown
>blob on my example map.
>
> I think you are on the right track with the TIGER County Subdivision
files.  I did a lot of work on the town boundaries in New Hampshire and it
probably had the same sort of mess you are seeing in Maine.  Boundaries for
towns were imported multiple times by different users and the county
boundaries were imported using a much lower precision source.  At least in
New Hampshire, the county boundaries are most often coincident with town
boundaries, so it makes sense to have these in OSM as Relations (not as
closed Way polygons) and sharing the Way objects among multiple
boundaries.  Also, the TIGER CDP shapes were erroneously used in place of
the actual town boundaries.  The CDP for a town is often generally where
the most of the people live, but the actual boundary for the town is a much
larger area and this can be verified against town-line signs on the ground.

The TIGER boundary data seems to want to share nodes with TIGER road data
even though it doesn't actually make sense.  For example, town boundaries
are often straight lines, but in TIGER these lines are slightly jagged so
that they can share points with roads that are close to the town line.  If
you look at resources like property tax maps that some towns make
available, you can see that in many cases the TIGER boundary data should
just be made into straight lines.  And straight lines in OSM should just be
represented by a Way connecting 2 Nodes (your simplify step in QGIS will
get you most of the way there).

Tagging on the Ways is completely optional as that information should all
be in the admin Relations.  However, the consensus is that the Ways should
have the admin_level be the lowest number (for example admin_level=6 for
Ways that make up both a town and a county boundary) rather than trying to
have multiple values separated by semicolons.

Generally, I have used admin_level=9 for areas inside of towns that appear
to be separately administered.  In some larger towns, I used admin_level=9
for the wards or districts which correspond to seats on the local
government.

Towns have also been imported as single nodes, which could be influencing
your Nominatim results.  These place Nodes should be added to you admin
Relations with role "admin_centre" or "label".  You should also add
wikipedia tags to the boundary relations as this will help Nominatim
determine the place hierarchy.

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

Re: [Talk-us] changeset 25081346 spanning contiguous United States

[Peter Dobratz]

On Mon, Oct 6, 2014 at 9:43 AM, Eric H. Christensen <
e...@christensenplace.us> wrote:

> On Mon, Oct 06, 2014 at 09:33:08AM -0700, Peter Dobratz wrote:
> > After checking a few of the objects, it looks like this change removed
> > "shop=copyshop" and added "amenity=post_office".  I don't necessarily
> > disagree with this change.  In my mind I had been reserving post_office
> for
> > entities controlled by the government-run United States Postal Service,
> but
> > after reading the wiki I see that private companies can be also
> designated
> > as post_office, and the UPS Store certainly fulfills many of the same
> > functions as government-run post offices.
>
> Isn't there a way to specify the provider of the service (outside the
> name?)?  I've run into this while marking post_box for UPS and FedEx drop
> points that I run across.
>

For "amenity=post_box", I add an operator tag.  The official blue mailboxes
get "United States Postal Service" and I also tag other common providers
"United Parcel Service" and "FedEx".  These same values for the operator
tag could be added to post offices and UPS and FedEx retail store locations.
___
Talk-us mailing list
Talk-us@openstreetmap.org
https://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] changeset 25081346 spanning contiguous United States

[Peter Dobratz]

Hi,

I noticed the following changeset which touches UPS Store objects across
the whole contiguous United States:

http://www.openstreetmap.org/changeset/25081346

I have contacted user andrewpmk inquiring about communications prior to
making this change and I have not received a response.

After checking a few of the objects, it looks like this change removed
"shop=copyshop" and added "amenity=post_office".  I don't necessarily
disagree with this change.  In my mind I had been reserving post_office for
entities controlled by the government-run United States Postal Service, but
after reading the wiki I see that private companies can be also designated
as post_office, and the UPS Store certainly fulfills many of the same
functions as government-run post offices.

I don't think that sweeping changes like this across large geographic areas
should be made without communication of some kind.  One option would be
sending OSM Messages to all of the users whose work is being changed.
Another option would be sending a message to this talk-us mailing list.
 Is there some other communication channel that I'm not aware of?

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

Re: [Talk-us] GNIS tag removal proposal

[Peter Dobratz]

+1


On Tue, Aug 20, 2013 at 4:54 PM, Serge Wroclawski  wrote:

> Hi all,
>
> I've been looking at the GNIS data and it's quite a mess.
>
> As a step towards cleaning up the mess, I'd like to discuss removing
> some extranious gnis tags in the editors (just as we do with TIGER and
> other tags).
>
> I would like to suggest that the editors remove the following tags
> entirely:
>
> gnis:ST_num
> gnis:ST_alpha
> gnis:feature_type
> gnis:created
> gnis:state_id
> gnis:county_id
> gnis:county_name
> gnis:feature_type
> gnis:import_uuid
> gnis:reviewed
> gnis:edited
> gnis:description
> gnis:County
> gnis:Class
> gnis:County_num
>
>
> In addition, I suggest that we remove two other tags conditionally.
>
> I suggest we remove the "ele" tag unless the tag natural=peak is
> present and that we remove "source" if the value for that tag is "USGS
> Geonames" (which is just GNIS).
>
> I'd like to open a discussion about these tags, or other tags people
> suggest removing from gnis data.
>
> - Serge
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Tag for Tim Horton's

[Peter Dobratz]

I've never been to a Tim Hortons, but it looks like it's very similar to
the Dunkin' Donuts we have around here.  I tag this as amenity=cafe as they
are primarily a place that sells coffee-like beverages.

name=Dunkin' Donuts
amenity=cafe
cuisine=doughnut
drive_through=yes




On Wed, May 29, 2013 at 7:24 PM, william skora wrote:

>
> hi,
>
> I was just curious if there's a consensus on what tag to use for a tim
> horton's.
>
> I've found http://wiki.openstreetmap.org/wiki/Tag:cuisine%3Dcoffee_shop
> and http://wiki.openstreetmap.org/wiki/Canadian_tagging_guidelines
> which both mentioned variations.
>
> For what it's worth, this location was a single building and had a
> drive-thru as well.
>
> regards,
> will.
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] highway=primary, area=yes, leisure=recreation_ground?

[Peter Dobratz]

On Thu, May 9, 2013 at 8:03 PM, Martijn van Exel  wrote:

> http://www.openstreetmap.org/browse/way/29696644
>
> What gives? Looks like an undesired / undesirable side effect of a
> MassGIS import.
>
> Requires further investigation? Doesn't look like it's rendered, but
> it looks like it has very little ground truth to it either, apart from
> the weird tagging.
>

If you look at the history, the highway=primary tag was added by theflash
(a user with exactly 2 changesets).  So I would say that that definitely is
an error.

I would say that this entire polygon should be removed since it is really
trying to describe a linear feature, which we model in OSM using (unclosed)
Ways.  There are already such Ways in the database with tags like
highway=cycleway and railway=abandoned here indicating that this is a
former railroad which has been converted to a bicycle path.

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

Re: [Talk-us] Tagging Live indoor music venues

[Peter Dobratz]

On Sun, Feb 24, 2013 at 11:37 AM, william skora wrote:

>
> Hi,
>
> I was curious us to hear what others have been using to tag music venues.
> There's numerous places in my city that hold upwards of 1,000 people for
> music concerts (also called 'shows'). In the US, they're indoors, serve
> alcohol, and usually only open when there are shows. There's usually
> admittance fees to enter. I'm thinking of places like House of Blues (yes,
> there's restaurants adjacent to some of them, but the one i've been to is
> separate from the concert venue), (Cleveland places like Beachland
> Ballroom, the Grog Shop), and more famous places like Bowery Ballroom.
>
> I looked on the wiki, didn't find anything besides a dead proposal -
> http://wiki.openstreetmap.org/wiki/Proposed_features/Music_venue
> That uses amenity=music_venue and has 56 uses according to taginfo.
>

leisure=music_venue has 62 uses for what it's worth.

I put a music hall on the map a few years ago and ended up perusing the
JOSM menus and coming up with amenity=arts_centre, but looking back now it
appears that that really doesn't apply.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] new mapper tutorial explaining overlapping Ways?

[Peter Dobratz]

I'm trying to help along a newer local mapper, and I'm looking for any good
tutorials or references to help show how to create Ways without
overlapping/self-intersecting.

http://www.openstreetmap.org/browse/way/206875817

These are the kind of things that MapRoulette picks up and other users fix,
but it would be helpful to not create them in the first place.

Maybe it's just as simple as "oh yeah, you have to double-click to end a
Way in P2, not just retrace back to the beginning point."

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

Re: [Talk-us] Changeset 14828923 review

[Peter Dobratz]

On Sun, Feb 10, 2013 at 12:20 AM, stevea  wrote:

> **
> ...
>
Marsland Street (way 12700863) has many extraneous tags which are nonsense
> and can/should be removed:  access, area, bridge, cutting, embankment,
> junction, oneway and tunnel, if not more. ...
>

I believe the presence of all these tags is the user clicking through all
the tabs in Potlatch2 and filling changing them from "unset" to whatever
seems to make sense for them.  By convention, we know that streets are
assumed to be two-way and there is no need to add the oneway=no tag, but
now that I look at P2, there's no indication of this and even if you jump
to the wiki, it doesn't really give any suggestion to omit the oneway tag
for two-way streets.

http://wiki.openstreetmap.org/wiki/Key:oneway

For access restrictions, there's this complicated page which lists defaults
values by country.

http://wiki.openstreetmap.org/wiki/OSM_tags_for_routing/Access-Restrictions#United_States_of_America

I wouldn't expect a new mapper to know that.

I can't honestly decide if this is a real noob (no offense to noobs, I
> enjoy helping noobs...) or somebody bent on subtle but foolish vandalism.
> ...
>
>
I'd lean toward a new mapper that needs guidance here.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Reaching out to Local User Groups

[Peter Dobratz]

On Wed, Feb 6, 2013 at 5:10 PM, Paul Johnson  wrote:

> I wish there was a way to see locations of mappers beyond the first N
> closest and perhaps a way to filter by date of last activity.
>

Have you seen this map?

http://resultmaps.neis-one.org/oooc?zoom=9&lat=36.11773&lon=-95.87058&layers=B00TFFT
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] how to handle far away user making global changes

[Peter Dobratz]

I noticed recently that all of the facebook tags that I had been adding to
local businesses, have been changed to link:facebook by user
Shmias.
The facebook tag is something that I was using often in addition to, but
sometimes in place of, the website tag.

Although the facebook tag is not documented on the wiki, other people have
been independently using it in a similar fashion.  Then Shmias creates a
proposal:

http://wiki.openstreetmap.org/wiki/Proposed_features/link

I can't find any evidence of anyone actually supporting this proposal, but
then Shmias decides to change everything globally using the following
changesets:

http://www.openstreetmap.org/browse/changeset/13418262
http://www.openstreetmap.org/browse/changeset/13418199
http://www.openstreetmap.org/browse/changeset/13418158
http://www.openstreetmap.org/browse/changeset/13418123
http://www.openstreetmap.org/browse/changeset/13418028
http://www.openstreetmap.org/browse/changeset/13417925

I've posted to the talk page of the above proposal on December 3 and send a
private message to Shmias on December 5, but I haven't heard anything back
yet.

I guess at this point I would like to pursue reverting these changes, but
I'm not sure about what the next step is.

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

Re: [Talk-us] First vs 1st

[Peter Dobratz]

On Tue, Nov 27, 2012 at 11:55 AM, Richard Welty wrote:

> the rule we frequently follow is to use the format of the name on the sign
> as
> the guideline for what to put in the name tag for a way.
>
> there's a neighborhood in Rotterdam, NY where the signs for the side
> streets along the main drag (Hamburg Street) are spelled out (e.g., Fourth
> Street)
> but the signs interior to the neighborhood are abbreviated (e.g., 4th
> Street.)


I think the key word here is abbreviated.  In the OSM name, the
un-abbreviated form should be used:

name=Fourth Street

To record the abbreviated form, you could use the following:

short_name=4th St

As with all abbreviations, sometimes they will appear on signs and
sometimes the full word will be written out (often depending on size
contstraints of the sign itself).
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] Abbreviating names in tools

[Peter Dobratz]

> Please consider also using abbreviated names that already exist in the data, 
> perhaps in a tag like abbr_name, when such data is present.

I prefer to use the key short_name instead of the key abbr_name.
short_name is documented on the wiki, abbr_name is not.

http://wiki.openstreetmap.org/wiki/Names


On 10/4/12, David ``Smith''  wrote:
> Please consider also using abbreviated names that already exist in the
> data, perhaps in a tag like abbr_name, when such data is present.
>
> Looking forward to seeing this kind of conditional abbreviation rendering
> deployed in more places...
>

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

Re: [Talk-us] how to select overlapping objects in Potlatch 2?

[Peter Dobratz]

On Thu, Sep 6, 2012 at 10:45 PM, Dale Puch  wrote:

> Sidetracking a bit, but is there a way to select overlaping items in JOSM?
>

Yes, in JOSM, you can use the middle mouse button to choose between
selecting multiple things that are either close to or even on top of each
other.  See the "Unglueing and untangling" section here:

http://wiki.openstreetmap.org/wiki/JOSM/Advanced_editing
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] how to select overlapping objects in Potlatch 2?

[Peter Dobratz]

A relatively new user has created a bunch of duplicate Ways around here:

http://www.openstreetmap.org/?lat=42.732&lon=-71.49769&zoom=17&layers=M

So far I've had good email exchanges with this new user, and I was about to
send him a message about this area, but I'm not sure where to begin.
Obviously, we don't want a bunch of duplicate objects that represent the
same thing with almost the same geometry.  Looking at the area in Potlatch
2, I can't figure out a way to select just one of the overlapping objects,
so I can't even explain to the new user how to clean it up using his editor
of choice.  Also, I'm confused as to how this happened in the first place.
Has anyone else seen something similar to this before?

I can clean it up in JOSM, but I want to get on the same page with the new
mapper so that he doesn't keep doing it.

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

[Talk-us] wetlands in potlatch 2?

[Peter Dobratz]

I contacted a new mapper in my area and asked if he had any questions about
editing the map.  He asked about adding swamps to the map using the
in-browser editor.  I was about to respond that he could use
natural=wetland for such things, but I wanted to make sure that I gave him
instructions that made sense.  Since I almost exclusively use JOSM, I
decided to fire up Potlatch 2 and try to add a swamp.  Using the simple
interface, there doesn't appear to be any good way to do it.  Did I miss
something?  Is the best answer to this question that he should switch to
the advanced interface and type natural for the key and wetland for the
value? (Or just switch to JOSM?)

Thanks,
Peter
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] MA county boundaries

[Peter Dobratz]

Personally, I think it's easier to work from smallest to largest.  I've
been trying to keep track of the boundaries relations that have been
created here:

http://wiki.openstreetmap.org/wiki/Massachusetts/Boundary_Relations

It seems to be easiest to go town by town and then once you have all the
towns in a county, to clean up the county boundary.  Then once you have all
the county boundaries, clean up the state boundary.

This is a potential source for the territorial waters in the sea:

http://www.mass.gov/czm/mapping/czboundary.htm

TIGER also has county subdivisions for MA that can be used.

Whether you work smallest to largest or largest to smallest, it should be
possible to work on this concurrently without stepping on each other's
toes.  As long as you are making the boundaries more accurate each time you
edit them, then we'll eventually get to where there are sections of the
county/state/municiple boundaries that are one in the same.

Note: use fiveisalive may prefer to communicate via wiki.  See:

http://wiki.openstreetmap.org/wiki/Talk:MassGIS#admin_boundaries_in_Boston_area.3F

Also note: portions of the town boundaries in MA where imported improperly
resulting in duplicate Ways being uploaded.  If you notice up to a dozen
Ways on top of each other, all with the same tags, then delete all but one
of them before adding to the new boundary relation for the town.

On Mon, Aug 20, 2012 at 10:59 AM, Metcalf, Calvin (DOT) <
calvin.metc...@state.ma.us> wrote:

> I’ve been slowly editing the county boundaries in MA and today noticed
> somebody else (a user named fiveisalive) is also editing them, so I sent
> him the following message and suggest we coordinate our efforts here.
>
> ** **
>
> Hey I see you've been editing the county boundaries for MA as have I and
> we seem to be working against each other.  I was merging the county
> boundaries with the state boundaries (same way, different relations) as a
> step one before as a step two fixing the state boundary, while you seem to
> be fixing the county boundaries while leaving the state boundaries alone.
> The two issues I can think of with your method are 
>
> ** **
>
> 1. I'm not so sure how accurate the off shore county boarders are.  Both
> as compared to their offshore town boundaries (which they updated 1 year
> ago vs 5 years ago for the counties) and as compared to OSMs definition of
> the boundaries. 
>
> 2. The state offshore boundaries are one and the same as the offshore
> county boundaries as all parts of the Massachusetts are within counties and
> all parts of counties are within cities or towns (note other state are not
> like this).
>
> ** **
>
> Feel free to message me back, or email me via the us-talk or import list**
> **
>
> ** **
>
> ** **
>
> Calvin  Metcalf
>
> Office of Transportation Planning
>
> MassDOT
>
> ** **
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] trying to figure out the former path of a trolley line in Derry/Londonderry, NH

[Peter Dobratz]

I'm updating parts of the map in my local area (Derry and Londonderry, NH)
and it looks like there was definitely a railroad of some kind here before.

There's a dirt path here:
http://www.openstreetmap.org/browse/way/89361695
This connects to a road called "Trolley Car Lane":
http://www.openstreetmap.org/browse/way/18863837

If you look at the Bing image layer, you can see a line of trees that run
behind 109 Hillside Avenue in Londonderry and then intersect Wyman Street
in Derry.  Apart from that I can't see any traces of the railway from the
aerial photos.

There aren't any plaques on the ground or anything to explain what was once
here.  Trying some internet searches, I can't seem to find any additional
information about this former railway.  Does anyone have any ideas about
where to look to find more information?

Thanks,
Peter
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] administrative boundaries and Nominatim

[Peter Dobratz]

> The solution to this is to replace nodes with polygons, period.

>
> Is it sensible to have a polygon for the boundary and also a node that
> locates precisely the cultural/logical center?  Does Nominatim ignore
> the node when there is a same-named boundary?  Or is there some way to
> have a relation for a town that has both a polygon and a center?
>

 http://wiki.openstreetmap.org/wiki/Relation:boundary

Apparently you can add a node to the relation with role admin_centre.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] OSM Meetup in Metro Atlanta (Late Notice)

[Peter Dobratz]

On 7/23/12, Martijn van Exel  wrote:
> Increasingly, I also explain the use of apps
> (OSMTracker, Vespucci) as most people bring a smartphone. iPhones are
> not so useful as there are not too many OSM specific data collection
> apps available (I remember iLOE but don't know if that's still
> maintained? Also MapZen, discontinued afaik.

I have recently switched from pen + notebook to using my iPhone 4S
almost exclusively for taking notes.  I agree that there are not a
whole lot of OSM-specific apps that I have found on iOS.  I do find
the OpenMaps app useful for quickly displaying the Mapnik rendering of
my current location.  Also, the GPSTrack app is useful for recording a
track that I later email to myself and load into JOSM.  Mainly, though
I just use the built-in Camera app and Notes app.  I usually find
details in the photos that I didn't notice while I was taking the
pictures, and if I forget where I took the picture, they are
geotagged.  I use the Notes app mainly for recording house numbers and
other things that seem easier to just type in rather than taking a
picture.

Peter

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

Re: [Talk-us] Fwd: Re: Post bot cleanup

[Peter Dobratz]

On 7/19/12, Charlotte Wolter  wrote:
>  So, now, what can be done to fix it? Are there tools?

As previously mentioned, the best way is to actually visit these
places, take notes, and then use your favorite map editor to update
the map.

If you can't do that, or don't have time to do that, but want to work
on improving the map, you can still make the map better using
available information that has been approved for use in OSM.  This
technique is known as "armchair mapping."

Bing has made more easily available a lot of high quality aerial
photos of the world.  You can use these for images for drawing map
features such as roads, and also for aligning existing roads.

Once you have the roads traced in, various goverment agencies provide
public domain data that includes the road names.  The following page
shows instructions for making an overlay layer showing the names from
the TIGER 2011 dataset (public domain):

http://wiki.openstreetmap.org/wiki/TIGER_2011

This layer is just mostly transparent, but it shows the centerlines of
the roads from TIGER 2011 and has their abbreviated names printed next
to them.

Most of the map of the US was based on TIGER 2007, so I use this
sometimes when I notice roads on the images that aren't in the map and
I am not able to travel to the location I am looking at.

Before staring Armchair mapping, read the following disclaimer:

http://wiki.openstreetmap.org/wiki/Armchair_mapping

Basically, aerial images can be out of date and/or mis-aligned (though
Bing is generally updated within 2 years and has decent alignment
(check other's uploaded GPS tracks)).  And check the history of what
is there in case a local mapper has made a more recent update to a
feature that might not be reflected in aerial images.  It's a good
idea to contact any local mappers before you make sweeping changes to
the area they are mapping in.

Peter

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

Re: [Talk-us] Reminder: OSM US Tile Server

[Peter Dobratz]

It sounds like these should be added to the lists for JOSM and Potlatch2 so
that more people discover them.

http://josm.openstreetmap.de/wiki/Maps
http://git.openstreetmap.org/potlatch2.git/blob/HEAD:/resources/imagery.xml

Is there any reason not to add these directly to the above?

On Sat, Jul 14, 2012 at 8:19 AM, Ian Dees  wrote:

> Hi everyone!
>
> I wanted to remind you all that OSM US has its very own tile server that's
> currently serving a couple very useful tile layers (and some other
> moderately useful layers):
>
>
> http://wiki.openstreetmap.org/wiki/Foundation/Local_Chapters/United_States/Servers/Imagery
>
> I want to especially point out the new USGS layers I added earlier this
> week:
>
>- Scanned Topos: Shows scanned USGS topographic maps. These are
>usually the source for GNIS points and tend to be relatively old
>(especially when zoomed in), so you can look for data changes or spot
>incorrectly-named features.
>- "Large Scale" imagery: This is an amalgamation of all USGS imagery.
>When you're zoomed out (it starts at zoom 12), it uses NAIP (~1 meter
>resolution), but as you zoom in it will switch over to using other, higher
>resolution sources if they are available. I believe it includes sources
>that aren't in the HRO imagery layer. This is probably the best source to
>use for general mapping.
>
> Feel free to use these for your day to day mapping. In some cases the USGS
> has data that Bing hasn't included yet, so it's definitely worth adding to
> your arsenal.
>
> -Ian
>
> PS: If you have any ideas for other layers you'd like to see, I'm happy to
> try to get them rendered or shown here.
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>
>
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] railway=abandoned and mapping things that are not there any more?

[Peter Dobratz]

On Fri, Jul 13, 2012 at 4:43 PM, Nathan Edgars II wrote:

> On 7/13/2012 8:45 AM, Peter Dobratz wrote:
>
>> On 7/13/12, Nathan Edgars II  wrote:
>>
>>> On 7/12/2012 11:43 PM, Peter Dobratz wrote:
>>>
>>>> NE2,
>>>>
>>>> So after I bring up that I don't think railways should be drawn through
>>>> buildings, and most people agree with me on that, you decide to do this:
>>>>
>>>> http://www.openstreetmap.org/?**lat=42.762886&lon=-71.430509&**
>>>> zoom=18&layers=M<http://www.openstreetmap.org/?lat=42.762886&lon=-71.430509&zoom=18&layers=M>
>>>>
>>>> Does 86 Central Street, Hudson, NH have remnants of a railroad running
>>>> through their living room?  No, because that's ridiculous.
>>>>
>>>
>>> Is it on the former route of a railway? Of course. And that's what
>>> railway=abandoned has meant since I joined OSM.
>>>
>>
>> That is the exact point being discussed.  Why are we singling out
>> railways as a feature that essentially never goes away entirely even
>> though areas have been completely redeveloped?
>>
>> In the case of the houses on Central Street in Hudson, these aren't
>> exactly new construction.  Here are the dates of construction
>> according to the town of Hudson's property tax assessor:
>>
>
> Sorry about that, and thanks. I did in fact have it on the wrong side of
> the street: http://historical.mytopo.com/**getImage.asp?fname=mnch41sw.**
> jpg&state=NH<http://historical.mytopo.com/getImage.asp?fname=mnch41sw.jpg&state=NH>
> http://www.wardmaps.com/viewasset.php?aid=5127
>
>
I guess the main thing that is problematic about former railways is their
verifiability.

Between TIGER and GNIS and other imports, when I come across a road or
church that no longer exists (or in some cases never existed), I just
delete them.  But before I delete the objects, I try to make some effort to
move them to their correct location where applicable.  How am I supposed to
tell if these railways are in the right spot?  In this specific case, I was
able to make an argument that the railway was in the wrong spot and you
managed to dig up a map that's more than 100 years old to help find the
correct former railroad location.

Where does the information about the geometry of these former railways come
from?  What steps did you take to verify the data before putting it in the
OSM database?  How are you going to document this information for future
mappers?  It would seem to make sense to add additional information to a
source= or a note= tag indicating where the information about the previous
existence of the railway came from.  Or maybe there can be more verbose
information added to the wiki somewhere.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] railway=abandoned and mapping things that are not there any more?

[Peter Dobratz]

On 7/13/12, Nathan Edgars II  wrote:
> On 7/12/2012 11:43 PM, Peter Dobratz wrote:
>> NE2,
>>
>> So after I bring up that I don't think railways should be drawn through
>> buildings, and most people agree with me on that, you decide to do this:
>>
>> http://www.openstreetmap.org/?lat=42.762886&lon=-71.430509&zoom=18&layers=M
>>
>> Does 86 Central Street, Hudson, NH have remnants of a railroad running
>> through their living room?  No, because that's ridiculous.
>
> Is it on the former route of a railway? Of course. And that's what
> railway=abandoned has meant since I joined OSM.

That is the exact point being discussed.  Why are we singling out
railways as a feature that essentially never goes away entirely even
though areas have been completely redeveloped?

In the case of the houses on Central Street in Hudson, these aren't
exactly new construction.  Here are the dates of construction
according to the town of Hudson's property tax assessor:

84 Central Street - built 1854
86 Central Street - built 1880
90 Central Street - built 1880
94 Central Street - built 1860
96 Central Street - built 1964

When was this railway that runs through the above buildings in
operation?  It must have been more than 150 years ago.  Maybe we can
agree to delete this instance of railway=abandoned as it either no
longer exists or it's clearly not in the right spot.

>> Maybe you could hold off dumping stuff on top of work that I've done
>> while we continue to discuss the matter.
>
> Back at ya. Don't delete something that doesn't interest you.
>>
>> Does anyone have any objection to reverting the following changesets?:
>>
>> http://www.openstreetmap.org/browse/changeset/12202043
>> http://www.openstreetmap.org/browse/changeset/12186087
>> http://www.openstreetmap.org/browse/changeset/12191431
>
> For the record, I do object. And the fact that you would consider
> killing a fly with a sledgehammer is disturbing.
>

A sledgehammer is doing edits over a large geographic area dispite the
objection of a local mapper.  If you're going to put stuff over things
that are already mapped at least you can make it mesh nicely.

My primary objection to imports is basically dumping things on top of
already existing things without regard to work that other people have
done.

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

Re: [Talk-us] railway=abandoned and mapping things that are not there any more?

[Peter Dobratz]

On Thu, Jul 12, 2012 at 4:21 PM, Nathan Edgars II wrote:

> I am of course one mapper who's been mapping former railways. (Russ Nelson
> is another.) There is certainly value in seeing how the current
> disconnected bits of railway infrastructure used to connect. I've also
> mapped the occasional highway=abandoned, e.g. where the old road was cut
> off by the Interstate.
>

NE2,

So after I bring up that I don't think railways should be drawn through
buildings, and most people agree with me on that, you decide to do this:

http://www.openstreetmap.org/?lat=42.762886&lon=-71.430509&zoom=18&layers=M

Does 86 Central Street, Hudson, NH have remnants of a railroad running
through their living room?  No, because that's ridiculous.

Maybe you could hold off dumping stuff on top of work that I've done while
we continue to discuss the matter.

Does anyone have any objection to reverting the following changesets?:

http://www.openstreetmap.org/browse/changeset/12202043
http://www.openstreetmap.org/browse/changeset/12186087
http://www.openstreetmap.org/browse/changeset/12191431

Thanks,
Peter
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] railway=abandoned and mapping things that are not there any more?

[Peter Dobratz]

I'm trying to get a better understanding of the railway=abandoned tag
and see what the community thinks about it.

It seems that there are a handful of railroad enthusiast users that
are systematically adding current and former railways into OSM, and in
some cases re-adding railways that I have removed.  I have been
operating under the assumption that if a physical feature is not
currently there, then it should be deleted.

Consider the following example area (Manchester airport MHT):

http://www.openstreetmap.org/?lat=42.94312&lon=-71.4307&zoom=16&layers=M

A few years ago the runway was shorter and New Hampshire Route 28 was
more straight.  Sometime before that there was a railroad running
right through where the airport is now.  For the case of the road and
any buildings that were removed due to runway expansion, these are not
in OSM, and even if we could find some record of them, I wouldn't
think that anyone would add them.  However, for the case of the
railroad, there is a Way in OSM tagged railway=abandoned where the
railroad used to be.

What makes railroads a special case?

Do we really want a bunch of railway=abandoned Ways running directly
through newly constructed runways, buildings, roads, parking lots,
etc?

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

Re: [Talk-us] Staleness view

[Peter Dobratz]

On Wed, Jul 4, 2012 at 1:05 PM, Martijn van Exel  wrote:
> Thinking about good locations for a mapping party, or just about where
> to go by yourself for a mapping session, can become harder in places
> where the map is pretty much complete looking.
> Looking under the hood can be helpful. For example at the age of the
> data. Older data is more likely to need updating that data that was
> just edited a little while ago. Of course, this is a significant
> simplification, but knowing where data has not been updated for a
> while will be helpful for QA and targeting mapping efforts.

Using the "keep right!" QA tool, I've found the HTTP error indicator to be
very useful for finding shops that have closed.  Of course not everyone
enters website URL for shops when they add them to the map.

http://keepright.ipax.at/report_map.php?zoom=15&lat=38.90738&lon=-77.01989&layers=B0T&db=osm_US&ch=0%2C170%2C411%2C412&show_ign=1&show_tmpign=1
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

[Talk-us] holes in Columbus OH?

[Peter Dobratz]

I currently have the opportunity to do some mapping in Columbus, Ohio.
 The admin boundary for the city seems a bit odd:

http://www.openstreetmap.org/browse/relation/182706

There seem to be a lot of holes in the multipolygon relation, and it
seems strange to me that there are so many places within the city that
are not part of the city.  Is anyone more familiar with this area that
give an explanation for this?

I'm focusing on adding some POIs and correcting streets, but it would
be nice to have the city limits cleaned up at some point.

--Peter

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

Re: [Talk-us] Abbreviation expansion - USS to United States Ship?

[Peter Dobratz]

On Thu, May 10, 2012 at 12:33 PM, Nathan Edgars II  wrote:
> Is this a good idea? It looks really odd to me:
> http://www.openstreetmap.org/browse/way/9061339

I bet that no one actually says United States Ship in conversation,
but rather U.S.S.

I recently added an IHOP restaurant, and I ended up using name=IHOP,
though I thought briefly about name=International House of Pancakes.

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

Re: [Talk-us] tagging cul-de-sacs

[Peter Dobratz]

On Tue, Apr 10, 2012 at 10:17 AM, Martijn van Exel  wrote:
> On 4/10/2012 7:46 AM, Peter Dobratz wrote:
>>
>> I'm experimenting with the Java code from Traveling Salesman
>>
>> http://wiki.openstreetmap.org/wiki/Traveling_salesman
>>
>> http://www.openstreetmap.org/browse/way/158307363
>>
>> Is this how people generally map these things?
>
>
> No, I ususally tag the end node highway=turning_circle. There are so many of
> them littered around US suburbia that mapping each and every one of them as
> a circular way would make for a lot of not too useful data, not to mention a
> lot of work.

So we have 2 people in favor of discarding the circular portion of the
Way and just using a node with highway=turning_circle.  I actually
also use this approach on roads such as this:

http://www.openstreetmap.org/browse/way/157338683

I distinguish between having a solid paved surface and a traffic
island.  This seems to make sense based on the recommendations for
dual carriage ways (presence of physical separation causing separate
Ways).

According to the wiki:

http://wiki.openstreetmap.org/wiki/Tag:highway%3Dturning_circle:
"There is no central island/reservation to a turning circle—it's
simply a wider bit of road."

Are you saying that you delete these circular portions of ways and
replace with a node?

Peter

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

[Talk-us] tagging cul-de-sacs

[Peter Dobratz]

I'm experimenting with the Java code from Traveling Salesman

http://wiki.openstreetmap.org/wiki/Traveling_salesman

I'm making library calls to the routing code and it seems that the
router does not understand cul-de-sacs mapped as a single
self-intersecting way. This got me thinking about different ways to
possibly map cul-de-sacs.  I generally use Way with
highway=residential or highway=unclassified.  At the end of the road
there is a loop that intersects the same Way.  Here is one that I
recently mapped:

http://www.openstreetmap.org/browse/way/158307363

Is this how people generally map these things?

One other possibility that I could think of was splitting the circular
part at the end and tagging it junction=roundabout.  However, this
would imply that the road is one-way, and I'm not sure that that is
the case.  Typically there is no one way sign on the ground and people
feel free to travel in either direction on these (though being a
cul-de-sac they don't have a lot of traffic).

--Peter

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

Re: [Talk-us] City boundaries on the Canada/US border

[Peter Dobratz]

On Fri, Mar 30, 2012 at 5:17 AM, Paul Norman  wrote:
> There are a significant number of cities in BC and Washington which have
> borders that in practice[1] coincide with the Canada/US border. Currently in
> OSM these are represented with many nearly-overlapping ways.
...
> I am considering replacing these with one set of non-overlapping border
> ways.

+1

I believe many borders were imported as Ways without regard to borders
that were already there.  I like using Relations for borders and
having as few Ways as possible.

--Peter

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

Re: [Talk-us] Getting ready for the license change

[Peter Dobratz]

On Fri, Jan 13, 2012 at 10:58 AM, Mike N  wrote:

> On 1/13/2012 10:49 AM, Martijn van Exel wrote:
>
>> * Remap data that remains at risk of being removed, following the
>> guidelines on the wiki (see links below).
>>
>
> Assuming that the April 1 delete is a smart delete, is it better to wait
> until afterward?   I'm thinking of this -
>
>  1 TIGER - agreer
>  2 Bridges, etc - decliner
>  3 Attribution - Maxspeed, lanes, etc. - agreer
>
>  Rather than just deleting the road and starting over from aerial or
> TIGER, will the April 1 roll back the decliner's edits, thus preserving the
> agreer's contributions?
>

I agree that it makes more sense to clean up the map after the deletes from
the license change have taken affect.  As far as I know, it's still being
sorted out exactly which portions of the map will be deleted.
___
Talk-us mailing list
Talk-us@openstreetmap.org
http://lists.openstreetmap.org/listinfo/talk-us

Re: [Talk-us] [Imports] US Golf Courses from GNIS

[Peter Dobratz]

On Thu, Dec 22, 2011 at 2:45 PM, Josh Doe  wrote:
> I'd suggest this be reverted tonight, keeping the two corrected nodes.
> Also, when we re-import this (more slowly), I don't think we need any
> of the gnis tags except for the ID, which should probably use
> gnis:feature_id.

+1 for not adding copious amounts of tags along with the import.

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

Re: [Talk-us] [Imports] boundary import Update

[Peter Dobratz]

So you've got a git repository with OSM data that you eventually plan
on uploading to replace data that is already in the official OSM
database?  But you've already demonstrated that your method of
uploading results in lots of duplicate Ways, so I'm not sure this is a
good plan at all.  Is there any reason you can't follow the normal
workflow (download from OSM, edit, upload)?

Peter

On Sat, Oct 22, 2011 at 8:30 PM, Metcalf, Calvin (DOT)
 wrote:
> it is up at http://www.github.com/calvinmetcalf/ma-admin
> at the moment it just has lines and the admin_level=4 tags. I wasn't planing 
> on posting it to github, I was planning on building it on github with the 
> help of anyone else interested, including the deletions etc.  If you look at 
> my account I have one for t data too.
> Sent with Verizon Mobile Email
>
>
> ---Original Message---
> From: "Greg Troxel" 
> Sent: 10/22/2011 8:12 pm
> To: calvin.metc...@dot.state.ma.us
> Cc: impo...@openstreetmap.org, talk-us@openstreetmap.org
> Subject: Re: [Imports] boundary import Update
>
>
> I think the nuke/replace notion is sensible.
>
> So I would suggest that you put up (github seems fine, doesn't really
> matter as long as it's broadly accessible) several things:
>
>  a pointer to the massgis data, and maybe a copy
>
>  exact recipe for going from that to the .osm to upload (as scripts
>  that others could run, if at all possible), annotated with
>  explanatations for why the various choices are made
>
>  the proposed .osm to upload
>
>  how you're going to execute the nuke-existing-ma-borders operation
>
>  an explanation of what happens to the borders with other states (the
>  notion of leaving two and manual fixup after studying their view of
>  the border makes sense)
>
>  an explanation of how this relates to the coastline and the 12-nm
>  boundary (saying that it is totally separate for now seems ok)
>
>
> For the record, I agree that things are getting better, even if a bit
> messy right now, and that we're on the road to having higher-quality
> boundary data in Mass.
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>

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

Re: [Talk-us] Signed shared roadway

[Peter Dobratz]

On Sat, Oct 22, 2011 at 6:17 PM, Martijn van Exel  wrote:
> With "signed shared roadways" I meant those that are explicitly marked
> with a sign that says that bikes share the road. I will tag those
> 'bicycle=designated', looking at the definition for designated that
> makes sense.

If you are referring to a sign that simply says bicycles "share the
road", such as these:

http://www.trafficsign.us/share.html

Then I wouldn't tag this at all.  It is really just supposed to be a
friendly reminder that bicycles can travel on the road alongside cars.
 But bicycles can travel on most roads that cars can, so it's really
redundant information.  It's just supposed to be a reminder to drivers
that they should be courteous to those of us on bicycles, especially
since there is no explicit bicycle lane.  Around New Hampshire, they
often put up the share the road signs right after a bicycle lane ends.

Peter

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

Re: [Talk-us] [Imports] MA towns

[Peter Dobratz]

Thanks for uploading these borders.

Building on my recently added relations for Tyngsborough and
Dunstable, I decided to create a relation for Chelmsford.  On the
Chelmsford/Westford border you created 7 overlapping Ways, so I
deleted 6 of them and then added the remaining one to the relation.
Also the Chelmsford/Billerica border there are 9 overlapping ways, so
I deleted 8 of them.

Browsing through the state, it appears that this problem is fairly
widespread.  These can be easily seen on the Mapnik rendering since
they are shown as thicker lines.  Once you realize what is going on,
it's not too hard to just delete all but one of the duplicate Ways in
JOSM as you are creating the relation.

On Thu, Oct 20, 2011 at 3:08 PM, Metcalf, Calvin (DOT)
 wrote:
> Ok uploading.
>
> The linework that didn't make it in (i.e. border's contagious with counties 
> or towns that were already done) can be found at the link bellow for those 
> who would like to help clean this up.
>  http://dl.dropbox.com/u/37626989/other.xml
>
>>-Original Message-
>>From: Greg Troxel [mailto:g...@ir.bbn.com]
>>Sent: Thursday, October 20, 2011 3:03 PM
>>To: Richard Weait
>>Cc: Metcalf, Calvin; impo...@openstreetmap.org;
>>talk-us@openstreetmap.org
>>Subject: Re: [Imports] [Talk-us] MA towns
>>
>>
>>Richard Weait  writes:
>>
>>> Starting with Stow, (is this the right Stow?)
>>>
>>http://www.openstreetmap.org/?lat=42.4237&lon=-71.5119&zoom=13&layers=M
>>> you might decide that the existing town border is
>>correct-enough, then
>>> merge the Worcester / Middlesex border to the town line
>>where they are
>>> expected to overlap.  Refer to other sources, perhaps aerial imagery,
>>> at the same time and you'll have the opportunity to fix
>>other problems
>>> in / around Stow as well.  Then do a couple of the adjacent towns as
>>> well.  Adjacent towns should be easier because you'll have already
>>> sorted out one of the shared borders.
>>
>>For the record, the Stow data was hand-imported by me from the
>>very same
>>database Calvin is using, and it's 100% ok with me to nuke what I did
>>and replace it with a new load of the same data.
>>
>>(I intentionally did only my town because I knew we'd be
>>getting to this
>>later stage at some point, and wanted to create minimal additional work
>>for the later people.)
>>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>

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

Re: [Talk-us] Mass towns

[Peter Dobratz]

sounds good.

I think some of the county borders are already in the map as
overlapping Ways, and I was planning on cleaning those up eventually.

On Wed, Oct 19, 2011 at 9:05 AM, Metcalf, Calvin (DOT)
 wrote:
> Yeah ways on top of ways would be bad, I grabbed the massgis arcs from survey 
> points without coast which is a file that has just the linework with off 
> shore boundaries,  the ways are currently already split at junctions, my 
> method is going to be
> 1. take out counties boundaries and state boundary
> 2. take out boundaries already in
> 3. check to see if anything already there corresponds to the border (river 
> etc)
>
> Would that be helpful or am I missing the source of duplicate ways/need for 
> splitting?
>
>>-Original Message-
>>From: peter.dobr...@gmail.com [mailto:peter.dobr...@gmail.com]
>>On Behalf Of Peter Dobratz
>>Sent: Wednesday, October 19, 2011 8:56 AM
>>To: Metcalf, Calvin
>>Cc: talk-us@openstreetmap.org; impo...@openstreetmap.org
>>Subject: Re: [Talk-us] Mass towns
>>
>>http://wiki.openstreetmap.org/wiki/Shp-to-osm.jar
>>
>>It's basically just a Java command-line program that reads in a .shp
>>file and produces an .osm file.  It can use a configuration (rules)
>>file that allows you to map attributes in the .shp to tags in the
>>.osm, but I just create an empty rules file and add appropriate tags
>>manually.
>>
>>If you find it easier to add Ways without Relations first that would
>>be useful.  But I would avoid adding Ways on top of each other, as it
>>would seem like just as much work to start from scratch instead of
>>having to split Ways and delete duplicate Ways before adding them to
>>Relations.
>>
>>On Wed, Oct 19, 2011 at 8:14 AM, Metcalf, Calvin (DOT)
>> wrote:
>>> So on ubuntu you can open up a self extracting exe with
>>archive manager (I smacked myself in the head when I learned
>>that, turns out massgis makes them on linux).   I'm not
>>familiar with shp-to-osm.jar link? I've been using Merkaartor
>>to convert, but that usually forces me to then use kate or
>>gedit to clean up the tags it makes. I wasn't going to add any
>>extra tags except maybe population (so that a renderer could
>>use the relations to label).   Would it make sense for me to
>>add the ways now with relations to come as people are able, or
>>does that sound like it wouldn't save much time?
>>>
>>>>-Original Message-
>>>>From: peter.dobr...@gmail.com [mailto:peter.dobr...@gmail.com]
>>>>On Behalf Of Peter Dobratz
>>>>Sent: Wednesday, October 19, 2011 8:06 AM
>>>>To: Metcalf, Calvin
>>>>Subject: Re: [Talk-us] Mass towns
>>>>
>>>>I had just added Tyngsborough and Dunstable and my plan was to work
>>>>along the NH/MA border downward, especially over the winter when it's
>>>>harder to go out and see things with the snow accumulation.  I also
>>>>plan on continuing to clean up NH towns as people often import them
>>>>such that you have lots of overlapping Ways.
>>>>
>>>>For NH towns, I had been starting with the TIGER shapefiles.
>>>>Generally I open the .shp file in QuantumGIS, then select a town and
>>>>save it as a .shp file using EPSG:4326 - WGS 84 coordinates.  Then I
>>>>run this .shp file through Shp-to-osm.jar with a blank rules.txt.
>>>>This gives me an OSM file without tags that I can open in JOSM in a
>>>>separate layer.  If there's already a Way in the map (mostly from a
>>>>border of a neighboring town), then I split and re-use part of the
>>>>existing way and add them to a relation.  If there are 2 or more
>>>>overlapping Ways in the map, then I delete all but one of
>>them and add
>>>>it to multiple relations.  If the way doesn't exist yet,
>>then I either
>>>>merge a portion from my other layer or for straight lines I
>>just merge
>>>>the 2 end-points instead of the tens of points all along a straight
>>>>line from TIGER.
>>>>
>>>>On the Way itself, I add boundary=administrative and admin_level=4
>>>>(state), admin_level=6 (county), admin_level=4 (city/town).  On the
>>>>relation, I add type=boundary, boundary=administrative, name=(actual
>>>>name of the town), admin_level=(4, 6, or 8), wikipedia=(appropriate
>>>>article link).  Each way of the border is added to the relation with
>>>>role as outer (unless there's an unusual border where an inner is
>>

Re: [Talk-us] Mass towns

[Peter Dobratz]

http://wiki.openstreetmap.org/wiki/Shp-to-osm.jar

It's basically just a Java command-line program that reads in a .shp
file and produces an .osm file.  It can use a configuration (rules)
file that allows you to map attributes in the .shp to tags in the
.osm, but I just create an empty rules file and add appropriate tags
manually.

If you find it easier to add Ways without Relations first that would
be useful.  But I would avoid adding Ways on top of each other, as it
would seem like just as much work to start from scratch instead of
having to split Ways and delete duplicate Ways before adding them to
Relations.

On Wed, Oct 19, 2011 at 8:14 AM, Metcalf, Calvin (DOT)
 wrote:
> So on ubuntu you can open up a self extracting exe with archive manager (I 
> smacked myself in the head when I learned that, turns out massgis makes them 
> on linux).   I'm not familiar with shp-to-osm.jar link? I've been using 
> Merkaartor to convert, but that usually forces me to then use kate or gedit 
> to clean up the tags it makes. I wasn't going to add any extra tags except 
> maybe population (so that a renderer could use the relations to label).   
> Would it make sense for me to add the ways now with relations to come as 
> people are able, or does that sound like it wouldn't save much time?
>
>>-Original Message-
>>From: peter.dobr...@gmail.com [mailto:peter.dobr...@gmail.com]
>>On Behalf Of Peter Dobratz
>>Sent: Wednesday, October 19, 2011 8:06 AM
>>To: Metcalf, Calvin
>>Subject: Re: [Talk-us] Mass towns
>>
>>I had just added Tyngsborough and Dunstable and my plan was to work
>>along the NH/MA border downward, especially over the winter when it's
>>harder to go out and see things with the snow accumulation.  I also
>>plan on continuing to clean up NH towns as people often import them
>>such that you have lots of overlapping Ways.
>>
>>For NH towns, I had been starting with the TIGER shapefiles.
>>Generally I open the .shp file in QuantumGIS, then select a town and
>>save it as a .shp file using EPSG:4326 - WGS 84 coordinates.  Then I
>>run this .shp file through Shp-to-osm.jar with a blank rules.txt.
>>This gives me an OSM file without tags that I can open in JOSM in a
>>separate layer.  If there's already a Way in the map (mostly from a
>>border of a neighboring town), then I split and re-use part of the
>>existing way and add them to a relation.  If there are 2 or more
>>overlapping Ways in the map, then I delete all but one of them and add
>>it to multiple relations.  If the way doesn't exist yet, then I either
>>merge a portion from my other layer or for straight lines I just merge
>>the 2 end-points instead of the tens of points all along a straight
>>line from TIGER.
>>
>>On the Way itself, I add boundary=administrative and admin_level=4
>>(state), admin_level=6 (county), admin_level=4 (city/town).  On the
>>relation, I add type=boundary, boundary=administrative, name=(actual
>>name of the town), admin_level=(4, 6, or 8), wikipedia=(appropriate
>>article link).  Each way of the border is added to the relation with
>>role as outer (unless there's an unusual border where an inner is
>>necessary to make a whole in the town or county).  Note: it looks like
>>you could just as well use type=multipolygon on the relation as the
>>values for the roles are identical (inner/outer).
>>
>>Rather than deleting Ways or Relations in the database, I would
>>instead split existing Ways and edit existing Relations if they are
>>not right.
>>
>>For MA, I had also been using TIGER as I looked at MassGIS, but the
>>only download I could find was an EXE and wasn't sure how to extract
>>it on Mac or Ubuntu.  If you can point me to how to get an .osm file
>>from this on a non-Windows system, that would be helpful.
>>
>>It is annoying that MassGIS tends to use hundreds of points to
>>represent roads where 10 or so would suffice, and I've been deleting
>>hundreds of points on roads near what I happen to be editing.  Also,
>>if you do end up pulling stuff from MassGIS, try to avoid adding tens
>>of tags to each way just because there is a MassGIS database field, as
>>they are probably not all relevant.
>>
>>As far as Boston goes, I would think you'd put the whole city in a
>>relation with admin_level=8.  There's no convention to map smaller
>>divisions within a city or town.  Looks like the non-agreement on this
>>has been summarized on the wiki:
>>
>>http://wiki.openstreetmap.org/wiki/Neighborhood
>>
>>Peter
>>
>>On Wed, Oct 19, 2011 at 7:11 AM, Metcalf, Calvin

Re: [Talk-us] Combining State/County Borders & Physical Features?

[Peter Dobratz]

I too agree that the same Way should be used for waterway=river and
boundary=administrative whenever the boundary is defined by the river.
 That Way should also be added to a relation for the boundary.  If the
river has changed course, but the boundary remains in the prior
location, then the map should be updated at that point to make two
separate Ways.

Having nodes and ways on top of (or really close to) each other is
generally much more difficult to edit.

On Sat, Oct 15, 2011 at 4:36 AM, Nathan Edgars II  wrote:
> On 10/15/2011 1:16 AM, Jeffrey Ollie wrote:
>>
>> I noticed in the past few days that user Alexander Roalter has been
>> converting administrative boundaries in the Midwest to relations
>> (which I think is good) but in some cases he's combined state and
>> county boundaries with physical features, especially rivers.  A prime
>> example is the eastern border of Iowa and western border of Illinois
>> now shares the same ways as the Mississippi River.  I personally feel
>> that combining administrative borders with other features is not the
>> right way to handle the borders - while the boundaries may originally
>> have been defined by the river that won't always hold, see the history
>> of Carter Lake, Iowa for an example.
>
> I think I would agree with using rivers in this case, if my assumptions are
> correct. That is, if the river shifts slightly in the same channel, the
> boundary moves to follow the main flow line (or however it's defined), but
> if the river moves to a different channel, the boundary remains. Since any
> such major relocation will require more to be changed than a single way, the
> boundary is not in danger of being moved incorrectly by anyone who knows
> what they're doing.
>
> Of course this is only true when the boundary is defined to be the center of
> the river, rather than a line near one bank (e.g. the Ohio River).
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>

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

Re: [Talk-us] Interstate Crossovers

[Peter Dobratz]

I had been tagging them highway=service, but with access=private.

It sounds like access=official is used in places like Germany to
denote access to a bicycle path where bicycles and only bicycles can
travel on it.  Typically in this situation, there is a parallel path
that pedestrians and only pedestrians can travel on.

Peter

On Fri, Oct 7, 2011 at 4:12 AM, James Mast  wrote:
> Hey guys, I'm curious, but how have you guys been tagging them when you add
> them?
>
> I've been tagging them as "highway=service ; access=official".
>
> I was wondering if I was using the correct "access" tag so that routers
> wouldn't route you onto them by accident.
>
> -- James
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>
>

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

Re: [Talk-us] Announcement: Address Improvement project

[Peter Dobratz]

For secondary house numbers, I use addr:flats as this appears to be at
least semi-documented on the wiki.  I also use addr:full in this case
to write the whole address so it's clear what's going on.  We have a
number of buildings around here that use Sutie/Apartment/Lot/Unit
numbers along with the regular house number to indicate their address.
 Some hypothetical examples:

A hypothetical mobile home:
20 Main Street Lot 5
Derry, NH 03038
addr:housenumber=20
addr:street=Main Street
addr:flats=5
addr:full=20 Main Street Lot 5;Derry, NH 03038
addr:postcode=03038

A hypothetical townhouse:
12 Main Street Unit 6
Derry, NH 03038
(on the way for the whole building)
addr:housenumber=12
addr:street=Main Street
addr:postcode=03038
(on the node for the entrance to a townhouse)
addr:housenumber=12
addr:street=Main Street
addr:flats=6
addr:full=12 Main Street Lot 6;Derry, NH 03038

Though I haven't actually tagged any apartment buildings yet, I would
use addr:flats=1-6 on the node for the entrance to a building with 6
apartments in it.  I would also use addr:full to write out the full
address.

Peter

On Wed, Oct 5, 2011 at 2:12 AM, Nathan Mills  wrote:
> On Tue, 04 Oct 2011 22:39:24 -0700, Paul Norman wrote:
>
>> In this case the A was part of the house number and 112A had no connection
>> to 112.
>>
>> What I see around here more often is suite numbers (e.g. 101, 102) that
>> are
>> placed in front of the number when written out, but sometimes are placed
>> after on forms. Eg http://www.openstreetmap.org/browse/node/1052967131
>> where
>> I mapped it as addr:housenumber=#104-7885
>
> I just use the building's address number in addr:housenumber and write
> something
> like "112 West Center Street Suite 700" in addr:full if all the entrances
> are
> to the same building and have the same street address but different suite
> numbers,
> prefixes, or postfixes.
>
> -Nathan
>
>
>
> ___
> Talk-us mailing list
> Talk-us@openstreetmap.org
> http://lists.openstreetmap.org/listinfo/talk-us
>

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

Re: [Talk-us] What does the community want from a US local chapter?

[Peter Dobratz]

Yes, OSMF US shouldn't mandate a certain tagging scheme, but they
could certainly help to facilitate a consensus among the community.

Peter

On Fri, Sep 30, 2011 at 1:03 PM, Martijn van Exel  wrote:
> I believe that the core tasks of the US Chapter should be 1) to
> support and help grow the community, and 2) engage with existing and
> potential professional users to stimulate awareness and adoption.
> Defining the map itself should not be among them, this is and should
> always be a community effort, even though it is sometimes a cumbersome
> process. If you're saying that the Chapter could do something to
> support reaching a consensus or define the issue more clearly, such as
> facilitating a working group or a wiki clean-up, I could not agree
> more. But I don't see the Chapter going in and defining a definitive
> (is there even such a thing?) tagging scheme for roads. That would
> have the Chapter claim a mandate that it was never given and would
> only serve to divide the community rather than strengthening it.
>
> Martijn
>
> On Fri, Sep 30, 2011 at 10:27 AM, Peter Dobratz  wrote:
>> On Fri, Sep 30, 2011 at 10:25 AM, Richard Weait  wrote:
>>> 1) What would you like to see the US local chapter do?
>>
>> Come up with a consistent way to tag roads for the country.  Somehow
>> get a consensus from current mappers about how tag
>> US/state/county/whatever highways.  Update this page
>>
>> http://wiki.openstreetmap.org/wiki/United_States_roads_tagging
>>
>> and then remove "There is conflicting information on this topic at
>> several places."
>>
>> Specifically, on the Way:
>> ref=???
>> highway=???
>> name=???
>>
>> Add Way to Relation
>> type=route
>> route=road
>> name=???
>> network=???
>> state_id=???
>> ref=???
>> modifier=???
>> direction=???
>>
>> For exit Node:
>> highway=motorway_junction
>> name=???
>> ref=???
>> exit_to=???
>>
>> Once we know how to map what we have, then we can document in the wiki
>> specifically how every numbered road should be documented.
>>
>>> 2) Why should this be done by the local chapter, and not by individual 
>>> mappers?
>>
>> Every individual mapper can have their own idea about how these things
>> should be tagged.  As it is right now, mappers look at the conflicting
>> information on the wiki and then make an educated decision, only to
>> have someone come in later and change the tags on the roads, often
>> leaving the map in an inconsistent state.
>>
>> I think there may even already be a committee setup to tackle this very 
>> issue:
>> http://wiki.openstreetmap.org/wiki/Foundation/Local_Chapters/United_States/Working_Groups/US_Tagging
>>
>>> 3) Why should this be done by the local chapter and not by the OSMF?
>>
>> Though there may be some global consistency with road tagging,
>> standards need to be established for each country.  Road tagging is
>> often influenced by US states making designations and signing each
>> road accordingly.  These signs do not necessarily have an
>> international equivalent.
>>
>> ___
>> Talk-us mailing list
>> Talk-us@openstreetmap.org
>> http://lists.openstreetmap.org/listinfo/talk-us
>>
>
>
>
> --
> martijn van exel
> geospatial omnivore
> 1109 1st ave #2
> salt lake city, ut 84103
> 801-550-5815
> http://oegeo.wordpress.com
>

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

Re: [Talk-us] What does the community want from a US local chapter?

[Peter Dobratz]

On Fri, Sep 30, 2011 at 10:25 AM, Richard Weait  wrote:
> 1) What would you like to see the US local chapter do?

Come up with a consistent way to tag roads for the country.  Somehow
get a consensus from current mappers about how tag
US/state/county/whatever highways.  Update this page

http://wiki.openstreetmap.org/wiki/United_States_roads_tagging

and then remove "There is conflicting information on this topic at
several places."

Specifically, on the Way:
ref=???
highway=???
name=???

Add Way to Relation
type=route
route=road
name=???
network=???
state_id=???
ref=???
modifier=???
direction=???

For exit Node:
highway=motorway_junction
name=???
ref=???
exit_to=???

Once we know how to map what we have, then we can document in the wiki
specifically how every numbered road should be documented.

> 2) Why should this be done by the local chapter, and not by individual 
> mappers?

Every individual mapper can have their own idea about how these things
should be tagged.  As it is right now, mappers look at the conflicting
information on the wiki and then make an educated decision, only to
have someone come in later and change the tags on the roads, often
leaving the map in an inconsistent state.

I think there may even already be a committee setup to tackle this very issue:
http://wiki.openstreetmap.org/wiki/Foundation/Local_Chapters/United_States/Working_Groups/US_Tagging

> 3) Why should this be done by the local chapter and not by the OSMF?

Though there may be some global consistency with road tagging,
standards need to be established for each country.  Road tagging is
often influenced by US states making designations and signing each
road accordingly.  These signs do not necessarily have an
international equivalent.

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

[Talk-us] East Coast Greenway

[Peter Dobratz]

Has anyone attempted to start mapping the East Coast Greenway as a
cycle route?  http://www.greenway.org/

This is a project to create a bicycle route along the east coast from
Florida to Maine.  I think the goal is to get everything off-road, but
currently they have a designated route that follows bicycle friendly
roads where trails have not been built.  I believe some sections of
the route are signed, but most of it is not.

Peter

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

Re: [Talk-us] access=destination vs access=private

[Peter Dobratz]

> Do you think it makes more sense to tag the apartment complexes as
> access=destination or access=private? The complexes are not usually private.

I'd even consider not putting access restrictions on them at all,
unless there is some rule that you shouldn't be using them as a
through street.  What if you are walking or on a bicycle?

I generally only put access=private on roads that explicitly say
private on the sign.  A lot of condo complexes around here have a sign
on the driveway saying private property, no tresspassing, etc.
Apartments complexes usually don't have such a sign.

I use access=destination whenever I see the sign "no thru traffic."
("no thru trucks" is hgv=destination).  Also, I use access=destination
for things like roads into cemeteries, since presumably they don't
want people drive through them to get somewhere faster.

Peter

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

Re: [Talk-us] appropriate tags

[Peter Dobratz]

On Tue, Aug 23, 2011 at 10:20 AM, Metcalf, Calvin (DOT)
 wrote:
> Trying to figure out how to tag certain things in Massachusetts and was
> looking for some clarification on appropriate tagging.
> 1. The things on the Mass Pike that people tend to call rest areas should be
> tagged highway=services, I think.

Yes, I've used highway=services for some rest stops on I-93 in NH:

http://www.openstreetmap.org/?lat=42.755158&lon=-71.220856&zoom=18&layers=M

It looks like there is now a highway=rest_area that might be more
appropriate for this particular rest stop without a refueling station.
 Of course, you should add amenity=fuel, shop=convenience and
amenity=fast_food as appropriate within the rest area.

> 2. Highway depots, salt sheds, fiber optic lines, and microwave
> transmitters I can't seem to find mention of, am I looking in the wrong
> place? should I make up a tag?

There's man_made=tower for things like mobile phone towers.  Also,
there's amenity=grit_bin, but that might not be what you mean by salt
shed.  I usually just trace the outline of things and put building=yes
on them for miscellaneous service buildings.  I add access=private for
roads behind an "authorized vehicles only" sign.

> or is there a process?

http://wiki.openstreetmap.org/wiki/Proposal

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

Re: [Talk-us] Use of ref-tag on state highways

[Peter Dobratz]

On Sat, Aug 20, 2011 at 6:04 AM, Henk Hoff  wrote:
> To my understanding our tagging-standard for State Highways is [STATE]
> [NUMBER]
> http://wiki.openstreetmap.org/wiki/United_States_roads_tagging#State_Highways

I agree that we should use the standard 2-letter state abbreviation
and then the number: "AK 7"

Also, for some reason in Massachusetts, most of the roads were
imported using "SR" instead of "MA".  I think I changed some near the
New Hampshire border to MA and then someone changed them back, but I
never followed up.

I've been using "NH" for New Hampshire state routes and at least that
seems to have stuck.  Currently, I'm more focused on adding buildings
and points of interest to the map, but at some point I was hoping to
formalize the conventions for numbered highways and then clean up the
major roads in New Hampshire and parts of Massachusetts.

Peter

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

Re: [Talk-us] amenity:fuel and fuel types for the US

[Peter Dobratz]

On Thu, Aug 18, 2011 at 10:25 AM, Toby Murray  wrote:
> On Thu, Aug 18, 2011 at 5:54 AM, Richard Welty  wrote:
>> On 8/18/11 6:30 AM, Peter Dobratz wrote:
>>>
>>> Does anyone know how to translate the US octane measurements to the
>>> European ones used (fuel:octane_91, fuel:octane_95, fuel:octane_98,
>>> fuel:octane_100)?
>>>
>>
>> rather than messing about, i'd suggest modifying the tagging system
>> to let us enter the AKI numbers directly with clear labeling.

> Yeah, I would just label them as it is signed. No one cares about
> octane rating systems. They care about what it says in the car owner
> manual and where they can get gas that matches that number. It is
> implied that the numbers follow whatever rating system is prescribed
> by law in the country the gas station is in. Since all cars sold
> within a given country will follow the same rating system as all the
> gas stations in that country, the rating systems are a non-issue.
>
> So just tag it as fuel:octane_87, fuel:octane_89, etc.

That sounds logical.  I probably won't end up tagging them a whole lot
since virtually all gas stations (tagged amenity=fuel) currently sell
3 variations of unleaded gasoline.

> Not Sure about
> the kerosene. fuel:kerosene=yes would seem logical but I'm not sure
> what the non-taxable use has to do with things.

Looks like there's a handful of people using fuel:kerosone=yes, so
I'll go with that.

As far as non-taxable use, I believe it is common practice in the US
and elsewhere to put dye in diesel fuel and kerosene to mark the fuel
for non-taxable use.  Taxable use is basically using it to power a
vehicle that drives on the public roads.  Non-taxable use is things
like kerosene lamps or heaters.  Farm tractors might also use dyed
diesel fuel.  Basically, if it has dye in it, it is sold cheaper than
if it doesn't (less tax).

> While we're on the topic of fuel types... I've been wondering about
> diesel. The JOSM preset has 3 different diesel checkboxes. One is for
> bio diesel which is fine. But is there any chemical difference between
> "Diesel" and "Diesel for Heavy Goods Vehicles"? Or is this tag just
> clarifying the physical characteristics of the pump? (enough space and
> clearance for a big rig to maneuver to the pump)

Yes, it is for the physical characteristics of the pump, but it has
more to do with the nozzle.  Fuel nozzles that are used on big trucks
don't necessarily fit inside diesel cars.  Big trucks have larger fuel
tanks, so they need a bigger nozzle to help them fill up faster.

Peter

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

[Talk-us] amenity:fuel and fuel types for the US

[Peter Dobratz]

Since I drive a diesel car, I've been adding fuel:diesel=yes to a
bunch of gas stations around here, but today I filled up and took some
notes at a gas station and I'm not sure how to tag all of it.  Here's
a list of the fuels for sale at Haffner's in Hudson, NH:

Racing Fuel - minimum octane 110
Super Unleaded - minimum octane 93
Unleaded Plus - minimum octane 89
Unleaded - minimum octane 87
Kerosene - dyed kerosene non-taxable use only
Diesel

The only clear tag from the wiki is fuel:diesel=yes.

Does anyone know how to translate the US octane measurements to the
European ones used (fuel:octane_91, fuel:octane_95, fuel:octane_98,
fuel:octane_100)?

The racing fuel seems like it would need a new octane tag altogether.

Also, kerosene doesn't seem to fit into any of the pre-defined tags either.

Thanks,
Peter

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