Re: [Talk-us] Boston speed limit too Re: Michigan speed limit changes coming soon

[Andrew Guertin]

On 01/07/2017 06:49 PM, Bill Ricker wrote:

On Sat, Jan 7, 2017 at 6:31 PM, Greg Troxel  wrote:


Also, we do have the implicit 30 mph tagged on many roads.   While there
are usually not signs, it is entirely verifable.  One only has to read
the law and measure the distance between houses (or observe that the
area is built up with businesses).   These two tasks are entirely within
the ability of a typical mapper.



​the question then is, can we tell (without driving in circles) is if an
existing ​30 mph tag in Boston was implicit or explicit ... to find which
might need fixing


Isn't this what the source:maxspeed and maxspeed:type keys are supposed 
to solve? So the answer is that yes, you can tell, if the original 
mapper added enough detail when they mapped it.


No idea if mappers in Boston have added that detail for you, though...


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

Re: [Talk-us] Railway crossing challenge for MapRoulette

[Andrew Guertin]

On 07/06/2015 10:46 AM, Martijn van Exel wrote:

(In particular I am never sure whether to use crossing or level_crossing.)


My understanding is that railway=level_crossing is where cars cross a 
railway, and railway=crossing is where pedestrians cross. I'm not sure 
what to use where bikes cross.


--Andrew


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

Re: [Talk-us] Best practices for high-density residential areas

[Andrew Guertin]

On 03/31/2015 01:07 PM, Steve Friedl wrote:

2) Are rectangular house outlines good enough?

So in my area I've been making the outlines look actually like the house, as
best as I can, but there's no way I'm going to do this to every house in
America.  For other areas, assuming house outlines are warranted, I can use
the building tool in JOSM (what a *great* tool) to make strictly rectangular
outlines that vaguely approximate the shape of the house. What are the
thoughts on this?

a) A rectangular outline is great, thank you
b) It's better than nothing,  but only marginally so
c) drawing squares on non-square things is inaccurate
d) something else?


Focusing just on this one, I often approximate buildings by rectangles 
when they're not technically but are pretty close. Lots of buildings 
seem to be a rectangle with a part of one wall that sticks out by a 
meter or two, or to have a bay window, or any other of an endless number 
of tiny variations. If it's close enough that a rectangle is the right 
shape at a lower level of detail, then there's absolutely nothing wrong 
with mapping it at lower detail.


Sometimes, like for a clearly L-shaped building, it's just better to add 
the two more vertices though.


--Andrew

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

Re: [Talk-us] Edits needing investigation in Tennessee

[Andrew Guertin]

On 10/10/2014 11:54 AM, Andrew Guertin wrote:

While investigating changes in my area, I noticed that
http://www.openstreetmap.org/user/Rondale has quite a large number of
changes recently where they changed highway=* to highway=residential (*
including at least service[1], unclassified[1], tertiary[1], secondary,
primary[2], and trunk[2]).

I have messaged them specifically about the changes in Vermont, but
there are other changes in Tennessee that I'm not local to and wouldn't
feel comfortable discussing or reverting. In Vermont, the highway
changes were paired with other changes (road deletions, road additions,
road additions through buildings[3]...) that may (?) have come from
working with old aerial imagery. Is there anyone here from Tennessee to
check if similar problems exist in this user's edits to that area? And
would feel comfortable reverting if appropriate?


Rondale never responded, and I reverted the edits in my area. Since that 
time, they have continued to make the same type of edits. Yesterday, 
they recreated one of the edits that I had reverted. I have contacted 
the DWG for intervention.


--Andrew.


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

Re: [Talk-us] Prima Facie Speed Limits

[Andrew Guertin]

On 09/08/2014 05:27 PM, Tod Fitch wrote:

[...]
instead there is a state wide prima facie limit:
source:maxspeed=US:CA:residential
[...]


My state doesn't have such a limit, but my city does. Supposing I 
started tagging things with source:maxspeed=US:VT:Burlington, would 
anyone be upset that Burlington and residential are in the same 
place in the hierarchy?


(I'm hoping the answer here is no one cares, the value isn't intended to 
be machine parsable, and both values are understandable by humans.)


--Andrew

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

Re: [Talk-us] Beach routing

[Andrew Guertin]

On 07/09/2014 12:50 PM, Elliott Plack wrote:

OSM US:

I've been using some routing engines to map fitness routes (e.g. Strava)
that use OSM data. Along our US coasts, there are beaches. The beaches I'm
familiar with are popular with walkers and joggers to go up and down the
shore, since access is generally open to anyone along the water's edge. I'm
considering adding a `highway=path` along the beach to facilitate this. I'd
add the connections to the walking paths between parking lots and the beach
as well.

For uninterrupted strips of sandy beach, would a path be appropriate to
indicate walkability?

How the map looks now in iD: http://i.imgur.com/2EQ06BR.jpg
What I'd propose to do (note the connections):
http://i.imgur.com/i8dj6lQ.jpg
Area of the examples:
http://www.openstreetmap.org/#map=18/38.45143/-75.04957

Thanks,


Contrary to the other replies, why not just teach the routers that 
beaches are something that can be walked (or ridden or driven) on?


Access restrictions can go on the beach itself, with bicycle tags if 
it's explicitly forbidden. There's no documented default value of 
surface for a beach, but sand is probably a decent guess. The beach can 
already be tagged with fee=*. Paths can connect to the beach area. All 
of this is already set and available for use by routers.


If you add a separate path, a router can't know whether it needs to 
apply the fee from the surrounding beach or not. If you also tag fee on 
the path, a user won't know whether having paid the fee for the beach 
also entitles them use of the path, or whether they can pay just for 
walking rights and not swimming. Surface needs to get tagged multiple 
times, as do any access restrictions. And in the end, it's really just 
not a path anyway.


That said, I understand the appeal of just making things work now, and I 
wouldn't be too beat up about it if paths do get added.


--Andrew

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

Re: [Talk-us] Beach routing

[Andrew Guertin]

On 07/09/2014 12:50 PM, Elliott Plack wrote:

OSM US:

I've been using some routing engines to map fitness routes (e.g. Strava)
that use OSM data. Along our US coasts, there are beaches. The beaches I'm
familiar with are popular with walkers and joggers to go up and down the
shore, since access is generally open to anyone along the water's edge. I'm
considering adding a `highway=path` along the beach to facilitate this. I'd
add the connections to the walking paths between parking lots and the beach
as well.

For uninterrupted strips of sandy beach, would a path be appropriate to
indicate walkability?

How the map looks now in iD: http://i.imgur.com/2EQ06BR.jpg
What I'd propose to do (note the connections):
http://i.imgur.com/i8dj6lQ.jpg
Area of the examples:
http://www.openstreetmap.org/#map=18/38.45143/-75.04957


Today I learned about 
http://en.wikipedia.org/wiki/Ninety_Mile_Beach,_New_Zealand , which is 
officially a public highway. Curiously, it's mapped in OSM with a 
separate way marked highway=path, bicycle=yes, which doesn't really 
match up with the Top Gear video of Jeremy Clarkson passing a tour bus 
at high speed.


I'm sure this means something for this topic, but I'm not sure what.

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

Re: [Talk-us] Sidewalks as footpaths

[Andrew Guertin]

On 04/30/2014 11:38 AM, William Morris wrote:

Is there a general OSM policy on marking sidewalks as highway=footway?
User dolphinling appears to have gone crazy in downtown Burlington,VT
tracing the sidewalks and calling them footways. Which wouldn't be a
problem if footways weren't so cartographically distinct in everyone's
stylesheets:

http://www.openstreetmap.org/#map=17/44.47772/-73.21112

Should I:

1. Revert
2. Get in touch with the editor
3. Get over it

Thanks!

-Bill Morris
@vtcraghead


Hi Bill :)

I was following the tagging at 
http://wiki.openstreetmap.org/wiki/Tag:footway%3Dsidewalk


I'm personally not a fan of the way Mapnik renders footways (I'd prefer 
a thin grey line rather than a dotted red line), but it hasn't yet 
bothered me enough to propose a Mapnik change--in large part because at 
low zooms the roads cover the sidewalks.


Do you have any concerns other than display? The wiki mentions some like 
it being harder for a pedestrian router to say something like Follow 
the sidewalk along main street with sidewalks as separate ways. If you 
have this or any other concern, I'm happy to add more detail to help 
mitigate them.


--Andrew

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

Re: [Talk-us] Aerial Imagery for Chittenden County, VT

[Andrew Guertin]

On 02/12/2014 11:15 PM, Andrew Guertin wrote:

I'm working on the rest of the county now, and I'll put it in the
same place when it's available. I expect it to take a week or two of
cpu time.

--Andrew


Quick update on this: conversion from jp2 to tiff and merging into one 
large (107GiB) file are complete, but I don't have enough disk space to 
start the tiling. I'll investigate as I have time.


--Andrew

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

[Talk-us] Aerial Imagery for Chittenden County, VT

[Andrew Guertin]

The Chittenden County Regional Planning Commission 
(http://www.ccrpcvt.org) recently acquired 15cm per pixel orthophotos 
for all of Chittenden County, Vermont, and has made them available 
online[1]. They seem very high quality, in both alignment and visibility 
of detail. They are also more recent--spring 2013--than anything else I 
know of.


I have received confirmation from them that they consider these images 
public domain, so I've started processing them and making them available 
as tiles for easy use in JOSM (or any other editor). I have all of 
Burlington processed so far. It can be used with the TMS url 
tms[20]:http://www.uvm.edu/~aguertin/ccrpc/{zoom}/{x}/{-y}.png . I'm 
working on the rest of the county now, and I'll put it in the same place 
when it's available. I expect it to take a week or two of cpu time.


--Andrew


[1] I've withheld the URL as they haven't publicized it and I don't want 
to be the cause of their server being hammered, but if you have some use 
for the original data, let me know.


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

Re: [Talk-us] Bing imagery update

[Andrew Guertin]

Unfortunately, in my area (Burlington VT), it seems like the 2008-era 
high resolution images (zoom 20?) are no longer available, and only the 
2010? era zoom 19 are there.


On the Bing website I can still see the higher resolution images, but 
neither JOSM nor the Bing Imagery Analyzer is displaying them.


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

Re: [Talk-us] [OT] Anyone ever talked about adding more Land Ownership data to OSM?

[Andrew Guertin]

On 01/07/2013 10:45 PM, Richard Welty wrote:
 On 1/7/13 10:37 PM, the Old Topo Depot wrote:
 
 We do have an issue with US state and county borders, as some are 
 missing, incorrect, incomplete or incorrectly tagged.  Perhaps we
 can organize a cleanse the state and county borders project to
 improve the data quality and currentness.
 
 i would like to switch the USGS based state border of NY for what i 
 perceive to be the somewhat more accurate borders from TIGER.  i've
 temporarily stalled on bringing in town borders from TIGER because of
 discrepancies along the state line between NY and New England. i
 gather that many of the state borders are USGS and that the TIGER
 borders may be better. this is a non-trivial exercise as any
 county/town borders that share ways with the state borders will need
 to be fixed up.
 
 richard

If and when you do this, could you compare to the VCGI data for the
NY/VT border at
http://www.vcgi.org/dataware/?page=./search_tools/search_action.cfmquery=themetheme=003x=17y=6
(BoundaryOther_BNDHASH layer)?

I plan to import VT town boundaries from that file at some point, but
was not planning on touching any of the state boundaries (except to
connect town boundaries to them). I'd be interested to know how well
that data source matches others for the state boundary, too, though.

Further information about this file (licensing, etc) on the imports list.

--Andrew

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

Re: [Talk-us] Importing highway surface tags

[Andrew Guertin]

On 12/20/2012 05:03 PM, Adam Franco wrote:
 * Has anyone located a good source for state or national road surface data?
 The TIGER data doesn't seem to include surface information as far as I can
 tell.

The VCGI EmergencyE911_RDS file has a field for this. Unfortunately,
58773 out of 64302 values (91%) are Unknown.

The VCGI license doesn't explicitly give the permissions needed for OSM,
but when I asked to use the town boundaries layer they gave permission.
(I still need to get around to that...)

 * Is this a project that the OSM community in Vermont, the broader region,
 or nationally (assuming data is available) would support? I'd rather not do
 a lot of work to prepare it if there is no desire for inclusion in the data
 set.

I'd support it.

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

Re: [Talk-us] Burlington, Vermont road classification

[Andrew Guertin]

On 10/19/2012 07:55 AM, Greg Troxel wrote:
 
 Primary highways generally lack stop signs; however, stop signs may
 control major intersections in rural areas with low traffic volumes
 and occur rarely elsewhere.
 
 The most notable example of this is North Willard Street[2]. It is 
 part of US Route 7, but as can be seen with Bing Imagery, it is 
 narrow, made narrower by street parking on both sides, and is 
 controlled by stop signs. Similarly, Main Street is part of US 
 Route 2, but has many lights, and does not even satisfy the near 
 the highest speed generally allowed on surface streets note about 
 secondary streets.
 
 My take from dealing with this (around Mass):
 
 If it's a US highway, then it's highway=primary, period.  A US 
 highway is important simply by virtue of being designated US 
 highway.

Good to hear. Which area of Massachusetts is this from? My experience
with driving there is mostly on 93 and 2 and in Boston, so I don't
really have a good handle on what a US highway feels like there.

 Note that speed limits etc. should be tagged, so routing is not just
  on classification.

A good reminder, yes. I should add this to my plans.

 But I don't know anywhere where a US highway is not important in 
 terms of cultural/transportation geography, even if it isn't the 
 first choice for long-distance travel. [...]

 An example in vermont that's kind of iffy is 100.  I see parts of it
  are primary, and parts of it secondary.  As a non-local who's driven
  it only a few times I have no basis for questioning local judgement.
  But I would tend to think that 100 is more important than most other
  NS roads that aren't US5 and US7. But, the other state roads that
 100 are more important than should be secondary, so it's really in 
 between primary and secondary and thus a tough call.

I'm not actually too familiar with 100--I don't know that I've actually
ever been on it myself. VPR had a long repeated segment on it recently,
where a pair of commentators traveled its length and talked about one
town each week. My impression is that it has cultural significance, but
that for going from the bottom of the state to the top, most people
would find their way 7 or 91/89 first.

 US7 should really be primary.  Even if it's slow in cities, it's the 
 main road where it goes (I89 aside, and generally the 'is it primary'
 test discounts interstates).  I am assuming that if you are in
 Shelburne and going to Colchester (and we stipulate that interstates
 are unusable), you'd drive on 7, including North Willard street.  Or
 at least someone not really familiar with the area would. Is that off
 base?

I can't really speak for what someone unfamiliar with the area would do,
but I have made that trip many many times, and there are many different
ways, each approximately equally good:
* 7 the whole way
* 7 - Colchester Ave
* 7 - Cliff - Prospect - Colchester
* 7 -- Union - Winooski - 7
* (if starting farther south) Spear - East - Colchester
* 7 - 189 - 89
* 7 -- Pine -- Battery - Pearl/Colchester
* ...

Of these, I see most people take the interstate, followed by 7 -
Colchester Ave, with Spear Street being a popular choice for a calmer,
lower-traffic drive. Staying on 7 is probably more common than the
weirder routes involving Cliff or Pine, but it wouldn't get you there
significantly faster than them (or slower than the more popular ones).


I took a look at traffic numbers from
http://www.ccrpc.us/data/traffic.php?town=BURLINGTONyrs=Ayear=2011count=ATR.

The traffic on various parts of North Willard Street ranges from ~7000
cars/day in some areas to only 2900 for US 7 North of North St.
Meanwhile, both Colchester Ave and Riverside Ave are usually ~15000 and
never below 1.

I'm not really sure how to interpret this.

--Andrew

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

Re: [Talk-us] Talk-us Digest, Vol 59, Issue 20

[Andrew Guertin]

On 10/18/2012 07:58 PM, William Morris wrote:
 Third local mapper chimes in: As weird as the cartography will look (and
 I've seen it appear as such on OSM in other U.S. cities), Route 7 through
 Burlington has no business being listed as primary. I can hit a maximum of
 25mph on the sections between stop signs, and by character that street is
 more of a Residential Road.
 
 That said, it might be worth asking public works what they think; the city
 transportation layer on VCGI marks it as primary, but I wonder how they
 treat it locally (particularly with snow removal priority).
 
 Either way the summit drink of choice should probably be a switchback :)
 
 -Bill
 North Ave.

Oh, Hi Bill! I didn't count you because it seems most of your mapping
has been outside Vermont, but I should have realized you'd still have
the right combination of local knowledge and OSM experience. Sorry about
that!

--Andrew

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

[Talk-us] Burlington, Vermont road classification

[Andrew Guertin]

Hi,

There are two active mappers in the Burlington, Vermont area, and we
disagree about how the roads should be classified, so we're looking for
more opinions.

The crux of the problem is the answer to the question: Which is more
important, outside/official classifications, or physical characteristics?

The tagging pages on the wiki don't really provide clarity on this
matter. For example, from [1],
 Almost all other U.S. Highways get highway=primary. A primary
 highway generally provides the best route (excluding motorways)
 connecting adjacent cities or communities

 Even where U.S. Highways connect only smaller communities, they still
 merit highway=primary

but

 Primary highways generally lack stop signs; however, stop signs may 
 control major intersections in rural areas with low traffic volumes 
 and occur rarely elsewhere.


The most notable example of this is North Willard Street[2]. It is part
of US Route 7, but as can be seen with Bing Imagery, it is narrow, made
narrower by street parking on both sides, and is controlled by stop
signs. Similarly, Main Street is part of US Route 2, but has many
lights, and does not even satisfy the near the highest speed generally
allowed on surface streets note about secondary streets.

Of note, there is in fact no path to get from US 7 south of Burlington
to US 7 north of Burlington without stopping at at least one stop sign,
except for the interstate. Should this imply that there just aren't any
major roads here?


We're especially interested in input from nearby states--the rest of New
England and northern New York, but of course anyone with an opinion
please chime in!

Thanks,
--Andrew






[1] http://wiki.openstreetmap.org/wiki/United_States_roads_tagging
[2]
http://www.openstreetmap.org/?lat=44.48388lon=-73.20368zoom=16layers=M

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

Re: [Talk-us] Burlington, Vermont road classification

[Andrew Guertin]

On 10/18/2012 05:07 PM, Richard Weait wrote:
 On Thu, Oct 18, 2012 at 4:48 PM, Andrew Guertin andrew.guer...@uvm.edu 
 wrote:
 Hi,

 There are two active mappers in the Burlington, Vermont area, and we
 disagree about how the roads should be classified, so we're looking for
 more opinions.
 
 If you are both local mappers, I suggest that you actually meet face
 to face and share a beverage.  [...]

While not a bad idea, I don't think that this is necessary or helpful
for this case. We're both impressed with each other's work, and (it
seems through text at least) perfectly willing to accept the other's
viewpoint, it's just that now we've realized that the docs are ambiguous
enough to make *both* viewpoints valid, and we'd like to choose the one
that most closely matches the rest of the map, especially in nearby areas.

In other words, amicable disagreement, not a budding edit war. :)

--Andrew

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