Paul Johnson's diary

Recent diary entries

Proposal: Sunset ref=* on ways in favor of relations

Posted by Paul Johnson on 6 November 2015 in English (English)

ref=* on ways has become an unmitigated hot mess for maintainability, and now that relations have been around for a considerable amount of time. Let's kill this dinosaur like Linux killed ext filesystem support already. It's so long in the tooth it's starting to go from egregious to just sad.

Background: Prior to 2007 and the 0.5 API, OpenStreetMap had no "relation" data type, just nodes and ways. A stopgap measure to deal with routes was to tag them on ways. However, this quickly developed into a major issue as routes with their own refs for different modes and memberships cropped up as OSM developed past being merely a traditional street map. Relations were introduced, and quickly adopted as the standard tagging for routes of all kinds, with tags for most networks moving off the member ways to the relations. Presets for creating road routes with relations exist in (at least) JOSM and have for years now (I'd consider it a glaring, almost egregious, omission for any general purpose or highway-oriented editor at this point).

The problem: It's now been 8 years since the introduction of relations, and only route=road seems to be the sticking point for consistent route tagging, with qualities that should be part of the relation are instead still getting legacy-tagged on ways. This ends up being a major maintainability nightmare in places where multiple routes share the same way, especially whenever a new route is introduced or an old route is later deleted, and only gets worse the more routes share the same way (Illinois being an extreme example with some highways carrying as many as 9 signed routes), or the more road networks in play (Texas being an extreme example with 7 state level networks).

There's also the situation where ways themselves have their own refs often disconnected from the route (Oregon for sure for any highway with no route, all highways prior to 1928 and most highways 1928-2007, which is to say all but ~20-30 highways statewide) and possibly Pennsylvania (with the four-digit state highways) has this situation, but PA's case might just be a borderline case of routes with unsigned-ref).

Even in European cases, thanks to "E" routes in the EU, this is decidedly not a strictly North American situation.

In any case, adding or removing members and checking the relation's consistency is far easier, and easier to validate, than juggling multiple ref values in any sort of coherent fashion. And this is before even getting into issues of which order the refs go in...

The fix: I propose a goal of December 31, 2016 to eliminate ref=* as a method to describe an overlying route; this should be more than ample time for existing data consumers to catch up on doing a move and ensure data consistency for routes. Kill the dinosaur: Deprecate ref=* on ways from having anything to do with the routes they run over, use relations instead and phase out the use of route-describing refs on ways (be it removal of the tag or replacement of the key's value with a ref that actually applies to the way instead of the route).
Stop rendering this key and instead render the relations in opencarto and other featured layers.

Vandalism in Vancouver...

Posted by Paul Johnson on 5 September 2015 in English (English)


Posted by Paul Johnson on 17 July 2015 in English (English)


Location: Mingo Trail, Tulsa, Tulsa County, Oklahoma, 74117, United States of America

Addresses in Oklahoma

Posted by Paul Johnson on 4 May 2015 in English (English)

Looks like address quality in Oklahoma is going to continue to suffer, at least for a while. Some good news and bad news from the OKGIS community: First, the good news: Addresses are available under a license compatible with OSM. The bad news: $50 per county, and Oklahoma's got 77 counties, and only centroids. So, it's going to cost more than my truck did to be able to import a complete set of Oklahoma addresses. Damn!

Location: Mingo Trail, Tulsa, Tulsa County, Oklahoma, 74117, United States of America

Changing strategies

Posted by Paul Johnson on 17 May 2014 in English (English)

I'm trying to add lane detail and relations to every US, Interstate, Turnpike and State highway in Oklahoma. My previous strategy seemed a bit too resource intensive (start at one end of a highway, work to the other, across the state). So I've basically restarted, and I'm going to focus one county at a time, in order of population. This should have the advantage of being "more right" for more people sooner than the relative scattergun approach of doing it by numerical order. That said, next up: Oklahoma County.

Location: Barrows, Oklahoma City, Oklahoma County, Oklahoma, 73103, United States of America

Operation Cowboy: Tulsa

Posted by Paul Johnson on 3 November 2012 in English (English)
Location: West 3rd Street, Tulsa, Tulsa County, Oklahoma, 74103, United States of America

Generating ref=* from relations?

Posted by Paul Johnson on 6 July 2012 in English (English)

Mapdust bugs going down

Posted by Paul Johnson on 17 September 2011 in English (English)

Looks like software that supports Mapdust is making it more difficult to create an invalid bug, the result being the number of bugs open has been steadily declining since August. Now's a good time to start using the Mapdust plugin in JOSM so you can spot and close bugs in the areas you're editing in realtime.

Lake Shore Drive

Posted by Paul Johnson on 23 July 2011 in English (English)

This one seems odd...just noticed it in my RSS feeds.

Something tells me that 90 isn't the actual speed limit for Lake Shore Drive in Chicago.

Location: Little Vietnam, Uptown, Chicago, Cook County, Illinois, 60640, United States of America

MapDust cleanup

Posted by Paul Johnson on 30 June 2011 in English (English)

If you're in North America, I'm going over Mapdust cleaning out obviously invalid bugs and fixing what I can as I go along trying to get Mapdust back to something actually usable for the US and CA if folks could start following the georss feed of open bugs for their usual area, that would really help for future bugs.

Six more routes are now NCNs in the US

Posted by Paul Johnson on 26 June 2011 in English (English)
Location: Monumental Core, Foggy Bottom, Washington, District of Columbia, 20585, United States of America

Tulsa area freeways

Posted by Paul Johnson on 18 June 2011 in English (English)

OklaDOT moved I 44 again, I'm sure of it, between OK 51 and the Arkansas River. also, there's a new detour between Red Forks and the LL Tisdale Parkway ramp on the MLK Expressway...does anybody have the latest GPS traces for these two locations? If so, I'll be happy to do the edit if you can drive through the construction detours both directions.

Location: Riverparks West Bank Trail, Tulsa, Tulsa County, Oklahoma, 74107, United States of America

Heads up TriMet

Posted by Paul Johnson on 13 June 2011 in English (English)

I know there's a few folks at TriMet working on geometry and access issues, however I do have a couple friendly requests:

Don't squash "name=" without double-checking it first. Most glaring example: The Robert Baldock Freeway is not "I5 Freeway (North|South)bound." Nor is the Boone Bridge.

Don't squash "highway=" without looking. At least make an honest attempt at figuring out why something is tagged the way it is before arbitrarily changing paths to footways, cycleways to paths, etc, especially if they're already part of an lcn relation.

Don't squash "cycleway=*" without looking or ground knowledge. I've noticed several instances of opposite_lane being cleared or set lane when the lane is only on one side.

Cemeteries and the sense of mortality

Posted by Paul Johnson on 25 May 2011 in English (English)

I feel vaguely mournful when I map a cemetery, particularly if it's in a very small town or if it's a very large cemetery. Oddly, though, when I read the obituaries on some of the gravestones, such as when I hiked the Memorial Park Cemetery in Tulsa, I feel a bit less bad about how things will turn out. Then again, Tulsans tend to have a lot more noble deaths than Portlanders...

Ugh, I miss Tulsa. After getting let go from my driving job for poor job performance (out of 10 runs, I had six trucks break down on me requiring a tow, four automatic transmission failures, a broken tommy lift, and a truck with bad brakes that left me navigating a runaway through city streets(!), due to poor vehicle maintenance beyond my control), I more or less went nuts and drove back to Portland. I really miss Oklahoma...

Location: East 56th Street, Tulsa, Tulsa County, Oklahoma, 74145, United States of America

Google is pirating OSM

Posted by Paul Johnson on 18 May 2011 in English (English),-122.78183&spn=0.123633,0.308647&z=12

Knowing that while Sunset Highway is technically Northwest and Southwest by Portland standards (so it's not absolutely incorrect), yet, ODOT doesn't call Sunset Highway by either quadrant where no addresses front Sunset Highway, nor does any commercially available data or ODOT data set, I added "Northwest" and "Southwest" to that highway, after long suspecting Google was wholesale copying OSM based on prior, more subtle, detail edits I had made to OSM in the western Oregon and eastern Oklahoma areas (and thus, all of OSM) for months going on years. Since Google picked up the "NW/SW" gotcha on Sunset Highway, to me, this removes all doubt.

Does OSM actually enforce it's license and copyright? If so, how do we get this ball rolling? I'm /not/ going to work for Google for free if they're not going to provide attribution.

Help needed in DFW

Posted by Paul Johnson on 30 April 2011 in English (English)

Dallas/Fort Worth area could use some serious help, as reported by data consumers.

Location: Saint Louis Street, Dallas, Dallas County, Texas, 75201, United States of America

Attention Armchair Mappers!

Posted by Paul Johnson on 24 April 2011 in English (English)

I'm going to start adding bugs to Mapdust as I find them while on the road trucking. Given the generally good availability of aerial maps and GPS traces in the areas I'm driving in, these should be fairly easy to locate for an armchair mapper. Most of the locations will be in eastern Oklahoma, with some locations in adjacent parts of Arkansas, Missouri, Kansas and Texas. See for current bugs in this region.

OSM is being actively consumed by truck drivers in this region, and there's a *lot* of TIGER data needing review, so your efforts will definitely be appreciated.

Location: South 300 Road, Okmulgee County, Oklahoma, 74437, United States of America

Making Goods Van GPS traces available

Posted by Paul Johnson on 24 April 2011 in English (English)

I'm starting to upload traces taken from driving a goods van around Oklahoma and Arkansas. Using this to compare and contrast with aerial imagery will really help some pros on the road, since I'm sharing OSM with the other drivers at my company.

So, Tulsa type folks...

Posted by Paul Johnson on 9 April 2011 in English (English)

...anybody need a GIS tech?

Oregon and Fort Vancouver bicycle access to freeways updated

Posted by Paul Johnson on 26 March 2011 in English (English)

All motorways in Oregon and Vancouver, WA have been explicitly tagged bicycle=yes or bicycle=no based on Oregon state law as published by ODOT (in Oregon) or local knowledge based on signage (to the best of my knowledge in Vancouver). The following locations in this area are tagged bicycle=no:

* I 5 from I 205 in Salmon Creek to OR 217 both ways.
* I 5 from South Medford to North Medford both ways.
* US 26 from the PGE Park exit to I 405 both ways.
* US 30 from 26th Avenue to I 405 both ways.
* I 84 from I 5 to 122nd Avenue eastbound.
* I 84 from Wood Village to I 5 westbound.
* I 205 from OR 43 to WA 14 both ways.
* I 405, entire length, both ways.

The rest of I 5 in Oregon (and the southbound approach to I 205 in Washington based on several "BICYCLES MUST USE I 205" sign on I 5) is posted bicycle=yes, as is the rest of I 84, I 82, and US 26. There are no freeway portions of US 30 that permit bicycles (as the only freeway portion of US 30 is the above mentioned section). WA 14 is tagged bicycle=yes due to lack of signage and having personally used that segment by bicycle without WSP getting involved, as was a small portion of WA 500. Motorway links connecting to segments closed to bicycles were also tagged bicycle=no, whereas links that connect to open sections were tagged bicycle=yes.

I also changed a couple sections that were improperly tagged as motorway to trunk based on local knowledge (part of US 97 and part of OR 22).

In addition to the source:url for that changeset, the actual OAR (Oregon Administrative Rule) is available at

Older Entries | Newer Entries