All map programs have errors

[+user13371]

I noticed a funny mistake in Google Maps near my home, so I looked at the same area in Garmin City Navigator NT, Ibcyus, and Delorme's online map viewer in "guest" mode.

 

Each of them had different errors in the same neighborhood. City Navigator looked the best; Google and Ibcyus were both pretty good, Delorme had the most errors. But I don't know if looking at data.delorme.com as a guest gives you the same maps as a paying customer gets.

 

Google's funny mistake: Marking a footpath through a park as part of SE 101ST Avenue. This might be a "fair" mistake because it's where 101st would be -- if it really went through the park. And Trimet (the local bus company) has a bus stop there labeled as "SE Powell and 101st." Other than that, Google does a good job of drawing the streets to match up with their aerial photos.

 

City Navigator's only obvious mistake nearby: Omitting a part of SE 97th Ave that both Google and Ibcyus show. This might also be fair since it's not a public road -- at that point it's really just a service road for Trimet's garage/service area. Also in Garmin's favor, they showed SE 101st Ave is unpaved and other maps didn't.

 

DeLorme showed SE 101st Ave going from Powell to Holgate and completely omits Ed Benedict Park (which has been there for over 20 years). There are also a few other streets drawn where they don't belong, and street segments missing in the surrounding neighborhood. As I said though, I don't know if this is just in Delorme's online map viewer or also shows up in their commercial products.

 

Has anyone else looked at their own neighborhood in different map programs an found collections of varying mistakes?

Edited November 16, 2008 by lee_rimar

[+embra]

Your methodology is not bullet-proof, Lee, but I think I'll sign on to your conclusion anyway...that's been my experience. I haven't done a cross-sample of my area yet, but I bet I'll find errors. Given that there are a gajillion details to incorporate, I'm only surprised we don't see more errors. I can appreciate

 

Looks like you were probably seeing the same as the Delorme Topo7 street data. Here's a screenshot of the hybrid HiRes/Topo7 view. The satellites don't lie, but the park doesn't show as a POI or map feature in T7.

 

I saw the discrepancy in Googlemaps with this shot on SE Francis between 103rd & 104th. It appeared from zooming in on the HiRes that the gap is unpaved road...or even alley. It looked like two ruts. Heck, I can add a screenshot of that, too.

 

 

[+user13371]

... discrepancy in Google ... SE Francis between 103rd & 104th... gap is unpaved road...or even alley. It looked like two ruts...

Max, around here the phrase "unimproved roadway" makes folks chuckle knowingly

 

That stretch of Francis is passable by foot, mountain bike, or MAYBE car with a high ground clearance. I would hesitate to put it on a map. It's quite a bit worse than the stretch of SE 101st Avenue south of Ed Benedict Park. Truly, the footpath in the park is better paved than either of them.

 

Portland has many such unimproved roads. I'm sure they give mapmakers fits trying to make the judgment call you tried to make: road, alley, or just some ruts ...?

Edited November 16, 2008 by lee_rimar

[+embra]

I gotta say that guy's house just to the north of the 4WD trail looks pretty interesting from the sky. Looks like a little rainforest in there.

[+user13371]

That's nothing. Look at the NW corner of Holgate & 101st -- there's a house there, really

Edited November 17, 2008 by lee_rimar

[+MartyBartfast]

Lee I just stumbled across this post by accident.

 

Over here (the UK) one of the main sources for map information (and the oldest as they've been doing it for almost 300 years) is the Ordnance Survey; all their data is copyrighted and they deliberately introduce small errors on their maps (e.g. a road near me is about 1/2 mile long and dead straight but the OS map shows a kink in the middle of the road, there are also examples of them showing a building on a map where none exists). Several years ago they sued a road atlas company for using their data without paying for it by showing that the road atlases included these deliberate mistakes.

 

Maybe this is what's happened here.

[+GEO*Trailblazer 1]

You do know that all maps have errors.

 

Any how thats why I make my own from all the gleaned data.

If you have the right programs you can edit all these maps you mentioned.

But will not go into that detail here it is all in the forums somewhere.

 

I do GIS and GPS 911 mapping and use my laptop and GPS hooked together while navigating.

I update areas as I find them or work on them.

There is mega data to deal with so a few missed spots by the other mappers is not that much to work on.

At least you have most of the data already in front of you.

 

And those new homes,buildings,roads,shopping centers and Highway Construction being done needs to be updated so really it is just a never ending progress so as soon as a map is put out there it is already out of date.

 

I can show you some HUGE mistakes in mapping but will refrain from that topic for now.

Yes Old data is still used and they do not want to change it.

That is why we have a better set of maps than they do.

We have more precise equiptment now than they did when it was done so yes there is errors in everything.

 

But that 1795 Law for the US that says a corner can not be changed regardless of error still stands.

[+Renegade Knight]

My street doesn't exist on most maps. If it does it's wrong. In the past year I've seen one map service fix it.

 

While I dont work in the GPS map biz, I do work with mapping for different reasons. There are always mistakes. The key thing is the how robust the effort to correct mistakes is. Different approaches can lead to one company having a heck of a lot less mistakes than another.

[SiliconFiend]

I'm surprised noone has mentioned OpenStreetMap yet. See a problem, fix it yourself. Presto!