Google Crowdsources Map Editing 149
An anonymous reader notes that Google now makes it possible to edit the map location designated by (almost) any address. Registered Google users in the US, Australia, and New Zealand can move incorrect markers for their homes or businesses to the correct locations. Access to some listings is restricted — hospitals, government buildings, and businesses whose listings have been claimed through Google's Local Business Center. In addition, moving a place marker more than 200 yards (or 200 meters) from its original location requires a moderator's approval before the change shows up on the map. Once a marker has been moved, a "Show Original" link will direct users to the original location.
with great power comes great responsibility (Score:1, Interesting)
That's great, but what about moving correct markers to an incorrect location? Or the location of a competitor to the address of the local hog farm?
For those who can't read past the very first line. (Score:2, Informative)
Google covered their bases. All their bases.
Re: all your bases (Score:2, Funny)
Re:For those who can't read past the very first li (Score:3, Interesting)
Re: (Score:2)
Re: (Score:2)
I wonder, does it check to make sure my IP address is near the location I'm moving, or is it just a glitch?
Re: (Score:2)
I moved my home's marker just fine, but when I went to move my parents', I got this error: "Because of technical restrictions, you cannot edit this location at this time."
I wonder, does it check to make sure my IP address is near the location I'm moving, or is it just a glitch?
More likely they used cookies to mark you as having moved your address marker. They may have then noted that "you" were trying to edit a second home address.
Consider that there is no mapping from IP to physical address that works for all IP addresses.
Re: (Score:2)
Re: (Score:2, Insightful)
No, but maybe a bad unscrupulous competitor can have a hog farm built within 200 yards of their location :-)
If it's on the same street, the moderators may have a tough time verifying the placement anyways; short of going there, finding another map, or sending a fax to the business to verify their location on a written map.
It's not like a Google rep can call residences/businesses on the phone and easily ask them to verify that X is their right physical location. (No way to communicate a map in a
Re: (Score:2, Funny)
Re: (Score:2)
Google: Hi, is the the Hog's Head bar and grill?
Business: Yes.
Google: Great. Are you in the building two houses south from the corner of 5th and Main Street, or the pig barn 200m down the road?
Business:Yep. Two doors south of 5th and
Re: (Score:2)
Move out to the countryside and you will see the Google map and reality differ even more.
You'd get close using your method, but there's still a way to go.
Re: (Score:1)
Re: (Score:2)
I just fixed the one for my new house (Score:5, Funny)
-- Hillary
Re: (Score:1)
--Exxon Mobil Corporation
Re: (Score:2)
I cou
Re: (Score:2)
Support openstreetmap instead... (Score:5, Informative)
Re:Support openstreetmap instead... (Score:4, Informative)
Why help any of them? The US data is FREE anyway... 99% of the people who pay for the data just dont realize that.
Re: (Score:2)
Sorry for being US centric in that last post... I can definitely see helping them in non-US areas.
As for US areas, I am baffled that they have no data for anywhere in my area (30 minutes from NYC in a major suburb) - well, to be truthful, they have an outline of the geography (land and water boundary line) - but thats it... especially since it is free, and doesnt take too long (for them) to import all 70GB of it into a database... for us to feed the data through their web server would take months...
So,
Re: (Score:1)
OSM wiki [openstreetmap.org]
Here in the UK our taxes pay for the Ordnance Survey mapping, but we have to pay for the results. I don't know the exact licencing rules, but it could be that a charity would have to pay to include a detailed map to show where they were holding an event. They could use the OSM map for free. Many cities are very well mapped, but there is still a lot of work to do. I've been doing my village when I have time. I've already covered some feat
Re: (Score:2)
Are you referring to TIGER/LINE data? Yes, that's free but companies such as Navteq and TeleAtlas have better quality/more up-to-date data. 99% of people who pay for it care about accuracy.
Re: (Score:2)
Really? More up to date? Sorry, but it isn't more up to date... the government doesn't give them a special release of it.
More accurate? In some instances - due to the already mentioned online correction tools that their customers can use to update incorrect info...
And from testing against that data I can tell you, those "corrections" (assuming they are accurate) make up a very small percent of the data. As of this day, my road, (and houses on it) built in 1940, in a county with 3 million people, still is
Re: (Score:2)
If you think that Tiger data is just as good as what you can get from TeleAtlas and NavTeq-- then you obviously haven't seen these datasets. I have. No commercial GPS unit vendor, or map web service uses Tiger data because it sucks.
Tiger doesn't contain sign data. It doesn't describe one-way roads or turn restrictions. It doesn't include form-of-way information. It doesn't tell you how large roads are. It doesn't offer a fully conneted road network. And on, and on...
Re: (Score:2)
http://www.theregister.co.uk/2007/10/31/garmin_tomtom_teleatlas_bid_war_mapping_gps_satnav/ [theregister.co.uk]
Thanks for bringing this up, I wasn't aware it was going on. Looks like Garmin are winning at the final hurdle.
Re: (Score:2)
As usual, slashdot posters get it wrong.
As usual, slashdot posters don't read. The discussion was about finding ADDRESSES using such tools. And hate to break it to you, but neither NavTeq or TeleAtlas are much more accurate than the TIGER/Line data that they are based off of.
For instance, my road, with the houses built over 50 years ago, with NO changes in at least the last 35 years (that I can personally confirm from living there) still shows a house 1-9 through those services... they dont exist, never did.
If you think that Tiger data is just as good as what you can get from TeleAtlas and NavTeq-- then you obviously haven't seen these datasets. I have. No commercial GPS unit vendor, or map web service uses Tiger data because it sucks.
That's where you are wrong... or m
Re: (Score:2)
NavTeq and TeleAtlas operate in Europe and the far East as well as in the U.S. There is no TIGER data in these areas. Therefore, trivially, NavTeq and TeleAtlas data is not always "based off of TIGER," although that may be true in some cases in the U.S.
Re: (Score:2)
Re: (Score:1)
Yeah cool. The wikipedia of mapping. For when you want to sent up the creek literally.
Re: (Score:2, Funny)
Good point (Score:2)
We all need to be very careful to not help unless absolutely forced to, since a more worthy cause may appear later. Don't be caught helpless!
Thanks for the helpfull reminder!
Suggestion (Score:5, Funny)
Other map crowdsourcing tools (Score:5, Informative)
Same as open source (Score:2)
'Crowdutainment'
Depends on why you contribute. In my community of practice (see sig) contributing is not necessarily done for fun, but might actually happen for job-related reasons. The same could be said of open source, my team participates at open source software projects because we directly benefit from these specific projets. Example; some commercial applications use OpenStreetMap data.
(and to answer the anonymous coward who replied to my parent post, OSM is the best map crowdsourcing project, no one said it's the bes
Interesting issues it raises (Score:5, Interesting)
There IS a reason why Google's (and everyone else's) data is incorrect. I'm wondering if Google got their data directly, or wasted money paying for it from TeleAtlas or NavTeq or one of the other companies that gets it for free...
The US Tiger-Line Data it is based off of (SAME errors in data - I know, I've got the whole Tiger-Line set to use for comparison) clearly states in the massive 369 page "Technical Document" (well I think 369 pages is kinda large) that the data is purposefully innaccurate to ensure that it cannot be used to pinpoint the exact location of any residence to help ensure some level of privacy for each citizen.
By allowing users to correct the information, it also means the interpolative data for other addresses becomes accurate or more accurate... for instance, if my neighbor corrects his location pointer, and you look addresses on the street, even if his is in the database as an exception rule, you can easily spot the exception and re-plot the rest of the data.
For reasons of National Security (second reason cited in the Tiger-Line Docs), that also can be bad, because figuring out a pretty near exact location of sensitive areas just requires someone(s) who live on each side to correct their info.
Especially considering the data set works with 6 decimal places of latitude or longitude precision (which is about 13" give or take for most US locations... in Alaska it is far more accurate on the longitude portion at 6 decimal places)...
I'm still up in the air as to whether this ends up being a good thing or a bad thing...
Re:Interesting issues it raises (Score:5, Insightful)
Re: (Score:3, Interesting)
True... though they dont hide it, it just isnt acurate, even though the information is stored in 6 decimal places...
And as for the gang checking out our site, you will find only one difference in the data compared to Google's - I dont adjust it to the left or right to make it pretty... the Tiger-Line Data specifies road direction, and whether the address is on the "left" or "right" (ie: if the road segment goes north, left addresses are on the west side). Google takes the exact same results, and moves the
Re: (Score:1)
Re: (Score:2)
The THEORY (ie: not mine, and dont know or care how valid it is) is that if the map data was accurate, bad people could pinpoint an exact location to within 13" +/- a few inches to do bad things, like use their imaginary WMDs to to pinpoint bombing - and (my guess) also to make the average citizen feel slightly more anonymous since the locations on the map dont actually correspond to their real location in the real world (most people wont make the connection you made - that the directions and location info
Re: (Score:2)
Re: (Score:2)
You said this was a privacy issue. How in the world does "OMGWMDZ!!!" equate to privacy?
Re: (Score:2)
Why are you asking me?
Let me re-quote what I wrote so you can re-read it...
The THEORY (ie: not mine, and dont know or care how valid it is)...
There... I even bolded the appropriate section. It Isnt MY claim, it's what our very own government wrote in the TIGER/Line Manual.... not me.... as I said, I dont care either way whether their claim/theory/lie is accurate or not.
Re: (Score:2)
Targeted solicitation or personal attacks would probably become more precise. For example, many lawyers, doctors, and others in positions of risk or notoriety may choose to muck around with home/deed records to obfuscate where they live. For example, crafty doctors or lawyers may create a fictitious business name, then op
Re: (Score:2)
Interesting that the inaccuracies are by design.
Re: (Score:2)
Actually, Google seems to get their data from a variety of sources, NavTeq being one of them... trying out different zoom levels and different locations will show you different "contributors" for the data that Google uses.
For instance, try New York, NY... you will see the copyright statement says "Google Imagery, NASA, TerraMetrics, MapData & NavTeq" - other areas will have one or more of those and/or others that arent in that list...
As for the innaccuracies, I doubt they are by Google's or NavTeq's
Re: (Score:2)
But the thing is, on my address lookups the data is like on the wrong side of the street. I don't see a p
Re: (Score:2)
Some of the data is missing and gets interpolated from the rest... for instance, my house range isnt in the MAIN data. What I mean by that is, the even house numbers for my road segment arent in the dataset (the odd ones are). Figuring out where my house "is" (as accurately as any of the data will allow) simply requires using the odd numbered data to calculate it - or looking through the secondary data sets (where my house range IS located). Many sites, (and even some of the large companies that use it for
Re: (Score:1)
Re: (Score:2)
Yes, this is funny, but in light of the parent post: how much funny and how much true?
Re: (Score:2)
Sounds like marketing nonsense to me. If they direct you to the right neighbourhood, then they might as well be directing you to the right address, as any stalker will just use street signs and house numbers once they're in the vicinity to get a finer grained location. The real reason map data providers insert errors is so they can dete
Re: (Score:2)
Yes, that is part of the problem (as I mentioned elsewhere)... and that problem pertains to rural and urban areas both... your example fits both - for relatively the same reason. A building in NYC may have a much larger footprint than the one next to it, but since there is no specific coordinate for each building, the software has to interpret it in a linear fashion (unless another address exception was added by someone to redefine the address segments - which is relatively simple).
But part of it is becaus
Re: (Score:2)
The problem I see arising from such situations is that many companies do not properly implement the TIGER/Line data. There is an entire table with such address variations listed, as well as another one that (usually) breaks the segments into smaller ones with the addresses on the correct side of the road. The main files do not have that supplemental information, and many companies that base their data off the TIGER/Line data seem to only use the main files... which is a very innaccurate method for dealing w
Fixing wrong business information (Score:2)
Thanks
Not bad! (Score:5, Funny)
Re:Not bad! (Score:4, Funny)
For what it's good (Score:5, Funny)
Following Calvin and Hobbes strategy, one never knows who hits from above.
Can't this be automated? (Score:2)
--
Educational microcontroller kits for the digital generation. [nerdkits.com]
Re:Can't this be automated? (Score:5, Informative)
Hi... it's not quite that simple. Here is how the data works... let's assume you live on a straight street...
The data has StartLong, StartLat -> EndLong, EndLat and corresponding StartHouseLeft, EndHouseLeft, StartHouseRight, EndHouseRight for that segment (which may or may not be your whole street - depends on whether the street curves somewhere, or is intersected by other streets, etc). Google, Tele, Nav, etc take what address you enter, and interpolate it based off that data so...
If the data set says your street starts at #0 and goes to house #40 and yours is house #20, it interpolates your address to be dead center on the segment and calculates that lat/long point based off that... but... what if half the houses on your street have 150' frontages, while the other half have 80'? Well, then the data is innaccurate... or what if (which seems to be the standard) the data starts your street at 1, but your street actually starts at 14? (Mine is exactly like that... so the whole first segment is highly innacurate). And the segment data dont take into account the WIDTH of interesections... so segment one (when it hits an intersection) ends in the middle of it. Segment two starts at that exact point. If the intersecting road is a rural or suburban local road, it may be 30-40' across... if it is a highway, it may be a couple hundred feet across (depending on median size, # of lanes, etc). That also makes all data even more innacurate (because the start address gets located on the highway - as the corner is represented by a point intersection instead of by a 2D road and highway width intersection.
So, no, there is no way way to fix it - because even though the data does say what type of road each segment is, that still wont tell you how wide the road (or any median on it) is. For instance, Interstates in the middle of no-where are often 2 lanes each direction... or in Norther Jersey, hit 6 or 7 lanes each direction... they both show up as the same road type.
Re: (Score:2)
I'm curious about that too... if they are leaving the data in "node-like" structures, it should. But personally, I have no idea.
It is as easy as this though:
-Segment info currently defines house 2-100 even (followed by 100-200, etc)
-You correct your location info for house #68
-They revise the end node (100) to be 68, insert another dataset going from 68-100
-The result being 2-68 and 68-100 searches are more accurate.
The way the original data is set up though, that can alternatively be done by adding
Summary Inaccuracy (Score:4, Informative)
Re: (Score:2)
The claim may not be true, but it is indeed made in the article:
Re: (Score:2)
In any case, I slightly misread the summary; registered users from all over the world can edit locations in the US, Australia and New Zealand. This feature is not implemented when viewing British maps.
The moral: laziness leads to the wrong conclusions.
Is there anything these guys do which isn't smart? (Score:2)
I hope they never turn evil, because come 10 years, 95% of internet connected users are going to be using their services.
Add missing data (Score:5, Insightful)
Re: (Score:1, Redundant)
http://slashdot.org/comments.pl?sid=366869&cid=21431207 [slashdot.org]
Same time stamp and everything!
Re: (Score:2, Interesting)
Google mostly uses NAVTEQ data for maps.google.com (they seem to use TeleAtlas when it's an embedded map).
For NAVTEQ data, you can use their Map Reporter [navteq.com] to submit information. Once they get around to [1] incorporating the new data, it may then take another few months to filter back to Google and all their other customers.
[1] My street (built circa 2000) is missing from NAVTEQ, so I submitted a report describing the street, and all the house numbers about a month ago. It doesn't seem like anybody ther
NAVTEQ Reporter Tool = No Response (Score:2)
For NAVTEQ data, you can use their Map Reporter to submit information. Once they get around to [1] incorporating the new data, it may then take another few months to filter back to Google and all their other customers.
Don't hold your breath. When I moved into my new house 16 months ago, the first thing I did was report my missing subdivision (which had already been there for 12 months or so at that time) using the NAVTEQ tool. I've been checking it once a month and there has been ZERO action on it. No comments from staff, even though I've left several asking for status updates. No NAVTEQ update, no GMap update. Not worth a damn if you ask me.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
How about correcting roads? (Score:2)
I can locate my house from satellite, even see it. As soon as I switch to hybrid or "roads" it becomes a little silly. I've checked with old timers in the area; meaning they still have their farms and family plots; and all agreed that the two roads in question never were there. (I haven
Re: (Score:2)
Imagine when they didn't have the web and online maps.
How did they do then? Communicate over phone? Send a map by postal mail? OMG
Re:Add missing data-- Deadtech, anyone? (Score:2)
I called them, back in 2001/02, when I HAD a home (lost it (sold it orderly) to the dot-commie economy crash...). It was listed on country property tax rolls, correctly, as a 2533 sq ft, 2-story, 4+ bedroom, 3 full bath, LR, DR, FR, nook, kitchen, 2-car atta
Re: (Score:2)
Of course, when we moved there it was 1988 and 75th street was considered "far south" Naperville (before Rickert even existed)...now I think it might qualify as "central" or "north central"!
Stupid McMansions in Will County...
Go Brookdale Bobcats! rofl
How about adding locations? (Score:4, Insightful)
Real estate records would have been better (Score:5, Interesting)
It would have been better if they tied the map data to real estate ownership records. Much of that data is available in machine-readable form. It would be cool, and useful, to zoom in and see the property lines. Displaying the ownership information would be even better. It's a public record, after all.
Or if they recognized house numbers in the imagery taken by the StreetView truck.
Re: (Score:2)
Re: (Score:2)
Some locales don't have the data, but the support is already there.
Re: (Score:2)
US-only? (Score:1)
meter == yard? (Score:3, Interesting)
In that case I would move markers by 200 meters, it gives you 18 yards more power.
I assume that you can move a marker only once, so that you can't keep moving markers 10 times in a row to move it 2000 yards (or 2km).
I foresee edit wars, markers that move constantly in a radius of 200 yards (or 200 meters).
And how would a moderator know if the edit >200 yards (or >200 meters) is correct or not? Maybe the company moved to a different building and google's info isn't up to date yet.
Re: (Score:3, Funny)
Most companies don't take the building with them...
Re: (Score:2)
Thats nice but... (Score:1)
Here's my house [google.com] (please keep arial bombardments aimed at my neighbours)
Not only are the maps not lined up with the satelite imagery, but my property is aparently twice as big as its supposed to be... unless I now inadvertantly own two houses and some crazy people who say otherwise are living my second house.
If they trust me enough to let me fix the marker, can't they let me fix the maps too?
somewhat unsafe (Score:2)
Isn't that very insecure for those who have unlisted numbers and non-published addresses?
Re: (Score:2)
1) Bother to correct your address on Google Maps if it's wrong? You don't want people to know what it is anyway, right?
2) Use your real name when doing 1?
Wonderful... (Score:2)
Naturally, that information is not editable. I've complained about this to my Google employee friends for years (not in a whiney sort of way), and even they have no idea how to correct it. On the other hand, I suppose I can live with being placed in Ventura, CA rather than Carpinteria, CA.
being off (Score:2)
Does it not work for anyone else? (Score:2)
Re: (Score:2)
<AOL>Me Too</AOL>
Tried using Firefox (2.0.0.9) and IE (7.0.5730.11) and got the same results. My house marker moved about 75 feet, but snapped back immediately. Close browser, open and search for the same address and it's still at the original location.
Ah well...
Re: (Score:2)
I now pronounce this Damn Spiffy!
How about crowdsourcing directions (Score:2)
What say you, Slashdot?
Re: (Score:2)
Except I use Google Maps to measure bike commute distances, so 'optimal route' for me isn't the same as 'optimal route' for a car or truck driver.
Re: (Score:2)
Re:Crap. (Score:5, Interesting)
Re: (Score:3, Interesting)
- RG>
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)