Hacker News Reader

Comment by enneff | Hacker News Reader

(Disclosure: I work for Google, but not in anything to do with maps. These opinions are my own.)

I've been to this national park before. It's a beautiful place, but not one you want to get stuck in unprepared. The temperature estimates given by the police are accurate. It gets damn hot out there. The roads are gravel at best, and often just packed dirt or sand. You can very easily get bogged, if you don't know what you are doing. I got to a point where the road became unsuitable for my car, and I had to drive backwards for nearly 100 metres to get out as there was nowhere to turn. I would be concerned for inexperienced drivers with two-wheel-drive vehicles.

Map data is serious business, and this particular case is an egregious error.

View on HN

The thing I really don't understand is this seems like such a simple thing to test! Search for X in our data set vs. competitor data set: flag for review if we are not within 1km radius. Don't release until the number and scope of flags hasn't been reduced to an acceptable margin of error (don't even have to test the full data set, just need a good enough statistician to help you figure out how bad your overall data is based on your sample tested data).

OpenStreetMap has been tested like this http://povesham.wordpress.com/2009/07/15/openstreetmap-and-o... http://www.ucl.ac.uk/~ucfamha/OSM%20data%20analysis%20070808...

"Apple steals Google Maps data then bans them"...news at eleven.

He's not advocating outright plagiarism. Checking your own results against someone else's is perfectly acceptable and recommended.

Not for the media. This would be a media gold mine.

Depends on the definition of "someone else". Check the terms and conditions of all big/high quality map data providers.

Not if the terms of use of the "someone else" explicitly prohibit such activity. This is particularly common with local information providers (eg business/event listings).

2 problems here, first, can you use competitors data for comparison and making your data better? Second, calculating difference is very difficult and size of difference could be overwhelming. But first issue is the main problem.

1. Yes.

2. Difficult how? You don't have to perform it on items that can't be automatically searched for. Then you do subtraction, square, check if you're over a kilometer.

I've been to a lot of similar places in Australia, though I'm not sure about this one.

The thing I don't get is how/why people keep going until they get stuck. Is there not a point where you think 'this doesn't look like the road to Mildura anymore' ?

Places like that national park should only be entered in a 4x4 and you need to be carrying a water supply, spare fuel and various other bits of equipment.

I know how having experienced it myself. In my case I was almost stuck in the French Alps using Google Maps on the iPhone. It had taken me off the autoroute onto a national highway that turned into a country road that eventually turned to a dirt road. The rental car's built-in GPS (which I had switched off earlier because it had given me bad directions before) actually came through this time. On the way back I used the directions one of the locals gave me to use the newly opened autoroute instead.

I used to laugh when I heard stories like this in the news. But in my case it's dark, we're all tired, in an unfamiliar location, and I'm busy looking for sign/landmark/anything to get me pointed in the right direction. And, it managed to snow (this was in May). So when you're doubting yourself and not sure what decisions to make guess who doesn't have that problem? The little 4-inch box sitting on your dashboard who is speaking with absolute authority that you need to make a right in 400-meters.

And that is why people get stuck in the places they do sometimes.

Yeah, sorry, I'll trump your tale. IOS maps got me chased out of the Baikonaur cosmodrome by the Russian Army and "special police" at 4 in the morning. The roads in Kaz are dirt tracks for thousands of miles, frequently, so getting lost is easy - particularly when the road vectors have a one-point-per-10km res.

Fortunately, they thought it was hilarious that an Englishman in an 80's Merc should rock up at their launch facility looking rather tired and confused, otherwise I wouldn't be telling you this.

Yeah, it's a very unfortunate error.

A couple of years ago, Google Maps placed Sunrise, FL in Sarasota, FL -- a 200 mile error, so somewhat larger than this one. But not as dangerous.

All of which is true, but it leaves me wondering if Vic Police have informed Apple - seems like a far more effective way to save lives than issuing a statement that will be read by pretty much no-one...

I mean as an absolute first step, they should as a minimum report the problem by using the -in-app error notification system...

From TFA:

> Police have contacted Apple in relation to the issue and hope the matter is rectified promptly to ensure the safety of motorists travelling to Mildura.

Not to mention that a LOT of people have just read about the issue. Also, not to mention it, but that site is really targetted towards the Australian news media.

Oh, is that what that .au means at the end of the URL?

No. You said it would be "read by pretty much noone". I pointed out that the Australian news media monitor the site.

Pretty stupid comment, really. No need to get uppity, I'm an Australian also.

My point being that your comment was the one that was getting uppity. What did you think that it added to the conversation? Yes, I did miss the line in the article saying that Vic Police had reported the problem to Apple. Fair enough, I made a mistake, and the first response called me out for it. You then decided to pile on with a zero content response. I mean seriously, your comment added exactly zero to the conversation. There's no point coming off all hurt by my sarcastic rejoinder - if you want to throw stones, you have to expect them to be thrown back...

Actually, I added to it. Sorry you took it the wrong way. I merely mentioned that HN had picked it up, and that the Australian media monitors the website. I think it might have been the "not to mention it" bits I put into the post that might have been misconstrued... I guess I didn't realize they'd be seen as aggressive. That was not my intent!

None of these points had been brought up before, so I'd hardly call that a "zero content response".

Peace, I wasn't throwing stones. And I don't see how it's helpful to any conversation on HN...

The manager of Apple Maps almost certainly got this clipping on their desk this morning with a memo of FIX THIS NOW. Seems pretty effective to me.

Every time I turn my Garmin GPS it warns me that maps might not be correct, so it's not just Apple.

That said, Apple should have written a big fat check to a company with much better data. (I still think Apple needed their own mapping and apparently maps improve as more users use them.)

That and they are using Tom Tom data for navigation. There have been 100's of stories about sat navs taking people on wrong turns, but when it's related to an iPhone you'r going to get more hits!

https://www.google.co.uk/search?q=sat+nav+wrong+turn+site%3A...

Agree, mapping is not something you can fumble your way through as an amateur.

It isn't a case of buying a better dataset, Apple needs to use the good data they already have correctly. See my comment below; in this case Apple is giving co-ordinates for the center of the Shire Of Mildura (a local government region), instead of the co-ordinates for the town of Mildura. Both co-ordinates are in the free GeoNames dataset that Apple already has access to.

Agreed, the search is worse than the actual data.

When iOS 6 was first available, I hit a link to a restaurant on a web site that opened Maps to the restaurant's supposed location. The location it opened actually had the restaurant's exact, correct, latitude and longitude. But Maps decided it knew better and showed me where it thought the restaurant with that name was, which was about a half mile off. If I removed the name and left the lat/lon then it was correct.

Or just search for "emergency room" and notice how, even today, it still shows tons of medical stuff but no actual emergency rooms.

Jobs is turning in his grave and would have had none of this. Little details collect like snowflakes and create avalanches if not put in check.

"Jobs is turning in his grave and..."

He was cremated and he wasn't above screwing up.

But when he did screw up, he yelled at people until it was fixed - instead of sitting on his thumbs while desanguinating from a thousand paper cuts.

What makes you think Tim Cook is sitting on his thumbs. Did he not fire two top-level execs over the maps fiasco?

Looks like that's effectively solved their maps problem.

If you believe his actions are ineffective, you can criticise them, but you can't say that he's not doing anything.

I didn't say he wasn't doing anything. He's certainly doing something. Maps still don't work properly.

Reply on news.ycombinator.com