Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google Maps' Moat

justinobeirne.com

Post reply on HN
Print view View original

161–170 of 619 posts

Re: Google Maps' Moat

#161

Post by ancarda »

dvt wrote: post #31

I accidentally used Apple Maps once as I was going to a (new) doctor's office. I realized I was using Apple maps halfway through my drive and I joked with myself that it wouldn't bring me to the right place. I was heading to a fairly large medical complex in Orange County (CA) and I sort-of had an idea of where I was going anyway, but boy, was I wrong. According to Apple maps, the doctor's office was literally in the…

I was using Apple Maps a few weeks ago, and wanted to get something to eat. I saw a place called “Shake Shack” and decided to go there. Not only did it not exist, I checked online, there has never been a Shake Shack in my town, there’s only a handful in my country.

How Apple Maps got that wrong, I honestly don’t know. Where did the data come from, and why is it that incorrect?

I reported it, and it’s since been removed, but I can’t do that for everything in my town, and there’s so much missing; I can’t report that, or edit the map like you can on OpenStreetMap.

Honestly, why does Apple Maps exist? Either it needs to be like Google Maps (and just work), or be like OpenStreetMap and let Apple fans edit the map and get more involved.

I gave them a lot of patience and understanding years ago (mapping the world is hardly an easy task), but my patience has run out. It’s been a shit show from day one and I honestly don’t see that changing anytime soon.

Re: Google Maps' Moat

#162

Post by GeneralMayhem »

irrational wrote: post #133

One thing I really appreciate about Google Maps is how accurate the times are. There have been numerous times where google maps says is will take x minutes to get there and I think "Nah, I'll be able to get there faster than that." Nope, I've never had google maps be off by more than a minute or so. I think it must take into account not just road conditions and traffic, but also how much I am likely to go over the sp…

The crazy thing to me is that it's pretty decent at estimating even in dynamic traffic conditions. Consider US101 in the Bay area at rush hour - if you leave San Jose around 6pm, you'll drive straight into traffic, but by the time you get to SF it'll be significantly lighter. A lesser algorithm would give you a time prediction based on current traffic speeds at all points along your path, which would give you a wildly pessimistic estimate when you leave and only converge when you get close to your destination. But Maps has enough data to know that at that time of day traffic is trending better, and by how much.

Re: Google Maps' Moat

#163

Post by angch »

londons_explore wrote: post #38
EGreg wrote: post #32

These building footprints, complete with height detail, are algorithmically created by taking aerial imagery and using computer vision techniques to render the building shapes. I could have guessed that before the punchline. And it's not that difficult to add for a company with tens of billions in cash. What are you doing, Apple? Fix your OS and Maps!

You need aerial imagery from different angles. Thats surprisingly hard to buy from imagery providers, because most of them only want to point their cameras directly downwards.

They fly their own planes these days.

https://blog.google/topics/inside-google/google-earths-incre...

Too bad it's a video. Fast forward about 2 minutes.

Or: https://youtu.be/suo_aUTUpps?t=121

Fly plane. Zig zag. 5 cameras, then photogrammetry. StreetView, from planes.

Re: Google Maps' Moat

#164

Post by daveheq »

Google Maps downloading building structure and parked truck boxes is WAY too much to download, especially over my limited cell data plan... That's pretty ridiculous, I have almost no need for that.

Re: Google Maps' Moat

#165

Post by sidcool »

sytelus wrote: post #142

Many people don't realize why Google maps are so much better than anything out there. I would actually give lot of credit to Marissa Mayer who suddenly became in charge of maps from otherwise more higher position. Thanks to her influence in Google SLT and ability to make impressive arguments, she was able to make a case for maps as core pillar in Google's offering and consequently obtain huge investment and large tal…

That's a very good point. Many a times engineers have an excellent use case/rationale for investing in a piece of technology, but they are unable to convert it into the language that business would understand. Marissa and many others have mastered that art.

Re: Google Maps' Moat

#166

Post by derekp7 »

leereeves wrote: post #89
dvt wrote: post #31

I accidentally used Apple Maps once as I was going to a (new) doctor's office. I realized I was using Apple maps halfway through my drive and I joked with myself that it wouldn't bring me to the right place. I was heading to a fairly large medical complex in Orange County (CA) and I sort-of had an idea of where I was going anyway, but boy, was I wrong. According to Apple maps, the doctor's office was literally in the…

Google Maps has sent me astray a few times as well. Strangely, I've had the worst experience with GM in the Bay area. Going to the Golden Gate Recreation Area, Google Maps sent me to a trailhead parking lot instead of the visitor's center. And for a small business near the Bay Model, GM directions stopped about half a mile away, and the map pin was on top of the Bay Model building.

The problem I have with GM is that it will often send me past my destination, then have me do a U-turn, in an apparent attempt to make sure the building is on the right side of the street. Even if there is no center divider and a left turn lane. Even if there is a traffic-light-controlled intersection to get me into the parking lot of the building.

Re: Google Maps' Moat

#167

Post by djsumdog »

jgh wrote: post #150

Honestly I _hate_ Google Maps, but it seems like the best thing available. I wish they'd spend less time on building outlines and more time on making the UI not totally suck. Someone should make a maps startup that isn't awful (note: these examples are from the web app, which I prefer, because the mobile app is infuriating to use except for driving directions) For example if I google an address and click the link to…

I really agree with you. I've tried using Here Maps and Waze. Waze is just too much ad/game crap and Here works nice for navigation, but tends to crash the OS or disable mobile data randomly.

With most of this article I was like "So what?" So what if Google has intelligent building vectors. It's UI is shit. You can't go back and fourth between different routes easily. You can't easily add waypoints. You can't do a lot of the stuff you could do with traditional maps.

And the worst thing, half the time you can't even see the names of streets:

https://www.justinobeirne.com/what-happened-to-google-maps/

This is especially difficult if you're walking around and just want the name of a god damn street. You gotta zoom in, jiggle things around. Fuck aerial views of buildings. When I'm walking I need to match up actual street signs to find my way around.

Open Street Maps is actually pretty nice, although the views are extremely busy. It can be a bit of an overload.

Re: Google Maps' Moat

#168

Post by djsumdog »

spaceflunky wrote: post #45
ajkjk wrote: post #9

Things like this make me think that it's a shame humanity is duplicating all this work instead of collaborating to do it once. One of the awkward failures of capitalism, I guess. At least when your product is something physical you end up with twice as many outputs. When it's just data you just do the same work twice.

This is the wrong way to think about things. Competition is good. If it weren't for the need to compete, would Google have spent billions on innovations like Street View, Satellite imagery, Google Earth, etc. Or would they have just said, "MapQuest has won the internet maps game, let's just let MapQuest do their own thing. No use in doing the same work twice,"

What if all the data was public and shared and different people could build their own representations. What if everyone collaborated?

I disagree with the competition is always good. There are many cases where you are duplicating work. The best standard rarely wins (PCI vs MCA .. MCA was superior in many ways technologically .. was even hot pluggable, but PCI had better licensing).

We get into a lot of repetition and the 'best' product rarely wins. It's more about marketing your product in a way people desire it. Example: Beats Audio Headphones.

Re: Google Maps' Moat

#169

Post by sonar_un »

irrational wrote: post #133

One thing I really appreciate about Google Maps is how accurate the times are. There have been numerous times where google maps says is will take x minutes to get there and I think "Nah, I'll be able to get there faster than that." Nope, I've never had google maps be off by more than a minute or so. I think it must take into account not just road conditions and traffic, but also how much I am likely to go over the sp…

Their acquisition of Waze really gave them the edge here. Waze has been getting better and better at predicting times. I find that in most situations it is rarely off by more than a minute or two.

Re: Google Maps' Moat

#170

Post by SEJeff »

on_and_off wrote: post #114
stefanpie wrote: post #23

Earlier quoted context omitted.

I've heard stories, guesses, and claims that the current traffic data shown on google maps is potentially aggregated from smartphone GPS and accelerometer readings on android phones or though their Google apps. It's crazy and amazing to just think about it.

I always thought this was officially a feature of maps ? I worked for a couple of years in a valley with only 2 access points. Quite often, when it is time to go back home, the trafic would be jammed. I activated Google Now (or at least I think it was already called that way ?) and without fail it would warn me that there is no point in leaving the building right now since everything was jammed. The same info was vis…

And no doubt they pull from obviously free public sources such as sigalert as well. More data to feed to their hungry compute engine.
Post reply on HN