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

On the Timing of Apple’s Map Switch

daringfireball.net

Post reply on HN
Print view View original

21–30 of 58 posts

Re: On the Timing of Apple’s Map Switch

#21

Post by recoiledsnake »

ghshephard wrote: post #18
arpit wrote: post #2

I dont get why Maps releases are tied to OS releases. If the Maps api to other apps doesn't change, why can't maps be updated as other apps are?

Maps are operating system features that any application can rely on being present. Their are OS APIs that require that the mapping infrastructure be present.

Can't the APIs stay constant while the supplied data/backend gets changed? After all, isn't that the whole point of having a public API so that the implementation can be changed without changing the calling apps?

Re: On the Timing of Apple’s Map Switch

#22

Post by mitchty »

rphlx wrote: post #13
ryanb wrote: post #8

Earlier quoted context omitted.

His main suggestion is that Apple couldn't get enough data to make it good without launching the app, and they had to go ahead and release it at some point. The software works quite well - it's the dataset that's inferior.

Apple could have rolled out their map-app gradually to a few hundred thousand "specially selected" beta users before totally dropping Google Maps in iOS. Given Apple's religious following - many users would have probably paid to participate. It was unnecessarily risky and basically a product-management-screwup to force-convert everyone simultaneously.

They did that in the iOS 6 beta, the users were known as developers. And yes, we do pay $99/year to be able to test things like this. The forums there listed the same things people here are complaining about. It was all said on those forums, we were just under NDA not to talk about it.

Which leads me to believe this was an upper management push and they are willing to deal with the PR fallout. beta 1 maps was really really bad, gm is workable. It has improved a ton during the betas, and that was a 2 month-ish spread. I'll give Apple a few months to clean up the data, it makes Core Location way more useful than when it was tied to Googles' api. Note this is for things like the vector road parts of the map, you can basically have your location clamped to the road. Sometimes gps signals are... erratic when driving in cities, this helps things out immensely. But we'll see if it is another ping or not.

Re: On the Timing of Apple’s Map Switch

#23

Post by recoiledsnake »

ryanb wrote: post #8
swang wrote: post #5

Gruber essentially tiptoes around the main problem. The iOS 6 map app was released before it was polished up to what people expect from Apple. "Oh Apple had to release it now or else Apple would have to release in a non major release!" Are you kidding me? The alternative being to release a terrible version that tarnishes their brand, that is way better alternative

His main suggestion is that Apple couldn't get enough data to make it good without launching the app, and they had to go ahead and release it at some point. The software works quite well - it's the dataset that's inferior.

They could've tried licensing far better map data from Navteq/Nokia like Amazon recently did for the Kindles by using some of their 100 billion dollar stash. Amazon did not try to roll out its own maps with half baked data. Whether Nokia would have been willing to license to Apple is a different question.

Re: On the Timing of Apple’s Map Switch

#24

Post by mitchty »

recoiledsnake wrote: post #21
ghshephard wrote: post #18

Earlier quoted context omitted.

Maps are operating system features that any application can rely on being present. Their are OS APIs that require that the mapping infrastructure be present.

Can't the APIs stay constant while the supplied data/backend gets changed? After all, isn't that the whole point of having a public API so that the implementation can be changed without changing the calling apps?

They could, but Core Location in iOS 6 uses the backend data for things like getting your car route. If I recall correctly Google doesn't let you use their api for things like that.

Basically removing the Maps app is one thing, removing new OS features that use the same frameworks is another.

Re: On the Timing of Apple’s Map Switch

#25

Post by jkubicek »

cheald wrote: post #11
ryanb wrote: post #8

Earlier quoted context omitted.

His main suggestion is that Apple couldn't get enough data to make it good without launching the app, and they had to go ahead and release it at some point. The software works quite well - it's the dataset that's inferior.

Apple's been keeping logs of iPhone users' positions for ages. And they could have just sucked data out of user behaviors of the existing mapping solution if they needed something in a map context. This smells of revisionism.

No they couldn't have. The "logs of iPhone users' positions" you refer to were stored on the device and never sent to Apple. It's highly unlikely they could have ever been sent to Apple anyway both from a legal perspective and a PR perspective (considering what a mess it was when they didn't send logs off users's phones).

Re: On the Timing of Apple’s Map Switch

#26

Post by brianobush »

ghshephard wrote: post #19
cheald wrote: post #16

Earlier quoted context omitted.

Okay, I can buy that. But, that basically means that Apple has drastically regressed the user experience in order to force users to help them build a product so that one day it might be able to compete with the product that they were already using for years, and suffer through cruddy data in the meantime. And they expected people to be okay with this? Google is so far ahead on mapping that I doubt Apple will ever cat…

"Google is so far ahead on mapping that I doubt Apple will ever catch up. " We're not talking rocket science here. If Apple wants to catch up to Google on mapping, they just have to invest time, money, and attention. We're talking at most two years for Apple to get where Google is today.

Google has far more data to enhance its maps. Does Apple have the internet lying around to analyze?

Re: On the Timing of Apple’s Map Switch

#27

Post by wizzard »

swang wrote: post #5

Gruber essentially tiptoes around the main problem. The iOS 6 map app was released before it was polished up to what people expect from Apple. "Oh Apple had to release it now or else Apple would have to release in a non major release!" Are you kidding me? The alternative being to release a terrible version that tarnishes their brand, that is way better alternative

Gruber is trying to make it out to be this chess game of when to release the new map app, but never addresses the main issue: Was it ready?

I guess if you're crowdsourcing map data the only way to start improving it is to get it out there. But don't act like it's all mind games and strategery and say that Google knew that Apple knew that Google knew that Apple was going to drop Google for maps. Reading the past couple of posts was exhausting and doesn't comfort me at all when my phone tells me to drive into a lake.

Re: On the Timing of Apple’s Map Switch

#28

Post by mcantelon »

ghshephard wrote: post #17

This is one of those articles, that after you read it, you face palm, and go, "Of course." I feel stupid not having come to the same assessment after hearing about when the dates were set.

What he doesn't answer is why they chose to ship a half-backed map implementation rather than continue to pay licensing fees until they got it right. It's not like Apple's starving for cash.

Re: On the Timing of Apple’s Map Switch

#29

Post by fpgeek »

ejdyksen wrote: post #10
philwebster wrote: post #7

Earlier quoted context omitted.

Check the footnote. Devices using still Google Maps will most likely function without a change for a few years, but Apple won't be allowed to sell/activate more phones using Google Maps.

Yeah I just saw that. I'm sure the deal takes into account legacy devices, but I wonder how long that extended support would last.

I suspect it would last for quite some time beyond the officially supported window (possibly until the vast majority of those devices are no longer using Google's map data either via upgrade or retirement). Absent Apple or someone else unreasonably taking advantage of the "legacy device loophole", Google doesn't have much incentive (and some disincentive from the potential PR fallout) to shut it down even no matter what the contract allows. I'm sure sometime after that support windows they'll want to roll out something where it's easier to break existing clients than to keep them working, but until then...

Re: On the Timing of Apple’s Map Switch

#30

Post by fpgeek »

brianobush wrote: post #26
ghshephard wrote: post #19

Earlier quoted context omitted.

"Google is so far ahead on mapping that I doubt Apple will ever catch up. " We're not talking rocket science here. If Apple wants to catch up to Google on mapping, they just have to invest time, money, and attention. We're talking at most two years for Apple to get where Google is today.

Google has far more data to enhance its maps. Does Apple have the internet lying around to analyze?

Very, very true.

Think of web pages with POI data (store and restaurant locators are probably the tip of that iceberg), a long history of location-related searches including data from "front door" web searches, data on how the results of those searches were used (which Google uses to great effect in other places) and so on.

Post reply on HN