On the Timing of Apple’s Map Switch | Hacker News Reader

On the Timing of Apple’s Map Switch

daringfireball.net

35 pointsjoshus57 comments

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.

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.

It's not position data that Apple needs, it's bug reports. Regardless of how much work you put into map data, at some point the only way to know if it's correct or not is to go there and see what things look like on the ground. Crowd sourcing is the only way to get to the quality level of Google Maps.

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 catch up. You're talking about volunteer crowdsourced data versus a company that literally has a fleet of cars and employees out scouring the world and mapping it in detail. Unless Apple decides to make a major play in mapping - not just in iOS, but in building a first-class mapping product to compete with Google Maps - I doubt that they're ever going to have data that'll match peoples' friends' Android-based mapping experiences, and that's going to leave a lot of people very sour on what is supposed to be the premium-brand smartphone.

I have a number of friends and acquaintences who have either already switched or are soon switching from their iPhone 4 to some Android-flavored phone specifically because they're so disappointed with the iPhone 5 + iOS6 combination, and no longer feel like it's the best option on the market. Pissing people off and losing them as customers is a pretty terrible price to pay for wanting to gather bug reports on location data.

"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.

...at which point Google will be two years ahead of them. Mapping is Google's wheelhouse; it's an afterthought for Apple. I wouldn't bet on the hardware company ever beating the data company at mapping. Apple would have to become a fundamentally different company to seriously challenge Google's mapping offering.

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.

I'm not making any claims about Apple catching up to "searching" - That truly is a field in which google has demonstrated they have unique, and proprietary advantages over the competition. They have deep research and superior search algorithms, and I would say only Bing can give them competition, and I don't believe Apple will ever be able to provide similar capability when it it comes to that.

But the elements of mapping that are not related to search, that is, routing, displaying tiles, walking directions, cycling directions, turn-by-turn, aerial-view, points of interests - these are all elements that internet data is less useful, and where being able to invest billions of dollars in acquiring suitable cartographic information (as Nokia did) should be sufficient to put together a world class map environment for the customer. My perspective on the challenges is mostly informed through several hours of reading historic postings here: http://blog.telemapics.com/?p=399

In all likelihood, after several billions of dollars invested, Apple 2014 will equal Google 2012 in the areas I just described (mapping sans search).

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).

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.

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.

Or they could have tried not bringing a water gun to a thermonuclear war.

I think Apple doesn't really care at this point where the iPhone is essentially a cash cow five years after launch with 600K apps in the app store. Incremental improvements are enough to sell millions on launch and later with many people guaranteed to buy it regardless of how many flaws it has.

They could have rolled it out as a standalone app first, with the clear goal being that they want users to try it out and report issues. There's no reason the first version had to replace the google-backed one.

I'm sure there'd be thousands of people willing to give it a go, and people would be much more forgiving of errors if it wasn't the only option.

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.

I'd say it was, or at least as ready as it could have been. Problem is that everyone is behind Google on Maps. They're on Mars, everyone else is playing with baking soda and vinegar at Cape Canaveral.

Apple would have to acquiesce to Google on a lot of fronts to get better maps features (that are a major improvement in many places), and get stuck working on something that could never compare in perpetuity. Launch or die, right?

> Technically, they could roll such a thing out in a 6.1 or 6.2 update, but major changes — and I think everybody can agree this has been a major change, for users and app developers alike — should be delivered only in major new OS updates.

Well, I personally can't wait for OS XI, since by that criteria, Apple apparently hasn't released a major OS update since OS X yet.

Version numbers schmersion numbers. Ship it when it's ready.

(I understand that you’re probably being somewhat sarcastic, but for the sake of semantics…)

It’s pretty clear by the amount of change (and Apple’s marketing efforts) that each OS 10.#.0 release is a "major" OS update, on par with the relative significance of each iOS #.0 release.

But Gruber may have over-generalized here. Indeed, Apple has added major new features in iOS .# updates, e.g. AirPlay and AirPrint in 4.2, or the Personal Hotspot feature in 4.3.

What they haven’t done is change key functionality in a point update, let alone introduce regressions. (In fact, before this incident, I’m struggling to recall Apple ever removing/crippling existing functionality.) If they were going to perform a full Maps switch-out, maybe they felt that now would be the best time, if only because there is so much positive press out there to temper the backlash.

Edit: Regarding removed features, I was mostly thinking of Apple’s record with iOS. As gurkendoktor points out, they have certainly removed ("simplified") functionality in Mac OS X, often creating great frustration.

> In fact, before this incident, I’m struggling to recall Apple ever removing/crippling existing functionality

Mac OS X has had plenty of that. You buy a Mac Mini with a remote so you can use Front Row, and bam, Front Row's gone. But that is probably why Mac users are a lot more careful with updating their systems than iOS users are/were.

My tongue is firmly in cheek, yes. I'm poking fun at the silliness of attaching some holy meaning to version numbers.

My point is that version numbers for end-user software are borderline meaningless; they're marketing fodder. Just look at Chrome and Firefox, for example. Firefox didn't suddenly start improving 10x faster than it was before, but they changed their versioning scheme in a marketing play to compete with Chrome's ever-inflating version, as the casual user assumes that Chrome 37 must be much better than Firefox 6 because its major version is so much bigger.

Apple could either ship a regression in a new major, or ship a less broken regression in a new minor. In either case, there's going to be a regression. Regressions are unfortunate anytime they have to happen, but the assertion that you could only replace Google Maps with an in-house solution in a major is just silly post-hoc justification. You can't avoid disappointing users except by not shipping a regression at all, but you can lessen it by shipping a less-broken product.

As an aside, I can think of at least one other major breaking change/regression in a point release - the total swap in functionality of the lock/mute switch in iOS 4.2. There are also things like dropping support for older devices in point releases (4.3 dropped support for the iPhone 3G) that make the argument look even weaker; device deprecation is generally the sort of thing that is sacredly reserved for major releases, yet Apple seems to have had no problem doing it in a point release, and we're supposed to believe that yet somehow Apple wasn't willing to replace an app in a point release just because it isn't a major?

Problem with this analysis is that if the replacing maps on iOS6 was always going to be this year then it just compounds the culpability of Apple for shipping it half baked with no fallback. ie. in trying to assuage the criticism of Apple for shipping maps too early (what I read as the main point of his piece here), Gruber only begs the question of why they weren't better prepared for it.

Maps are hard.

The original Google Maps was pretty bad. Mapquest beat it much of the time. There was one road in my hometown, and none in many countries.

Google devoted enormous resources to acquiring data (petabytes worth), processing it, correcting it, to give us the product we have today.

Even though Apple acquired some neat companies that made data look good, they obviously underestimated how much data was required to make it useful. No amount of QC could have prepared them, frankly.

The upside is that Google has been fairly closed with their map products. They're miles ahead of Apple, who will need every partner they can find to close the gap. Hopefully it will give the data providers a bit of leverage to get refinements back out so it can be used in endeavours other than showing where the nearest pizza joint is.

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?

Any big feature change makes for a good slide or two in the keynotes. They get once or twice a year to make a big impression on the press - dribbling out updates to core apps in the meantime would remove the buzz.

It's certainly made a big impression, though perhaps not the kind they were after. Revealing something in a keynote talk leads to more scrutiny so making a big deal out of something that isn't ready doesn't seem wise to me.

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.

Wouldn't moving from Google to Apple maps be adding the new OS features rather than removing them?

That is what I said. My point being removing/adding Apple Maps isn't the hard part, it was merely to point out that there is more to the mapping api's than an application. And if legally Apple cannot do turn by turn using Google apis, then they cannot provide one to one compatible api's with both. So they just dropped the Google api backended MapKit.

Map data will certainly be updated constantly, not even requiring an update to the app. Any time there's a road closure or the map changes in some way, it needs to be reflected live. So I wouldn't be surprised if other aspects of the map are updated server-side as well.

Ok, can someone help me? I’m quite confused. Apple’s maps have undeniable problems. Pretty big ones in some cases.

But are they really unusable? From my perspective it seems as though Apple’s maps are merely inferior, not unusable. Some aspects are even better than they were before (on iOS), some (very few) are better than Google Maps (traffic info in Germany is quite impressive).

Everyone here seems to take it as self-evident that the maps are unusable. I don’t get that. To me it looks like a slight (again, depending on use case) regression in the short term.

I also don’t get the deification of Google Maps. I remember they weren’t that great in Germany only two or so years ago. I still encounter mistakes regularly. Google Maps are awesome despite those flaws.

Apple's maps are really good. Google's maps are much better, generally, though.

There are places where Apple has an advantage over Google, but Google has been pouring resources into Maps for a long time.

Garmin, TomTom, Navteq, etc. won't be nearly as good as Apple or Google because they aggregate sources.

They aren't unusable, and are better than if you rely on a dedicated GPS. But they can still be improved. Apple doesn't seem to be weighting proximity for search as effectively as Google (unsurprisingly).

Honestly, Apple is the new #2, but they've got a mountain ahead.

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.

> rather than continue to pay licensing fees until they got it right

It's not like you just buy the data at Wal-Mart -- you have to negotiate licensing terms with Google and the terms that Google wanted were, reportedly, onerous.

Gee, that almost sounds like something worth considering before starting a thermonuclear war.

That's pretty much what the entire article is about, no?

It's about the timing of the decision, not the underlying rationale for the decision to move away from Google Maps at the cost of user experience.

Gruber says only:

>If Apple had stuck with Google Maps for another year they would have been forced to renegotiate with Google in a situation where both sides at the table would know that Apple ... had to agree to whatever terms Google demanded to extend the deal

Does Apple somehow have less leverage in the past? They can always use App Store gatekeeping as leverage. Perhaps Gruber is implying that Apple's antagonism towards Google's ally Samsung means it will receive unfavorable terms?

Even if Apple has to pay a bit more one would think it would be worth it given that Apple has lots of cash and the risk of shipping a poor maps implementation is losing further marketshare to Android.

If I'm reading this right, does that mean when the contract expires, Google stops serving map tiles to iOS clients using the Google Maps client?

That would mean the iPhone, the iPhone 3G, and the original iPad (none of which support iOS 6) will lose all mapping ability starting next year. Unless, of course, Apple backports their maps app to iOS 5 (and even that wouldn't help original iPhone users--if there are any left).

You're reading Gruber's assumptions correctly. Whether those assumptions are accurate or not is another matter entirely.

I can't imagine apple would have ever negotiated any deal where their maps simply stopped working when a contract expires.

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...

This still doesn't explain why they didn't ship it in iOS7 then. If it wasn't ready and they only want to ship such things in major versions then hold it till the next major version. It wasn't ready. And bug reports aren't the solution to get it ready either. They needed better expertise and better quality data to start with.

How does it not explain it? Maps deal expires 6/13; iOS 7.0 launches 9/13. What do you ship from 6/13 to 9/13?

New maps deal to get you up to 9/13.

Remember he usually calls half-cooked products or apps as "turd"s.

Who cares about the timing? The maps experience, even with turn-by-turn, feels worse than what I had in iOS 5. I hate it when companies deliberately make things worse.

really ? I'm liking the turn by turn directions in ios6 a lot and to me the update was worth it just for that. I still have the mobile google maps on there if I need to search anything.

Apple Maps feel very snappy and I like the Yelp tie-in.

But I have never used the transit options so I can't say I'm missing it. If I had to use it I figure I could just jump over to web based google maps.

Curious to know what don't you like about the new apple maps that can't be solved by the web based google maps ?

To be fair I've only used it for about a week-and-a-half now. If you give it a specific address it works pretty well for the most part. One thing I love compared to my Galaxy Nexus is that under the lock screen the next direction will turn the screen on.

Another thing that works better is integration with Siri. I felt that on Android there were a few more hoops to go through to do things like "drive to home" or "drive to Bob's Chicken Shack".

The things I don't like are the lack of detail on the maps currently. I don't use the satellite layer so it looks really sparse and I feel like I'm driving to big abstract blobs.

To be honest this is pretty good for a first try. Getting native turn-by-turn that works 95% of the time is pretty impressive.

While I'm not a fan of the new iOS maps, I'm reasonably happy that Apple chose to break the relationship with Google rather than throw my privacy under a bus to get a contract extension.

Blah, blah, blah, blah, blah, Gruber. IOS6 Maps still suck balls.

tl;dr : Apple screws up. The totally-oh-so-unbiased Gruber defends them anyway. For the rest of us who don't give a rat's ass, life continues.

Reply on news.ycombinator.com