Crucial insight, there. From the perspective of an observer who has no special affection for Apple, it's surprising that a company of that competence would succumb to such a basic process error.

1) "Social Experiences" - Ping failed where... who succeeded?

2) "Data Harvesting and Search" - These are two terms that are either exactly the same, or extremely different. Neither points at a clear criticism. You realize Apple pioneered desktop search, right?

3) "Data synchronization" - Where's the problem, exactly? Is there any company other than Dropbox that can claim the crown for this? Do you know how hard this is? Who is third place after Apple's second?

4) "and Security" - Again. Huh?

5) I'm terribly sorry you had a software failure and only had a single solitary backup of 8GB of your vital that was difficult to retrieve when Apple was clearly spiking 8 petabytes of traffic in one day. In this clearly entirely hypothetical scenario you've invented.

I hate defending the only company that finally brought UNIX to the desktop just because I happen to wear designer glasses.

Let me give another example where iCloud fails: iTunes in the Cloud and iTunes Match. I'm regularly unable to download or stream certain songs for days. I can play the rest of the album fine, but one or two songs, you cannot play nor download.

Sometimes, I cannot access my matched music at all. iTunes will do an iTunes Match update, which fails for some reason, and then suddenly all songs are greyed out. Usually, a match update will only succeed after a few hours of trying. During that time I can only play music that is locally available.

Match is a big mess and I will probably not renew it after a year.

Facebook + Spotify. I actually see people sharing the music they listen to. Not to mention, it is indeed very possible to fail without someone else succeeding - i.e., launching a product nobody ever wanted (say, Ping).

> "You realize Apple pioneered desktop search, right?"

And now the competition is much better than they are. Your point?

> "Where's the problem, exactly?"

iCloud. Have you written an iCloud-enabled app recently? I was in the iCloud session at this year's WWDC, and damn that was a tense session - developers approached this session with "hopefully this shit finally works", because iCloud up until that point was broken, and still largely is.

If you look behind Apple's hyperbolic claims and actually look at the API and work with it, you will realize iCloud is pretty much loosely held together by string. It breaks often, the API is obtuse and more or less undocumented, and there are very few ways to look under the hood when it does break, even while in dev, much less production.

Apple really, really sucks at data synchronization. Here's an interesting test: take an iCloud-enabled app, go into your settings and delete the app's cloud data. You would expect this to delete your data and let you start fresh right? But nope, none of your devices associated with this Apple account can ever use iCloud with this app again.

http://www.powayiliad.com/2011/01/itunes-ping-an-apple-failu...

http://www.buzzfeed.com/scott/ping-fail-5-reasons-to-avoid-j...

Even Apple considers Ping a failure, so I'm really not certain where you are trying to go with your argument.

http://news.yahoo.com/blogs/technology-blog/apple-declares-p...

2) Apple is fairly good at doing things on the desktop, however their server work leaves much to be desired: the way they attempt to rank search results in their various online store fronts is a commonly cited example of this. I have explicitly been told by people at Apple "we are not server people" when reporting issues. (Also, "harvesting" is the collection of data, and "search" is the process of finding data you need in the massive pile you have collected: these are orthogonal concepts.)

3) There are so many issues with this that there is a small industry surrounding tools and teaching people hacks to make this work correctly; I'm kind of assuming you have sufficient context in the Apple community to know much about the specific issues, which is why I felt it was perfectly acceptable to just list "known failures" rather than having to say "this is how why this one failed".

http://www.maclife.com/article/howtos/how_solve_all_your_mob...

http://www.cultofmac.com/82971/how-to-fix-mobileme-sync-prob...

http://www.google.com/search?q=+site:discussions.apple.com+m... (12,600 results)

4) You realize who you are talking to, right? Do I really need to defend this? The sheer fact that I have been, for years, operating what is probably the largest private-sector man-in-the-middle attack ever (many tens of millions of devices worldwide, literally billions of stored signature blobs), on Apple's products, an attack that would be trivially solvable by techniques literally taught in intro-level encryption classes, should be sufficient (but of course when you do what I do you have tons of stories to tell).

5) I haven't had this happen to me, as I don't use iCloud: this has happened to numerous other people, however. Here is one example user (of many), from May, with the non-saurik names changed:

--- Day changed Tue May 29 2012 01:04:34 <xxx> this iCloud shit is getting ridiculous 01:04:59 <saurik> xxx: ? 01:05:16 <xxx> saurik: I upgraded my phone, but my iCloud backup was 7GB, i've spent the last two days straight trying to restore it 01:05:38 <xxx> (if the download is interrupted for one second, it fails, and there is no download resume or anything similar or way to download to iTunes or anything) 01:05:50 <xxx> and on my slow internet, 7GB takes about 13 hours to download 01:06:10 <saurik> ouch 01:06:20 <lmrpq> I backup everything minus the Camera Roll, so my backups stay small. If I update, I use iTunes for a one off backup/restore. 01:06:56 <xxx> yeah; i should have disabled the camera roll 01:07:33 <xxx> saurik: I don't think it fails if the screen doesn't turn off, so right now I'm sitting with it for the ~9 remaining hours on my 7th attempt tapping the screen. 01:07:35 <lmrpq> However, the backup I made before upgrading to 5.1.1 turned out to be corrupt (wtf?) so I had to restore from iCloud minus Camera Roll. 01:07:46 <xxx> lmrpq: how do you restore minus camera roll?? 01:07:50 <xxx> oh, iTunes backup 01:08:02 <xxx> i want to restore from iCloud without camera roll, for maybe ~500mb 01:08:08 <lmrpq> No, my iCloud backups don't have camera roll, so I had to do that 01:08:12 <xxx> hmm 01:08:15 <lmrpq> I was not happy. 01:08:19 <xxx> maybe i could finish setup 01:08:20 <xxx> jailbreak it 01:08:26 <xxx> remove the "setup finished" flag 01:08:32 <xxx> hook into the setup script 01:08:41 <xxx> skip parts of the download 01:08:45 <xxx> :< 01:09:25 <xxx> hmm maybe i could also go somewhere with > 10mbps internet 01:09:37 <lmrpq> Apple has a lot of money. I wish they put some of it into making stellar cloud services. 01:11:18 <saurik> xxx: what kind of device? 01:13:03 <saurik> ah, I'm going to resume 4S? (as you said iphone) 01:13:47 <xxx> yaeh 01:13:55 <saurik> I guess its time for the drinking bird --- Day changed Wed May 30 2012 00:07:33 <xxx> iCloud restore failed. 00:07:37 <xxx> :( 00:07:42 <xxx> time to start again!

The best part of the entire report (which you will note was still plaguing this user a day later, as near the end the timestamps reset for May 30th):

00:42:52 <xxx> all i really want is my chrono trigger save game :(

2) Okay. I've tried to buy large numbers of servers from Apple in the past. They really aren't server people, it's true. I was trying to buy these servers because I was working for a company whose primary business is searching through petabytes of data. Apple was and still is kicking the shit out of everyone with Spotlight.

3) I'm not saying there have never been bugs in any of Apple's data synchronization software, but Dropbox is the only company doing a better job.

4) I guess I don't realize who I'm talking to. Apple has a great reputation when it comes to malicious code. Until I finish posting this and get to researching your history, I'm going to assume you're behind jailbreakme, or similar. Thanks! Great job! Really, it's hard for me not to come off as hypocritical here, but that kind of work is incredibly valuable.

5) Eh, I have difficulty garnering sympathy for anyone complaining on IRC about ~160KB download speeds for downloading their personal backup from Apple while Apple is also distributing iOS6 binaries to at least 15 million devices or so (based on napkin math derived from here: http://allthingsd.com/20120920/usage-of-apples-ios-6-hits-st...)

You're not wrong about Apple's hubris in general. These are just bad examples.

http://www.huffingtonpost.com/2011/07/18/left-itunes-for-spo...

4) I currently host jailbreakme.com, but comex developed that specific jailbreak. I helped with Corona (5.1 exploit) and run Cydia, Substrate, etc.; I guess thinking about it more, I produced the protection fix for one of comex's JailbreakMe PDF exploits (making me somewhat relevantly related to that project).

Apple does not, in fact, have a good reputation when it comes to malicious code: they are simply sufficiently small players that people don't target them. A lot of people believe otherwise, but as far as I can tell this is because their knowledge of Apple products comes only from Apple's marketing efforts.

In fact, in 2010, reknowned security researcher Charlie Miller (who was winning Pwn2Own every year until he decided to stop attending to protest a rules change) was fuzzing PDF renderers, and found many more exploitable PDF files against Apple's Preview (30-60 failures) than in Adobe's Acrobate (only 3-10).

^ This, combined with first-hand experience with the zero-day PDF exploits from comex (where the second exploit was to the same mechanism as the first, as Apple apparently failed to fix it the first time around) are the reason I install Adobe Acrobat and deactivate Preview on my Mac: at least Adobe manages to fix the bugs that are found.

5) iOS 6 was not released in May. In fact, no iOS version was released concurrent to that reported issue, AFAIK. How is this relevant to the example I dragged up and posted? I specifically went out of my way to find an example that would not fall to simple "but the bandwidth is too much for Apple/Akamai to handle!" arguments, and you didn't even pay attention.

4) I said hello on IRC a little bit ago. I'm not going to argue the point on security when I'm clearly outclassed.

5) Whoops. I'm just being careless there. Sorry about that.

I'm really only arguing any of this because of the remark "Apple actually makes this kind of mistake often." This maps thing is a rare, but colossal fuckup from Apple. First other thing that came to mind was when they cut the original iPhone sticker price by $200 only a couple weeks after it came out.

As an example, 5.1.1 was subject to Rocky Racoon for many months throughout the beta releases of 6.0. This exploit requires physical access to an unlocked (as in, not at the lock screen: PIN code already entered) device; once the device is unlocked, you can just use it, getting access to e-mails, the address book... whatever you'd like: it isn't really a serious security hazard to also be able to jailbreak it.

I personally believe that not doing this (either caused by not finding it important, or by an endemic challenge to the action by the communication medium), and not anonymity (as many people in the "use real names" debate claim) is what causes commenting community chaotic collapse (of the form of YouTube, 4chan, and random blogs).

Maybe the format of a message is confusing? Each line is "timestamp <from> message", with the convention that people use a colon to do a targeted address; were I talking to you, the message might be "12:00:00 <saurik> blub: this is my message". HN then collapsed it to a single massive line, and I figured that the contents was sufficiently unimportant that that was better not to expand into an equally-massive vertical wall of text.

If there's one thing that Apple hates as much as a half-baked UX, it's dependency on another tech company. It seems pretty obvious to me as a casual developer and observer of Google maps over the years that global mapping is really really hard. How could Apple not have known this? I don't think even they have this much hubris.

Now we don't know why negotiations with Google for the maps API failed. Maybe Apple was overconfident, maybe Google was playing hard ball for Android's sake, maybe a bit of both. But regardless of the reason, the contract was not renewed and they had to release something. The iPhone without maps is not the iPhone.

The fact that they used an algorithmic approach to QA says more about the timelines than Apple's beliefs about the overall best approach. There was simply not time to get people on the ground to do a proper QA job. Apple's hands were tied.

On top of all that, I think Google has more business intelligence than to actually let the contract for iOS maps go. Maybe both sides were just bluffing and never thought the one party would cut the other loose, or something like that, but you'd think with a deal this big at some point someone would come back with their tail between their legs and reopen negotiations.

Of course, it's all speculation at this point anyway. It'd be nice if someone with knowledge could come in here and elucidate, though the likelihood of that is really small.

(And I know that for all its faults, Nokia is smart enough to apply human QA to their mapping system, because in my last job I worked on the Web site that those human contractors used.)

They have made a deal with Tomtom (teleatlas).

Concerning Garmin, they don't do maps but use navteq (a subsidiary of nokia). Apple should probably have bought nokia, they would have acquired the best mobile maps technology.

I'm not sure what all that deal entails, so it's hard to say whether it could have been better executed upon, other than to say it's clearly not ready yet. I find it hard to believe, however, that all of the incorrect placement of destinations and, in some cases, even the inclusion of destinations that have not existed for decades, also occur in TomTom's car navigation systems.

The author of the article is suggesting the problem is the integration of lots of other data in with the TomTom data is the problem. Not the TomTom data

This is Apple's "Vista" moment and will inevitably draw the "wouldn't have happened under Jobs" comments. How Apple recovers will form business school study material for years to come.

- try to make the deal with Google happen; Apple has so much money it's hard to understand why they couldn't buy their way into Google Maps for another year or two

- make a deal with someone else (Microsoft / Bing Maps, other map provider) while still working on their own, in-house solution

It's great that Apple decided to build its own map system; the problem is they released it when it was far from ready; Apple didn't use to present half-baked products to the world.

Nice straw man and shoving words into another's mouth.

When the product suits one's purpose, it's viable. When it doesn't, it's half baked. When the original iPhone was released, the bar for viable was much lower. Other smartphones were a mess, UX-wise. Just fixing the UX with a smaller feature set won over a mainstream audience.

Now, the bar for viable is much higher, and it carries the expectations of a Maps application that works at least as well as Google Maps for iOS.

Customer expectations change over time. Pretending that they don't for the sake of argument is either foolish or dishonest.

What insight?

Apple makes most of their stuff themselves -- or gets them from multiple sources.

Maps were an exception, as a key technology that was in the hands of a competitor.

But not because Apple could help it at the time: they made a foray into a new territory, mobile phones, and Google had all the available technology and was a non competitor at the time (the first Android phone was out a year later, Oct 2008).

It's a few years that they have already started the process of moving to something of their own. They bought some mapping companies, etc. Something forced their hands to make the switch before that was over.

In any case, you can't just jump over 8+ years of refinements that Google maps have (not to mention all the "driving endless miles for Google Street View and road info, etc).

0 - http://www.theatlantic.com/technology/archive/2012/09/how-go...

If the _best_ Google could come up with was throwing manpower at the problem then anybody else should think really carefully about their solution if it doesn't involve similar manpower.

Given that the Youtube app was also pulled, there's probably some political maneuverings going on. Google pretty much pulled the rug out from under Apple's feet (whether this was Google taking their ball and going home, or Apple's hubris we don't know). If Apple doesn't provide a coherent story in the next few weeks/months, this may actually have a bottom-line impact on the iPhone 5 and iOS6.

This just amazes me.

For the longest time there were really only two suppliers of data: Tele Atlas and NAVTEQ. Everything else wasn't worth touching b/c the quality sucked. Then Google collected their own in their little cars and stopped paying suppliers. I'm not sure why Apple didn't have the foresight to understand this was an enormous engineering effort from Google - not only collecting their own data but the whole platform itself.

I worked as a consultant for one of the two major data suppliers for 3 years rebuilding their backend. 400 years sounds like a reasonable swag.

Apple has no lack of hubris - in some things they succeed spectacularly beyond all expectations, and in others - well MobileMe, Ping, and now Maps...

We get it, they're bad.

EDIT: Frankly, I think it's pretty hilarious this article is so long and formal.

edit: on a related note, I always took google maps for granted, but now that I realize how fantastic of an effort it must take to map the world to the degree they do, I have much more respect for it, and I think it is amazing in terms of "cataloging the worlds information".

http://www.theatlantic.com/technology/archive/2012/09/how-go...

See e.g. http://blog.telemapics.com/?p=394

Thanks for the warning though

Read some of the other comments to find out how it can be interpreted.

He lays out very well that a human element is required to bring these streams of data together. And Apple is not a company with the DNA of big data.

Just algorithmic manipulation of the data is not possible or sufficient - they will need an army to integrate these streams and bring them up to par. And Apple is even less happy to deal with an Army than Google was.

Although frankly speaking, if anything 400 man-years seems very low for worldwide mapping.

edit: I... I skimmed the article the first time, but... The author actually claims that it's a data quality problem, not an algorithm problem. Huh?!

second edit: I guess my only real contention is that "data quality" is a fake idea.

Apple has had trouble with Big Data, in the parlance, but iCloud is still running. They've made some pretty dumb mistakes with their online services over the years, but they still do happen to be syncing data (if even just push notifications) to something on the order of a hundred million devices.

Catching up to where Google Maps is today will take somewhere between 10 days and 400 years. I'll bet good money that they have this nailed within a 95% tolerance before next year's new hardware model.

This is apart from MapMaker, and their dev teams, and their software improvements.

For instance, if you look at directions in India now, they're awesome, since they tell you landmarks to look for on the way [pass by * on left], at the intersections, etc. http://goo.gl/maps/b8w15 for instance.

My guess is 95% will take 3-4 years - once they accept they have a problem.

On the other hand, this actually is one of the rarer problems where throwing 10,000 contract workers from India for 5 months could solve much of the problem, especially since Google has already paved the way on the harder parts.

What I really want to see is a serious quality comparison between Apple and OSM.

edit: And the million dollar question... How much of Google's map data was correlated/verified against iPhone positioning data?

I can see how that would help reveal issues, but I don't see how it would help much in fixing them.

Fixing map data is.

It's just that it is hard and that it will take Apple a lot of time and investment to bring the mapping functionality to a better level. I'm pretty sure Apple knows that.

It's just the first iteration. Apple works that way. Bring a product or service in a first iteration and then improve from there. This is how the iPod evolved, for example. Aperture, another example.

The current maps application in iOS already has a feedback interface. This helps to improve the data.

But there are a few things which need more consideration:

* the 3d view looks ugly when looked at close to the objects. The 3d reconstruction algorithm which creates a 3d view from images is problematic.

* the angle of the data from Tomtom is for car users. Other users have less benefit: there is a lack of detail and the usage perspective renders the map in a certain way. For example here in Europe there are a lot of local public transit users. They have a hard time identifying useful informations on the current maps.

* combination with all kinds of POIs. You need to get that data and have it constantly updated. Where is a shop, when is it open, where is a museum, where are interesting views, where is a difficult road condition, ...

Probably the mapping domain is the toughest Apple has touched in years. You need a really good idea how to deal with the challenges. Personally, I think it is worth it, but it will be a lot of work (and not of machines, but also of humans) and very expensive for Apple. I'd wish they would use more of Openstreetmap and that there would be a benefit for the Openstreetmap community.

When you try to compete with an entrenched product that is feature rich and highly polished (such as google maps) you either need to differentiate your product or come to the table with an equal level of polish. Apple has failed to do this. They have made a classic "strategy tax" blunder.

This move is straight out of the playbook of the old, bad Microsoft and the old, bad IBM. It's the sort of fundamentally bad idea that a big, dumb, lumbering company makes. If there's anyone in the world that is happy about this it has to be the people on the surface team at Microsoft. Because it shows how massively vulnerable Apple is to competition now. Whatever magic Apple used to have, it seems to be gone.

Sure, all software works this way. We all know the "release early, release often" mantra. The difference is that Apple replaced functionality that many users depended upon with something completely inadequate. It'd be like Sun or IBM going in and replacing all the super specialized custom microprocessors in your server farm with experimental souped-up 386es. Sure, they're the same type of thing, and hypothetically they should serve the same function, but you don't just go out and replace completely functional stuff without at least the basics firmly in place.

It is clear that Apple released iOS Maps very prematurely, mid-development. This was stupid, and there's really no use splitting hairs about it. Some bugs or problems are to be expected, but a critical function that most of your users depend upon, that has real, serious consequences if failure occurs (lost in a dark alley, etc. etc.) ought to be better than what Apple has put out. Add to this Apple's draconian policies that forbid Google from offering "Google Maps" as a third-party app on the App Store (though I guess today it looks like they may be waffling a bit on this), and you have a very serious problem.

I guess we can say that it may be good for Android adoption. :)

From the article

1. Completeness – Features are absent and some features that are included seem to have erroneous attributes and relationships.

2. Logical Consistency - Expecting the data across different sources to be completely consistent ie 'An example of this could be having a store’s name, street number and street name correct, but mapping it in the wrong place'

3. Positional Accuracy – is considered the closeness of a coordinate value to values accepted as being true

4. Temporal Accuracy – particularly in respect to temporal validity – are the features that they map still in existence today?

5. Thematic Accuracy – particularly in respect to non-quantitative attribute correctness and classification correctness.

The first iPod didn't do a lot, but what it did, it did correctly. If the first iPod had had the scale of quality issues we're seeing out of Apple Maps, there wouldn't have been a second.

All things considered, given how complex maps must be to implement, it seems like Apple did a pretty good job for day one.

Furthermore, this furore reminds me of the storm over Siri. Tech pundits work themselves into a frenzy proclaiming that Apple is losing its edge. Average consumers however pay no heed and the company rolls on to the next product launch largely unharmed. The critics miss the bigger picture: a company with so much momentum that it can easily afford to crowdsource the refinement of challenging big-data projects such as Siri and Maps. While the critics stand around prophesising doom, Apple iterates, improves, and by the time the next big hit comes out the last "disaster" is ancient history. Ignore their strength at your peril I say.

And why exactly should Apple care about this? Isn't the vast majority of their profit coming from higher then normal margins on hardware? Their bread and butter is suppose to be making compelling tech products that have high build quality and are easy to use for any type of user. Why would Apple care where ad dollars are going as long as they kept doing what they are suppose to be doing to justify their high margin devices? Apple isn't in the ad business.

Of course, with their map move they've just compromised the quality of one of the core services of a mobile device, that should be a bigger deal then any of this nonsense about location based recommendations being the future of ad revenue.

Anyway I think there are plenty of good reasons for Apple to pursue its own maps solution; eventually all technology will be fully location-gnostic, so it seems like a pretty critical technology to me. If I were Apple I wouldn't be happy leaving that in the hands of a malicious competitor either. And when you have one of the world's largest customer bases to beta-test with... why not? Seems like pulling a tooth to me - best just to get it over with quickly, and you're glad once you do.

You call it compromising the quality of a core service, Apple might argue that effectively unless that service was in-house it was not adequate, so effectively never existed, and what they've done is get started on building the service they should have had a while ago.

I don't know the particulars of the agreement they had with Google and other providers, but couldn't Apple have just rolled out their own driving directions application with restaurant and location search services but kept Google on for the time being? This would have allowed them to launch, then roll out a fully fledged Maps application of their own.

Nobody, including Google, consistently delivers functional map data to mobile devices every time, at least not in the Atlanta metro area. I travel there with some frequency for various soccer tournaments as a referee or parent. It is not uncommon for one or another device to provide faulty routing to the people involved in a match. Sometimes it's Garmin. Sometimes it's Android. sometimes Bing.

People don't primarily buy smartphones for the maps any more than people primarily buy smartphones based on call quality. Apple knows this. Apple sells phones because of iTunes and brand positioning.

The issues with maps didn't even garner a comment among the Apple fans in yesterday's Facebook feed. The edge cases among iPhone users that will be lost over poor quality maps is more than offset by the Genius's sales pitch about how easy Apple's map application is to use.

What has changed as Apple has scaled is that they are willing to weather a PR storm.

Imagine if it couldn't make calls, or got the wrong person when you did call? That would incite a similar emotional response.

Sure, nobody's perfect. There's so much data that problems inevitably slip through.

But Google's maps are at least "good enough" in most important markets. Apple's are not. That is a problem.

"All new edits need to be reviewed by another mapper. The more you successfully contribute to Map Maker, the more trust you will gain in the system and the easier it will be for you to make and review other changes to the map. If your edit is still pending and not on Google Maps, it has likely not been reviewed yet. You can post the links to the edits you’ve made on the review edit requests forum, so that your fellow mappers can review your map changes, enabling your edit to go live once approved by enough people with enough trust in Map Maker. Learn more about how reviewing and the moderation process works here."

email #2 "Congratulations! We have made changes based on your suggestion on Map Maker. Soon you will be able to see your edit live on Google Maps. Thank you for your contribution to Google Map Maker, and happy mapping! "

it's got status 'published'. is it pending some another review? I've read the faq and learn more, but still have no clue, the terminology is vague and a post from 2010 on a discussion group says 4-6 months to live.

Reminds me of the days when Microsoft used to release software that customers would call "beta-quality" and wonder if they were being used as unpaid QA.

I wonder how long it will take for the outside world's expectations of Apple to drop to what they should be in the permanent absence of the gaze of Barad-dûr?

Only if they realized it. It took Google 3 years to realize that they couldn't solve it algorithmically.

This could be a very interesting case study on sales of an otherwise great product being hurt by a move that cripples a critically important element of said product.

I've already met several people with iPhone 4's (or older) who said they are not upgrading to either iOS6 or iPhone 5 because of the mapping issue. I am part of that group as well. I'll have to buy an iPhone 5 for development purposes but I don't think I'll have it as my primary phone until the whole maps business is sorted out.

Mapping seems to be one of those things that you can't design your way around. In other words, nobody cares about beautiful inaccurate maps. This could be one of the first challenges on Apples's desk that can't be solved with cute commercials and pretty design. It has to be good and at least equal to, if not better than, Google's offering.

Regrettably sometimes the only way to get good at something is to start doing it. At first you'll probably suck at it but, with time and effort, you'll get better and better. This is Apple getting on that path to excelling at mapping. It'll take time. There's no doubt that they have the financial resources to make it happen. Now it is about execution.

Apple lacks the ability to mine vast amounts of local search data, as Google was able to do when it started its mapping project.

Does anybody know what this means? Don't the queries I make go through their servers, and isn't my location a relevant parameter for those queries? I don't see why this can't be mined.

Having a lot of relevant questions (and hard data on which answers to those questions were used) is an invaluable resource in bootstrapping mapping and local search together.

EDIT: Here is an archived version found elsewhere: http://www.allhatter.com/showthread.php/13017-Google-Maps-am...

This is ridiculous in the year 2012. Nothing about this should be a headache (apart from people keeping their data secret for various reasons) yet it's still bafflingly hard.

Suppose that data source A has a point for “Logan Airport” and data source B has a point (a few hundred feet away) for “General Edward Lawrence Logan International Airport” and data source C has a separate data point for every terminal.... on a case-by-case basis it’s not hard to resolve questions like this, but when you try to come up with an algorithm that scales to tens of millions of map features, you spend a lot of time scratching your head saying “why did it do THAT?!?!”

And they can improve, as Google did. But customers are very forgiving of mistakes and iteration when your product is the first and best (so far); plus, Apple's key branding is quality.

> Apple lacks the ability to mine vast amounts of local search data, as Google was able to do when it started its mapping project

This is an interesting competitive advantage Google has that is really really hard to beat. Similar to what Facebook has. But... perhaps Apple also has data? e.g. from Siri queries? They can certainly gather it now, now that their mapping app is being used.

I just wonder why fix something that ain't broke and make it worse!

Google must be pissing themselves laughing. Just as Apple get their act together google will probably wade in with a killer app? OR will they? Perhaps they'll not bother and use it as a marketing ploy to push Android?

Fascinating article.

Wasn't iOS6 in public beta for a few months?

Surely it's going to end up pushing improvements in maps technology.