Ford to dump Microsoft's 'aggravating' in-car tech for ... BlackBerry? Ford Motor Company will part ways with Microsoft for the next generation of its Sync in-car infotainment system in favor of technology from Canadian rival BlackBerry, sources claim. According to a report by Bloomberg, switching from Windows Embedded Automotive to BlackBerry's QNX real-time OS will both save Ford money and …
The question is, How did this happen? I've worked as a vendor for Ford and they have an extensive six-month QA test program for all new technologies. What is described here would never had passed Ford's QA testing - unless there was a "deal" (wink, wink, nudge, nudge, Bob's your uncle) between MS and Ford's upper management.
"unless there was a "deal" (wink, wink, nudge, nudge, Bob's your uncle)"
It happens a lot, and I've been in the situation a couple of times where I was part of the process.
Everybody, and I mean *everybody* in the chain gets blamed for faults, whatever the reason, except the original vendor who's supplying and should be responisble for the crap in the first place.
Because MS always seem to get a free pass...
Any other vendor with such onerous licensing terms, poor security and dangerous level of lock-in would be excluded from any remotely sensible tendering process.
Various security standards have over the years been relaxed to accommodate MS, and in some cases actually require non-ms systems to comply with a much higher standard.
I had B&M in a 2006 Alfa and at the time it was miles ahead
of anything else from other manufacturers especially in term of
voice recognition.
Sadly Fiat never really upgraded it in the intervening years
and it has now been rightly canned.
Sadly the follow-up system is touch based and the voice
recognition is rubbish.
But the first generation Sync system in my 2010 Stang GT is not that much in my way. It worked pretty much with all smartphones I had since. It is not the most intuitive, but it's workable.
It's definitely not as good as Android in interpreting voice commands, but I wasn't expecting it to be. I am content with operating Sync by buttons only.
Based on how the 2015 Mustang looks, I would get a 5.0 even if it had MS-DOS on it. Actually, I would rather like it that way...
I used to drive a car with a very well-designed infotainment system - a 1978 Dodge Van. The system consisted of a radio with an AM/FM slider, volume and tuning knobs mounted on either side of the analog display, and a set of programmable preset buttons. (The buttons were programmed by tuning the desired station, then pulling the button out and pushing it back in.)
The entire thing could easily be operated by feel and had zero downtime for the lifetime of the vehicle.
Reliability and user experience have gone steadily downhill from there.
The trouble is that car manufacturers have spent most of the past ten years sticking technology and gadgetry where it isn't suitable, doesn't work and isn't wanted.
I can think of very few places where a touch screen interface is less suitable than in a car, particularly if it's supposed to be operated by the driver - and yet many manufacturers have used them for all kinds of operations which were previously trivially easy to manage purely by feel whilst driving.
I could not agree more. I was just talking to my friend about his new Lexus IS with haptic feedback controls and he goes like "you have to click like 20 times to do this and that but it is really really interesting..." WTF is it with this swipe and click for 5 minutes to get to a contact in a phone book madness? And now they put this into cars??? That is just simply dangerous, IMHO. On the old Nokias you make two keystrokes and there u dialing.. Touch is backwards and so are obviously 90% of the world, period! This is in no way an improvement.
I always thought that you were supposed to design a car interface as if the driver was blind, that way they will eventually not have to look at it to use it while driving. I absolutely hate touch screens as an interface in a car, I don't mind large displays for things like Navigation or media selection, but I don't ever want my grubby hands to touch it.
I can only assume that one big slab of virtual buttons is a lot cheaper than designing, sourcing, and assembling a dash full of real buttons. That unfortunately means that in the future there will be a huge screen and a steering wheel making up the majority of dashboards (cough, Tesla, cough).
I can think of very few places where a touch screen interface is less suitable than in a car, particularly if it's supposed to be operated by the driver
The only thing worse than a touchscreen is a turn and push interface, as on the BMW iDrive system. My son had one of these, and if you think using a touchscreen while trying to drive is tough, try twisting a knob left and right to move a cursor to the item you want to select, then pushing down to activate it...while trying to drive. It controlled *everything*...radio, heat/cool, nav system. Whoever designed it obviously drove only on unpopulated, glass-smooth highways, because it was impossible to use otherwise.
"Infotainment" systems...they'd be banned if they were on a smartphone, but somehow, they're just fine if they're integrated into the vehicle.
"QNX is Blackberry by name only"
Indeed but BlackBerry, despite whatever their faults might be, knew a good thing when they saw it and bought QNX for use by themselves. BB10 has certainly benefited from being based on QNX.
I don't really understand why MS's automotive offering has gone so badly wrong. Did they plug it as a finished solution for the car makers who had only to burn it to ROM, but then spoiled it by not polishing it, fixing the bugs, etc?
"I don't really understand why MS's automotive offering has gone so badly wrong."
Well I think I do know...
I spent a few years developing Windows CE systems from the ground up (writing the drivers etc). This was back in 2000 or so. At the time, WinCE was being pushed for multiple different roles including hand held, automotive, industrial etc.
The actual WinCE development and marketing team was tiny (it got that I knew a lot of them by email) and many of the offerings (eg. WinCE for automotive) were really little more than a two sided glossy pamphlet.
In the first 10 years or so, WinCE was never profitable for MS. Nor did they really invest enough in it to build it into anything useful. They just kept it alive.
On top of that they were not painting a clear road map for potential customers. They pushed PowerPC and SH3 support in their first release, but excluded ARM. x86 was for simulation only.
Then they dropped PPC and SH3 support, allowed x86 and ported to ARM. This road map confusion and erratic behaviour was very worrying for many potential customers.
Many of the products that they did get were highly subsidised "design wins" to try to promote WinCE. These tend to be driven by MS wanting to showcase themselves, rather than by the customers wanting to deploy their features. Net result is product that just does not really do what it should.
As a consultant, I still have some customers using WinCE but many of those are worried and are looking for exit strategies to Linux. They have little confidence that MS will support WinCE into the future and most of them want out.
This post has been deleted by its author
"As a consultant, I still have some customers using WinCE but many of those are worried and are looking for exit strategies to Linux. They have little confidence that MS will support WinCE into the future and most of them want out."
Sounds like the typical WinCE experience. MS really had zero imagination when it came to anything other than desktop and server software. That's why they missed out on the mobile revolution.
As for an exit strategy from WinCE, that sounds awkward. If they'd picked a POSIX-ish OS in the first place then their code base would be much more portable than it is. No doubt there was some engineer on the staff at the time WinCE was picked who said "This ain't a good idea" for that very reason!
Two reasons I don't design Microsoft into the user interfaces for our embedded devices:
1. Cost - licensing, limited hardware platform options (Intel embedded stuff is much more expensive than, say, i.MX or ARM. We're actually using RasPi with Linux on our latest UI)
2. Future - I can't be sure how long the software platform will be around
Our software guy loves the features of .NET, but there's so much "business baggage" that comes along with using MS in an embedded environment, I'm just as happy to do without the hassle.
QNX has also benefited a lot from being bought out by BlackBerry. Their new Car platform is essentially BB10 re-purposed. The delta between BB10 and pre-BB10 QNX was due to BlackBerry involvement. And some major components like the UI framework, the browser, and the Android runtime were BlackBerry-developed.
"QNX has also benefited a lot from being bought out by BlackBerry. Their new Car platform is essentially BB10 re-purposed."
Ah, I was wondering about that. I use a Z10, and a PlayBook for that matter, and whatever else one might think about it the BB10 UI framework certainly allows a dev to produce a well polished result.
I wonder how important the Android runtime is going to be in the automotive sector. There's currently no app-store type market place for in car infotainment systems, so I'm guessing that most of the applications are bespoke to or customised by each manufacturer. So if they're being written specifically for each manufacturer, why write it in Android when one might just as well write it for BB10? And, any sign of TomTom doing a BB10 version of their Satnavs?
"Ah, I was wondering about that. I use a Z10, and a PlayBook for that matter, and whatever else one might think about it the BB10 UI framework certainly allows a dev to produce a well polished result."
On top of QNX BlackBerry also bought an entire (Scandinavian, of course) UI/UX design team, famous for their beautiful, elegant, high-quality designs (for Samsung, Nokia, Sony Ericsson, Disney, Motorola, PayPal etc), to do a full reboot in BB10 - it was The Astonishing Tribe aka TAT: http://www.tat.se/
I, for one, love the UI design of BB10, it's far beyond anything I've seen on Android, no question about it (forget iOS and it's Jobsian cheesy-glitzy, low-brow older versions or its downright cheap-looking latest iteration.)
They were full of awesome mobile UI ideas even before BB acquired them in late 2010: http://www.youtube.com/watch?v=g7_mOdi3O5E
Although apparently seven of the TAT founders have left BB since and started their new firm called TOPP - see http://www.toppstudio.com/ - still, buying QNX and TAT were the two BEST-EVER acquisition by BlackBerry; these two practically saved them from completely going out of fashion, for sure.
Reliable fine, but Matag makes reliable appliances but that doesn't mean I want them designing a media and navigation interface.
I have had a QNX-based interface before and it sucked, was buggy, and once it was in my car, they clearly felt no desire to keep developing and issuing fixes. Who walks around with a 5 year old smart-phone, why do we allow car manufactures to basically lock us into the same thing. There are at least 5 distinct GUI and navigation bugs in my current car, and they will be in there until it gets crushed according to the manufacture.
That's why they're probably better off developing some sort of dock or holder for your already existing phone. Then connect with usb/lighting/whatever and just develop an app that interfaces nicely with the stereo/odometer/etc. That way, the app will get the updates and your car doesn't need them.
I'm not sure if MS involvement caused the whole mess with Sync. Unless MS - just like on the PC side - pushed touch interface just to prove this could be done. With only minimal exposure to sync (rental vehicle) all I can say is that car manufactures share majority of blame here (and just some recent review of Toyota showed their interface to be just as clunky and laggy). Automakers keep pushing touch interface where it does not belong. All this because some buyers are - initially - impressed by hi-tech looks and manufacturer can command/justify higher price. As much as I'd like to have nicely integrated mount for a smartphone (and some basic/functional interface) I'd rather keep all others controls as hardware buttons/knobs. It just works.
Every time I see another semi-embedded system running on a MS OS, I have to ask myself "what can these people have been thinking?"
Each time my car infotainment system crashes; each time I see yet another airport flight display with a DOS prompt on it; each time I see a cashier at a checkout rebooting their Windows system just so I can check out;
Hard to understand the rationale behind such thinking.
"Each time my car infotainment system crashes; each time I see yet another airport flight display with a DOS prompt on it; each time I see a cashier at a checkout rebooting their Windows system just so I can check out;"
Yet underneath all that crap there lies what is essentially the NT kernel and the bones of MS's desktop operating system (unless it really is DOS, but surely that's not been used on anything recent...). Windows in it's various (supported) guises doesn't have a huge reputation for conking out randomly, at least not these days.
So it's more likely that buggy bespoke device drivers and software written by hired then fired devs are to blame. A lot of manufacturers just don't get it; smarter systems absolutely require continual and substantial development, otherwise someone else will come along and steal your business.
It happened in mobile phones. Smart phones were great, then Apple showed everyone that actually they weren't and perhaps this iPhone thing is a better way, and everyone else has been playing catchup / going extinct ever since.
Cars are going to be increasingly sold because of their tech and the manufacturers have to get deadly serious about Doing Software Properly. Otherwise the likes of Google or Apple will come along and start controlling the market place. Hiring a few programmers for a few months to hack an infotainment system together before sacking them and shipping it isn't going to work against Apple and Google. They'll practically have to turn themselves into software companies with a sideline in car manufacturing.
[Having said that the prospect of a car running Apple Maps for a navigation system really won't sell at all well...]
"what can these people have been thinking?". Easy: free software & development services and subsidised hardware.
Most of them did not choose Microsoft as such. MS chose to "partner" with them. MS would have had a hand in designing the hardware, the software, the UI etc.
From MS's point of view this is a showcase for Microsoft. That it might be part of a Ford or whatever is secondary.
Result: you get touch screens and other gee-gaws where they are not appropriate.
even PCs have dedicated function keys. I have begun to hate tablets/phablets because of the touch interface. eg entering none trivial passwords is hard. In cars, dedicated knobs/buttons or switches have a lot to recommend them. As for entertainment systems, perhaps more people could learn to enjoy silence. Allows better concentration in high workload conditions. On long trips simple audio is sufficient in front, and multimedia players exist for back seats.
As for QNX, good move. It just works.
Strange that the car-manufacturers didn't go to air-plane manufacturers for some guidance in this cockpit-interface solutions (hybrid of buttons and touch ).
the big boss of Ford has to call his old buddy's at Boeing ... and the European manufactures can go to Airbus (wasn't Airbus who was the first to introduce this in a civil air-plane ? )
@AC good point. The hybrid control devices like Altair have both interfaces and seem easier to use. I know that the touch only interface of a WinCE device or Android running flight management software is irritating. Touch only interface is difficult and distracting. I wont go into UI design. The dedicated glass cockpit devices seem to have both type of interface, but I can't check that. This is an issue that is generating much discussion among instructors about the safety implications of additional pilot distractions, and rightly so. Additionally, touch can't work well in turbulence or the rough tracks that are becoming the norm for country roads. We glider pilots spend a lot of time in both circumstances, especially in Oz where we have had a fantastic season this year.
...is a bluetooth connection to the audio system in my car. I'd like to have a bluetooth connection to OBD (though I'm not really sure why, just 'cause). Unfortunately, my car is too old (1994) to have OBD2, so no luck on that one.
Other than that, I have a dash mount for my phone and access to everything I need from it without some auto manufacturer's re-imagining of how it should work.
Is a pocket for the phone and a bluetooth headset with voice activated dialling. Touch the headset, speak the name, it will confirm the party and dial the number. Look no car mounted software, sadly the mobile function is not supported in modern crud phones but hey, it works for me with my 8 year old phone so no need to down grade to a big flat look-at-me dumb phone.
Even the radio work by pushing a button or two, note button not slithery 'touch and miss' screen.
" What I need... ...is a bluetooth connection to the audio system in my car"
Your car from 1994 probably has a cassette player, if so you are in luck - There are several cassette bluetooth adapters available.
Here's one, no idea how much the sucker costs but it'll be less than a new player.
http://www.ionaudio.com/products/details/cassette-adapter-bluetooth
My car is a 2003 and has a cassette with a factory-fitted CD changer in the boot. The sound is fantastic, and I really don't want to change it. I have recently busted the lead on my cassette/mp3 player adapter after 2 years and was doing to look for a bluetooth one so thanks for that. My Z10 sounds great on it though.
However, going back to the original article, I'm not so sure a "fully immersive" ICE system is really what we need on our roads... is it?
"Your car from 1994 probably has a cassette player"
LOL, didn't think that one through. I actually do have the audio bluetooth connection because after market car stereos with new features can still be added to older vehicles.
But, thanks, DJO, maybe I could get one for my Boombox!
Never had the problems they've listed with it in my 2013 Kuga. Works with all phones I've tried it with. It has obviously been designed by 3 or 4 people working separately and with no UI design experience, but that's another issue. You'd think when you're doing sat nav/menus etc. that the rotary dial would move up and down, but no it does volume all the time. You can't scroll menus from the controls on the steering wheel either, they just do the menus on main instrument display. Stupid.
...is FAR more robust than anything MSFT ever produced - there's a reason why it's one of the most commonly deployed industrial-quality RT embedded OS, nothing comes close from MSFT (please, PLEASE don't even try to mention this abomination called Windows Embedded, that's an utter PoC and likely Ford just learned that in a hard way.)
And people often don't realize but it in-car systems DO make a very lasting impression. My single encounter of a rental but brand new Lincoln MKX couple years back during a NYC-DC-NYC round trip convinced me to never look at the brand, primarily due to its utter junk entertainment system (froze up around every ~15 mins, up until he point when it completely froze to death, around halfway back to NYC, leaving us with no music, no radio, nothing whatsoever - heck, even the driver's small console was frozen between the main dials! Yes, it was MSFT's In-Sync crap.
The Consumer Reports description: "poorly designed," "maddeningly fussy," "time consuming and cumbersome," and "distracting to use while you're driving." - that all smells like UI & apps, i.e. exactly where Ford is bringing their special expertise to bear. Rehousing the same atop a better OS core won't achieve much so have to hope they have the wit to do some real UI design too.
And for all that's holy, drop the animations and distracting eye-candy: buttons and touch areas need to be big, plain, and predictable, and stay unobtrusive unless they have something important to communicate (and that's "important for the driver", not "important for my team's bonus that we better highlight Twitter connectivity status"). Among the reasons VW gets repeat sales is that you know that the major controls in the next model (or even in different product lines) will look the same, work the same, and be in the same damn place.
"According to a report by Bloomberg, switching from Windows Embedded Automotive to BlackBerry's QNX real-time OS will both save Ford money and improve the flexibility and speed of its in-car systems."
I switched for MS Office to Libre Office, it saved me money and improved the flexibility of my office applications and increased system speed too.
"A year later, Ford's position in the JD Power & Associates Initial Quality Survey had plummeted from 5th place to 23rd"
Hang on - I get that it's an important bit of a car, but fifth to twenty third? How many marques are even extant? If the position was really for 'Ford', that would imply that it wasn't per-vehicle, and i'd be hard pressed to name 23 car manufacturers, even if you assume that sub brands like Lincoln/etc were separate... Who would even be left to be behind Ford? BYD? Geely? Something is missing here...
will have a working clock.
Since I got my Focus, with version 2.11 of MFT, Ford has released three updates without providing a working clock [or addressing other outstanding defects]. Ford has removed useful features, and added an unwanted animated compass. [And the BSoD claims to be 'performing scheduled maintenance']
I think that this is mostly a project management problem. To put it mildly, Ford is not seeking input from users/customers about defects and misfeatrues. [no issue tracking system, pathetic online forums and support]
It's been a long tradition here in the US of A, that vehicles should have a clock on the dashboard (an extra-cost option, of course), but that under no circumstances should that clock actually *work*.
Perhaps, somewhere, there is a US-made car with a working [analog mechanical] clock, but I've never seen one.
I have heard of coworkers "updating" their Ford media systems, and have never ever seen a QNX based upgrade. So either QNX never releases a system with any bugs or problems, or they simply don't care once it's in your car. I'm betting the latter, especially since the only way I can get my QNX based car to display the current song playing off the internal harddrive is to first go into the "Telephone" menu, then back to "Media" and magically it jumps to the currently playing song in the list.
Damnit, I shouldn't have to buy a new car to have the radio work correctly.
Yeah, right.
#1 The suit is in California. Any friends of Microsoft live in California? Any enemies of Microsoft live in California. Ah... So the suit is frivolous and there to generate adverse column inches
#2 Is Blackberry likely to be offering their technology at a discount at the moment? Mmm.
Seems like a post justification exercise. The story is that something from Microsoft is so defective that it alone has caused a rating slump. Now the clue is in the name MicroSOFT. That's 'soft' for 'software', you know, the thing you can change so the UI is what you want it to be.
Sounds to me like Ford have f**ked up royally and want to reduce costs because those nasty people in Redmond will not negotiate to a price Ford wants to pay. But rather than taking the blame for their own incompetence, Ford management has decided to shout loudly "It not me, its them, its them".
It's nothing to do with the os. It's the idiots who design the GUIs. They can bring any os to its needs with stupid design and illogical though processes.
I have a GUI visualization product that has been running wince 3.0 since 2002. Never upgraded it cause it was perfect. Never crashes never fails. It's been super reliable. Boots in 4 seconds out of the box no special changes. I have a few Linux projects too. Needless to say the grass is not greener on any other side. If anything I quite respect the leanness of wince. I see it as a positive.
Like in most cases it won't be a fair comparison with fords old system to the new. They will be going from some ancient 150mhz arm chop with 65mb ram to a dual core cortex with 512mb ram, OpenGL es etc. you could do the same with Linux, or wince and the new system
Would of course be superior.
There's been problems all the way through the program. Ford were desperate to lead the pack with providing advanced connectivity in the car and Sync was released too early and with a lock in to Microsoft for which I'm sure they paid a premium price. They compounded this by releasing the touch screen variant (Gen 2) only to have a slew of more issues with not just the touch screen but the update process as well. As a Sync Gen 1 (UK) user it's been sold as complete hands free connectivity but that is far from perfect with the manual fallback, for when voice activation doesn't work, relies on you taking a hand off the wheel and look at the centre console for a majority of actions.
The problem that Ford and Microsoft have is in not meeting expectations. Their customers have fruity and droid devices that in their eyes just work out of the box and neither company put sufficient effort in understanding the market, technology and expectations. Given that Sync was portrayed as the differentiator and then found wanting, I'm not surprised about the JD power results.
It remains to be seen how good QNX / Ford will become but the automotive maker had, IMHO, no option but to dump a failing product/vendor. For the Microsoft division this is a serious setback and something radical will need to occur to make their product competitive let alone leading edge.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
