[app_Developer]

It's interesting that in the article you quote it was suggested that the cost of inventory caused the company's financial difficulties that ended it in bankruptcy. Not, according to the company's executives, investment in software R&D.

Of course, because those guys weren’t foolish enough to go build an entire OS themselves! So their big problem was hardware. BB, on the other hand, tried to do both hardware AND their own OS, including their own kernel and driver set, which is why their break even point was much higher.

So they got out of the BB10 business, which lowered the break even point. However, they still failed to meet even that level of sales, and so they got out of the hardware side of it as well.

How software compatible are devices made by different manufacturers that use the same chipmaker's SoC and how much backward software compatibility is designed into each subsequent generation of a chipmaker's SoC?

I suspect much more than people here suggest.

What’s the reason you suspect this? GPU’s for example are hardly ever backward compatible with old drivers, except in some lowest common denominator bootstrap mode. Where have you ever seen this backward compatibility? What would be the point of even doing that? When you introduce a new GPU you always have the opportunity to ship it with compatible drivers.

[Troy Tiscareno]

It's interesting that in the article you quote it was suggested that the cost of inventory caused the company's financial difficulties that ended it in bankruptcy. Not, according to the company's executives, investment in software R&D.

Much like what happened with the Z10. It is also what I had suspected all along.

That it was up front cost of hardware not software which was BB10's money pit.

Actually, BB10 and Silent Circle can't be directly compared in software R&D costs AT ALL - they were almost completely different. While both were selling enhanced security, they otherwise couldn't have been more different.

Silent Circle was selling an ANDROID phone, and even without a Play Store license, they really didn't need to be overly concerned about an ecosystem, nor did they need to be concerned about their hardware not having drivers.

BB, running a completely different and non-Linux-based platform, had to not only develop (or pay to license) absolutely everything about the OS, including the stock apps, but they also had to pay for all of the infrastructure for all BB10 services (BB World, Blend servers, mobile payments, etc.), design and manage those services (and all the relationships), and critically, deal with all of the DEVELOPER-related costs, such as creating APIs, releasing and updating the development language (Cascades, a custom fork of Qt), the developer environment (to simulate the devices for development purposes), and all of the costs of trying to win and maintain relationships with developers. Oh, and let's not forget the one significant hardware cost: paying QualComm for optimized QNX drivers for any SoCs they were going to use.

As I said, the costs were orders of magnitude apart.

Certainly, the cost of mass-producing the phones themselves was also a massive cost - everyone was always pointing to BB's "money in the bank" without really understanding that almost all of that money was collateral for the phone hardware, so, it was effectively "spent" until they could actually sell the hardware and recover the money, only to "spend" it again for the next batch of phones.

So, with almost all Android devices being rootable the question ibecomes:

How software compatible are devices made by different manufacturers that use the same chipmaker's SoC and how much backward software compatibility is designed into each subsequent generation of a chipmaker's SoC?

Not as much as you'd imagine. SoCs change quite a bit each year, with newer CPUs, newer GPUs, newer memory controllers, newer peripheral controllers, obviously the big change from 32-bit to 64-bit a couple of years ago, and so on. That's why, even with the SoC manufacturers supplying Linux drivers, all of the manufacturers need to take a good amount of time to optimize everything for the individual hardware. Otherwise, you'd see Android phones supported with 4-5 years of OS upgrades like you see on iOS, despite having 200 different Android device manufacturers. Apple can do it because the number of devices they support is so limited and so tightly controlled, down to Apple designing their own custom SoCs, and also selling tens or hundreds of millions of each device model, which makes the cost of custom BSD-based drivers relatively trivial.

[i_plod_an_dr_void]

And how was Google/Android able to foot the bill for all of their software development costs (selling not enough Google branded phones in the beginning?), which is claimed to be the downfall of Blackberry?
Because they lined up an aresenal of consumer competitors, and the internals manufacturers and formed the open handset alliance.....the cartel to which the point of BlackBerry's troubles were caused. That has been my point...thanks for detailing all that development difficulty if you weren't going to sell a gazillion devices....yes the recoup for Google wasn't directly the number of sets it sold, but the mass behaviour of all the owners of the sets to which Google was able to monetize through its control of Android, any way you slice it, to me, it amounts to the same thing - cartel/ Open Handset Alliance.

On a lighter or creepy note: Perhaps Google was accidentally tipping its hat to its newly christened unofficial motto - One ring(smartphone) to rule them All - Lord of the Rings....
as just today it set up a donut shop to hand out those iced sugared rings to the masses (lol)

Google just opened a free donut shop in Toronto

[i_plod_an_dr_void]

No doubt Google experimenting with the taking out Tim Horton's in that(donut) market....after all, who can compete with free for very long, right BlackBerry?

[BBTIN]

In response to some posts stating no one would ever throw in good money after billions of losses: Look at smart vehicles (car brand of Daimler) first launched in 1998. After huge R&D invests, dwindling sales and heavy financial losses (billions), smart was liquidated! in 2016 and its operations were absorbed by Daimler directly. Although sales numbers have gone up again, it is still far away from a very profitable business but it has some strategic value to Daimler because of the industry change to car sharing and EVs. Of course, Smart is not at all comparable to BB10 OS but this example shows that in business you can never say never. However, will BlackBerry change its mind? Probably not just because of this Smart example.

[app_Developer]

In response to some posts stating no one would ever throw in good money after billions of losses: Look at smart vehicles (car brand of Daimler) first launched in 1998. After huge R&D invests, dwindling sales and heavy financial losses (billions), smart was liquidated! in 2016 and its operations were absorbed by Daimler directly. Although sales numbers have gone up again, it is still far away from a very profitable business but it has some strategic value to Daimler because of the industry change to car sharing and EVs. Of course, Smart is not at all comparable to BB10 OS but this example shows that in business you can never say never. However, will BlackBerry change its mind? Probably not just because of this Smart example.

Daimler has revenues of $200B, BB is now below $1B annualized. Different circumstances. Even $100M is big money for a small company like BB.

But the bigger difference is, as you said, companies will make strategic investments if they are in fact strategic. Daimler is still in the car business and plans to stay in the car business. BB, on the other hand, is out of the phone business and moving to, among other things, the car business.

So no investment in phone anything makes sense for BB unless it can produce immediate returns which they could then invest in their new car business or IoT business.

[kvndoom]

Call Daimler and see if they'll buy BB10. Probably be cheaper than the Smart acquisition.

Regardless of whether I come off as a smart***, the fact that all these internet posters have all these "amazing" ideas but not one, ONE, single corporate entity has put forth a single dollar towards making a new BB10 phone speaks volumes. There's pipe dreams and there's reality.

[Dunt Dunt Dunt]

Daimler has revenues of $200B, BB is now below $1B annualized. Different circumstances. Even $100M is big money for a small company like BB.

But the bigger difference is, as you said, companies will make strategic investments if they are in fact strategic. Daimler is still in the car business and plans to stay in the car business. BB, on the other hand, is out of the phone business and moving to, among other things, the car business.

So no investment in phone anything makes sense for BB unless it can produce immediate returns which they could then invest in their new car business or IoT business.

I do sometimes wonder if in the end BlackBerry was too small to really compete with their own ecosystem.

Even if Mike and Jim had been actual forward thinking people and released an new modern platform in 2007... would they have been able to beat Apple and Google? Android is free, Android OEM's work off pretty small margins compared to Apple and BlackBerry (back then). Maybe if they had been able to establish then new platform before Activesync made it too the market and they lost their real edge.... and been able to keep BIS in some form factor to allow them to have the position to force carriers to support them.

But BB10 as a OSAAS.... no, no and no.

[Emaderton3]

I do sometimes wonder if in the end BlackBerry was too small to really compete with their own ecosystem.

Even if Mike and Jim had been actual forward thinking people and released an new modern platform in 2007... would they have been able to beat Apple and Google? Android is free, Android OEM's work off pretty small margins compared to Apple and BlackBerry (back then). Maybe if they had been able to establish then new platform before Activesync made it too the market and they lost their real edge.... and been able to keep BIS in some form factor to allow them to have the position to force carriers to support them.

But BB10 as a OSAAS.... no, no and no.

I've wondered about this too especially since BIS was such a big part of their success.

Posted via CB10

[DonHB]

I do sometimes wonder if in the end BlackBerry was too small to really compete with their own ecosystem.

Even if Mike and Jim had been actual forward thinking people and released an new modern platform in 2007... would they have been able to beat Apple and Google? Android is free, Android OEM's work off pretty small margins compared to Apple and BlackBerry (back then). Maybe if they had been able to establish then new platform before Activesync made it too the market and they lost their real edge.... and been able to keep BIS in some form factor to allow them to have the position to force carriers to support them.

But BB10 as a OSAAS.... no, no and no.

People here that say BlackBerry had to write their own OS seem disingenuous to me because underneath the UI and apps much of the work was done with the exception of optimization for battery life in Neutrino. Writing apps is not the same thing as building an OS from scratch.

Also, in terms of cost of driver creation overhead, QNX has had do to that for each new car entertainment system they have developed (assuming minimal backward compatibility as has been said) and as we know Qualcomm open sources most of it so the source is there. Also, there is code commonality in drivers irrespective of OS.

Meaning, QNX should have great expertise with drivers. Qualcomms SoC designs targeting cars and handsets have been converging for years so more redundant work can occur between the two platforms.

Ultimately the real mistake, I think, was that BlackBerry didn't adopt the Android SDK and left native code support in BB10 for later. Porting Android to the native SDK could have been the proof of concept. The desire to support native code in the future would have been THE reason to not license Google Play Services.

Those that have read Losing the Signal know that this was considered early on in the development stage, but there was a push to leave Java behind. Not having to create an SDK, but to get Android working with Flow would likely have reduced time to market. But more importantly it would have improved developer ROI for BB10. For those thinking of OS/2, Android was an Open Source OS and BlackBerry could have been a counterbalance to Google's eventual privatizing of it. It is hard to believe that Android would come to this could not have been foreseen.

Its a shame that when John Chen became CEO he didn't chose to take the advice of Alan Brenner before essentially EOLing BB10 by licensing GPS. Of course we don't know for sure what BlackBerry's licensing terms are and now that they are not a device manufacturer they may be free to use Android as they like.

[conite]

Its a shame that when John Chen became CEO he didn't chose to take the advice of Alan Brenner before essentially EOLing BB10 by licensing GPS. Of course we don't know for sure what BlackBerry's licensing terms are and now that they are not a device manufacturer they may be free to use Android as they like.

He didn't EOL BB10 by licencing GPS. He was hired specifically by Prem and the BlackBerry BoD to end BB10 after burning through component commitments and sales contracts.

The game was up for a new platform by 2008/9, BlackBerry didn't find that out until 2013, and others still stubbornly wander the land of unicorns.

[Troy Tiscareno]

People here that say BlackBerry had to write their own OS seem disingenuous to me because underneath the UI and apps much of the work was done with the exception of optimization for battery life in Neutrino. Writing apps is not the same thing as building an OS from scratch.

When Google bought Android from Andy Rubin in 2005, there were 26 people working on it. Google immediately increased that number to 600, and continued to ramp that number up to over 1100 by the time Android was initially released in 2008. Today, there are over 2000 people working on the OS alone. Yet, Android was based on Linux - an OS that lots of people had already done lots of work on, and already worked on PCs. So, why in the world did/does Google have all those people working on it? Could it be that taking one type of OS and using it in a completely different way requires a whole lot of research and development work?

iOS was based on OSX, which itself was based on BSD, and Apple still worked on iOS for at least 3 years before releasing the first iPhone, and has continued to have big teams work on its development for the last 10 years. Why all of that work for an OS that already existed?

Why did MS have 2000 people working on WinPhone? Why did BB have 1800 people or so working on BB10 (just the OS)?

Simple: what you describe is simply not - even a little bit - how things actually are. Smartphone hardware is radically different from desktop-class hardware, and the OS that runs on it has to be a whole lot different too. Building a mobile OS (much less an entire platform, of which the OS is only a part) is a massive project, that requires a ton of manpower and a ton of money.

Look at SailFish - another Linux-based OS that is essentially just a continuation of WebOS, and yet after literally years of independent development, they managed to release a version (a year late) that an run on a single sub-model of a single phone, and STILL has problems with major subsystems such as BlueTooth. Yet, BT and everything else worked on Palm devices 10 years ago, so why does it take so much developing just to make it work on a single sub-model of one model of phone, and why do things that used to work no longer work?

BECAUSE IT'S FAR, FAR MORE DIFFICULT THAN YOU IMAGINE.

[markmall]

He was hired specifically by Prem and the BlackBerry BoD to end BB10 after burning through component commitments and sales contracts.

With due respect, I don't think you really know this as fact. I happen to disagree based on my close following of the company during this period.



Posted via CB10

[Invictus0]

Look at SailFish - another Linux-based OS that is essentially just a continuation of WebOS, and yet after literally years of independent development, they managed to release a version (a year late) that an run on a single sub-model of a single phone, and STILL has problems with major subsystems such as BlueTooth. Yet, BT and everything else worked on Palm devices 10 years ago, so why does it take so much developing just to make it work on a single sub-model of one model of phone, and why do things that used to work no longer work?

Sailfish is actually a continuation of MeeGo and it released a few months after BB10 in 2013, the Xperia X port was a recent development that probably took months instead of years (considering the base device only released last year). LuneOS is a continuation of WebOS (for phones anyway, WebOS seems to have found success in other products).

With due respect, I don't think you really know this as fact. I happen to disagree based on my close following of the company during this period.



Posted via CB10

Yeah I'll need to see sources for this as well. I think at minimum it's safe to say devices were de-emphasized after BlackBerry wanted to increase software sales but they still kept devices going hoping they could find profitability.

[DonHB]

BECAUSE IT'S FAR, FAR MORE DIFFICULT THAN YOU IMAGINE.

They had a head start because Neutrino was designed for embedded systems. That they didn't take as much advantage of the real time functionality of Neutrino is that SoCs for phones were not optimized for a real time OS. The architecture could be very different. Think what Apple's SoC would be like if it was using a real time OS instead of BSD. Perhaps, that is why top people from QNX have been hired by Apple.

Lots of conjecture here. Would like some sources too. Like evidence regarding the licensing terms of Android.

[Invictus0]

Lots of conjecture here. Would like some sources too. Like evidence regarding the licensing terms of Android.

This is probably as close as we'll get to be honest,

https://www.digitaltrends.com/mobile...samsung-tizen/

[conite]

With due respect, I don't think you really know this as fact. I happen to disagree based on my close following of the company during this period.



Posted via CB10

Big hint: Chen is a 100% software guy and a corporate turnaround expert.

[DonHB]

This is probably as close as we'll get to be honest...

The numbers of increasing personnel being talked about doesn't breakdown into what type of personnel and how many of them are customer facing. Developing a product and making it market ready is hugely different in terms of personnel numbers required. And we aren't even considering that many of these are marketing besides customer relations people, because this discussion has been about maintaining existing customers (and developers by extension).

[app_Developer]

They had a head start because Neutrino was designed for embedded systems. That they didn't take as much advantage of the real time functionality of Neutrino is that SoCs for phones were not optimized for a real time OS. The architecture could be very different. Think what Apple's SoC would be like if it was using a real time OS instead of BSD. Perhaps, that is why top people from QNX have been hired by Apple.

Lots of conjecture here. Would like some sources too. Like evidence regarding the licensing terms of Android.

Those people were hired to work on the car project. RT makes a ton of sense in a car.

What possible advantage does RT scheduling have in a modern smartphone? And pls don’t say preemptive multitasking. The only place that RT equals preemptive multitasking is on Crackberry.

[app_Developer]

They had a head start because Neutrino was designed for embedded systems. That they didn't take as much advantage of the real time functionality of Neutrino is that SoCs for phones were not optimized for a real time OS. The architecture could be very different. Think what Apple's SoC would be like if it was using a real time OS instead of BSD. Perhaps, that is why top people from QNX have been hired by Apple.

Lots of conjecture here. Would like some sources too. Like evidence regarding the licensing terms of Android.

So again here is the situation with every new smartphone SoC: the day they come out they are already running Linux, and they already have a full set of Linux drivers. Sailfish doesn’t need to worry about this. LG doesn’t need to worry about this.

Of the new smartphone SoCs that come out, zero of them run QNX because Qualcomm spends no time anymore doing QNX drivers. Chen explained this years ago. The cost of making QNX run on these new SoC’s is something BB had to pay for.

Why do you think they kept re-using the S4?

[Invictus0]

The numbers of increasing personnel being talked about doesn't breakdown into what type of personnel and how many of them are customer facing. Developing a product and making it market ready is hugely different in terms of personnel numbers required. And we aren't even considering that many of these are marketing besides customer relations people, because this discussion has been about maintaining existing customers (and developers by extension).

I was referring to your question on the license terms. You don't need a big team to work on an OS but it will impact the speed at which updates release and the type of support it receives. In a way, we can visualize this by looking at BB10 updates and seeing what size the team was at that time,

https://en.wikipedia.org/wiki/BlackB...ersion_history

BlackBerry Sweden was closed around 10.3.2.x and there were additional layoffs shortly after that IIRC.

Without any new devices a userbase will inevitably head towards 0, why would developers learn new tools at that prospect? And what will existing customers do when their devices break or need to be replaced?

[DonHB]

Those people were hired to work on the car project. RT makes a ton of sense in a car.

What possible advantage does RT scheduling have in a modern smartphone? And pls don’t say preemptive multitasking. The only place that RT equals preemptive multitasking is on Crackberry.

With most cell phone SoCs designed for Linux or similar not much. But with a simplified and cheaper architecture that does not compromise performance, a lot.

[DonHB]

]I was referring to increased staffing claimed at Google, Microsoft, etc and that the breakdown of staffing was not indicated. So, we don't know how many of the claimed 600 were doing OS development.

Without any new devices a userbase will inevitably head towards 0, why would developers learn new tools at that prospect? And what will existing customers do when their devices break or need to be replaced?

But that is why new devices based on BB10 with bug fixes are desired by many.

[app_Developer]

With most cell phone SoCs designed for Linux or similar not much. But with a simplified and cheaper architecture that does not compromise performance, a lot.

What do you mean? What simplifications would make RT more useful?

[DonHB]

What do you mean? What simplifications would make RT more useful?

Maybe a more powerful GPU could replace the DSPs in the SoC?