[app_Developer]

Some details regarding high level elements of any operating system. Each of these is comprised of multiple components with many interdependencies, and a team of specialist developers with deep domain knowledge. Many of these, such as memory management require very specialized experts re-engineering components for each generation of hardware. Just the maintenance and ongoing tuning of these subsystems requires several hundred engineers.

Kernel
Process Execution
Interrupt
Memory Management
Multitasking
Networking
Security
User Interface

And of course, none of this works without device drivers, which requires technical, specialized teams.

One estimate I've heard for the number of iOS developers at Apple is 5,000 people.

Z10 = BB10 + VKB > iOS + Android

Networking alone is hundreds of people over there. Many thousands work on Android (not including GPS and apps). Tens of thousands work on Linux and all its various distros today.

But again, these are people doing real engineering, not people making fantasies out of stuff they read in press releases or marketing materials.

[conite]

Will you ever be specific? You make statements, but can't provide a detailed explanation (or examples) of how it is "heavy work"?

You have been presented with countless examples over the months, but you don't seem to take anything to heart.

The fact is, it's very difficult to find common ground with someone who feels that almost anything can be developed by three interns in a garage.

I again draw your attention to this post:

/general-blackberry-news-discussion-rumors-f2/clock-work-true-story-backdoor-bb10-clock-1140464-post13168089/

This photo represented only 75% of the (direct) team who worked on the following (not including support and management personel):

"We developed the Clock and Weather apps from scratch, and we took over maintenance of the Compass and Calculator apps that had been developed in other parts of BB. (No, there aren't any hidden features in those apps.) Plus, we worked on the Android version of Docs To Go (which never got released as standalone product), fixing bugs and adding features (like Good Dynamics support)."

[Dunt Dunt Dunt]

Will you ever be specific? You make statements, but can't provide a detailed explanation (or examples) of how it is "heavy work"?

How is it not?

At one point there were over 4,000 people working on BB10... even long after the lauch. Heck BB10.3.1 was mostly done before Chen started laying them all off in 2014. The thousands that remained, were mostly there for minor tweaks and maintenance for later versions of BB10.3.1.xxxx on into BB10.3.2. Which is when the second big round of layoff happened. But most everything points to their being over 500 developers involved in BB10.3.3. These ended up transfer to BB Android or Ford....

Even when they had 4,000 people.... how long did it take to get email on the PlayBook? If you look at the PlayBook OS with email in 2012... how much work was really needed to get out BB10.1? Over 4,000 people working a year, and they still didn't have it "right".

Look at Sailfish... At one time they had hundreds of developers, working on an existing OS (Linux based MeeGo).... now they have a handful and can't do much more than refresh the UI a little. There are 64Bit Linux Kernels out there, lots of open sourced projects that include 64Bit processes... why can't a handful of developers move Sailfish to 64Bit?

[Troy Tiscareno]

How is it not?

At one point there were over 4,000 people working on BB10... even long after the lauch.

Prior to the launch, it was well over 6,500 - there was a round of layoffs at the end of 2012 when initial development (finally) ended, with the "core team" being about 4,800 developers. That's not counting the the business, sales, and support staff, which was another 3000+ people, NOT counting the 1800 outsourced field reps.

[DonHB]

There’s no remaining in the business as they exited over three years ago. BlackBerry Limited went in a different focused business strategy. No resources anymore either. You’re a minimum 3-5 years too late.

Technically, BlackBerry remained in the business with BlackBerry Mobile and may still be in the business with BlackBerry Secure.

They did stop making handsets, if that is what you intended from your statement.

[DonHB]

Some details regarding high level elements of any operating system. Each of these is comprised of multiple components with many interdependencies, and a team of specialist developers with deep domain knowledge. Many of these, such as memory management require very specialized experts re-engineering components for each generation of hardware. Just the maintenance and ongoing tuning of these subsystems requires several hundred engineers.

And of course, none of this works without device drivers, which requires technical, specialized teams.

One estimate I've heard for the number of iOS developers at Apple is 5,000 people.

Z10 = BB10 + VKB > iOS + Android

I refer to my earlier comment regarding good enough. If it performs better than the Passport and has better Android and browser compatibility I would be satisfied. Because this is neither competing against iOS nor Google's Android, the development investment can be focused. Also, no one has addressed the question if choosing a handset with a SoC and a modem that would be the basis of a future car chipset could improve the development bottom line. It could be Exynos based using Samsung's 5G modem, if it would make that company more interested. I also suggested that customer facing support would not be that different than now. So, investment could be dedicated to development and QA. With sufficient buy-in customer facing support could be improved. Everyone should wrap their heads around this being a LOW priority endeavor which builds on assets from other projects and product lines. It is not intended to compete with iOS (despite Apple's presumed focus on privacy) or non-AOSP Android.

[DonHB]

So here’s an idea:

Someone could build a curated App Store for Android that has higher standards for testing and certification.

You could also just make a rating service for any Play store app. But with a store, and the developers permission and cooperation, you could add a sidecar to the app that monitors its activity, ensures that it remains in compliance and could detect any alteration or compromise after install.

If people are willing to pay $ for additional privacy, or companies are willing to pay for additional privacy and security, then they could buy certified versions of apps from this store at a premium (which you can share with devs)

Just a thought, and would test whether people are willing to pay real money for additional protection. And you can start for a lot less investment than a whole new OS. And you don’t need BB to do this. Maybe you can license Cylance for some part of it.

This could suggest building the platform is putting the cart before the horse. Android as the SDK should make changing the platform easier down the road due to the VM and what you said about gauging interest.

Maybe Jarvis and the various Cylance products being offered to their auto manufactures could be applied into making BlackBerry World into the curated Android application store you mentioned. Problem is, what will replace Google Services Framework to make developers' work easier and the apps more private? I would guess Amazon's SDK is a little better in this regard, but not good enough.

[app_Developer]

This could suggest building the platform is putting the cart before the horse. Android as the SDK should make changing the platform easier down the road due to the VM and what you said about gauging interest.

Maybe Jarvis and the various Cylance products being offered to their auto manufactures could be applied into making BlackBerry World into the curated Android application store you mentioned. Problem is, what will replace Google Services Framework to make developers' work easier and the apps more private? I would guess Amazon's SDK is a little better, but not good enough.

Not even Amazon had the money or dev support. There is no support for replacing whole parts of Android.

But you don’t have to do any of that if you just build a curated store. It’s straightforward and simple and could actually do a lot of good.

[Chuck Finley69]

Technically, BlackBerry remained in the business with BlackBerry Mobile and may still be in the business with BlackBerry Secure.

They did stop making handsets, if that is what you intended from your statement.

No, they didn’t remain in the business with BlackBerry Mobile as they don’t share in the risk or consumer facing anymore. Literally, they’re just an Enterprise vendor to BBMo like they are to auto companies with QNX products.

[joeldf]

Technically, BlackBerry remained in the business with BlackBerry Mobile and may still be in the business with BlackBerry Secure.

They did stop making handsets, if that is what you intended from your statement.

Technically, in 2016, BlackBerry Mobile became a direct subsidiary of (and owned by) TCL to sell the BlackBerry branded phones made by TCL. Taking BlackBerry Ltd out of the equation. And since last year, BBMo has pretty much been gutted.

[Chuck Finley69]

At this point, there’s nothing that Android/iOS can’t already bring to conversation more cost effectively and that satisfies majority of consumers. QNX developed into an updated BB10 or BB20 isn’t something demanded enough to be even closely viable....

[DonHB]

At this point, there’s nothing that Android/iOS can’t already bring to conversation more cost effectively and that satisfies majority of consumers. QNX developed into an updated BB10 or BB20 isn’t something demanded enough to be even closely viable....

You don't find it interesting that Google is working on a microkernel based OS and Huawei has introduced one to compete with Android? Interesting because most of the people negative on BB10 think it was a mistake to use Neutrino. Consider what could have been had BlackBerry headed Alan Brenner's advice to implement Flow using Android. BlackBerry would have been years ahead of both those companies. Cascades could have been used, back in 2013, when John Chen became CEO to remake the Android Player in Flow's image. Of course a Huawei executive claims that HarmonyOS' IPC is up to 3 times faster than QNX and 5 times faster than Google's Fuchsia, but this remains to be seen.

[DonHB]

Technically, in 2016, BlackBerry Mobile became a direct subsidiary of (and owned by) TCL to sell the BlackBerry branded phones made by TCL. Taking BlackBerry Ltd out of the equation. And since last year, BBMo has pretty much been gutted.

As far as I know BlackBerry Mobile never existed before TCL created it. Unless you consider producing software for handsets doesn't count, they are still in handset market. As long as BlackBerry is producing updates for BlackBerry Secure on Android they are also in the OS space in handsets.

[Chuck Finley69]

You don't find it interesting that Google is working on a microkernel based OS and Huawei has introduced one to compete with Android? Interesting because most of the people negative on BB10 think it was a mistake to use Neutrino. Consider what could have been had BlackBerry headed Alan Brenner's advice to implement Flow using Android. BlackBerry would have been years ahead of both those companies. Cascades could have been used, back in 2013, when John Chen became CEO to remake the Android Player in Flow's image. Of course a Huawei executive claims that HarmonyOS' IPC is up to 3 times faster than QNX and 5 times faster than Google's Fuchsia, but this remains to be seen.

But you’re ignoring the obvious. Either company has 1000x the resources now and 100x the resources then of BlackBerry at it’s asset peak.

If either company is mistaken, it’s a simple rounding error banished to “never talked about ever again land”

[app_Developer]

You don't find it interesting that Google is working on a microkernel based OS and Huawei has introduced one to compete with Android? Interesting because most of the people negative on BB10 think it was a mistake to use Neutrino. Consider what could have been had BlackBerry headed Alan Brenner's advice to implement Flow using Android. BlackBerry would have been years ahead of both those companies. Cascades could have been used, back in 2013, when John Chen became CEO to remake the Android Player in Flow's image. Of course a Huawei executive claims that HarmonyOS' IPC is up to 3 times faster than QNX and 5 times faster than Google's Fuchsia, but this remains to be seen.

It was a mistake to use neutrino because BB was too small to make that work. It added too much time, it added too much cost, and it limited the hardware it could run on. BB didn’t have time to waste. Or money. And BB had very little influence over the hardware manufacturers.

Huawei has more than 2,000 people working just on the kernel and low level Libs. Because they know it will take a lot of effort to make it work, but they have the incentive of having a complete Chinese owned OS. That is a strategic asset for them and for China and there is more than enough money there to make that investment.

BB is not Google and it is not Huawei. It is a tiny company compared to them. QNX offered no actual advantage in phones. None.

If Google decides to really push their new OS (which is not a certainty at all. Right now it’s just a project) then Qualcomm and others will get on board because it’s Google. Google wields a tremendous amount of power in this industry.

[DonHB]

I have no reason to trust Blackberry with my privacy any more than I do with Apple or Google. A new platform based on privacy should be opensource, so its privacy is verifiable.

It seems you also think there would be consumers who want a third platform based on BB10 to exist to combat the "Duopoly".
BB10 failed miserably in the market place already? Windows Phone failed, The Amazon fork failed, Palm failed. The Market seems fine with the Duopoly.

So IMO your sales pitch seems rather weak.

I thought to add that as far as I know licensees of SDP 7.0 can get OS source code. QNX has licensing specifically targeting Academic and similar institutions. So, having an independent auditor(s) of the software shouldn't cost more than doing the same for Linux and other OSS.

[DonHB]

It was a mistake to use neutrino because BB was too small to make that work. It added too much time, it added too much cost, and it limited the hardware it could run on. BB didn’t have time to waste. Or money. And BB had very little influence over the hardware manufacturers.

QNX offered no actual advantage in phones. None.

That is due to the very time constraints you mentioned and it being EoL'd. Why would Google invest in building a microkernel if it offers no advantage in mobile? What is its expected return on investment that BlackBerry can not realize? How incompatible are drivers for previous generation SoCes? For example, if the code that identifies the GPU is replaced would the driver work? If so, how compromised would the performance be? On the desktop the GPU vendor wouldn't want a (previous generation) driver to compromise the customers experience and the associated bad PR, but here we are talking about good enough and being particular as to where to invest development dollars.

[joeldf]

That is due to the very time constraints you mentioned and it being EoL'd. Why would Google invest in building a microkernel if it offers no advantage in mobile? What is its expected return on investment that BlackBerry can not realize?

Like... oh, I don't know... about a few hundred million more possible devices it could be used on than anything BlackBerry could hope for?

[app_Developer]

That is due to the very time constraints you mentioned and it being EoL'd. Why would Google invest in building a microkernel if it offers no advantage in mobile? What is its expected return on investment that BlackBerry can not realize?

QNX offers no advantage to users of phones. Google may be able to get some development/testing efficiencies out of their design at scale. We’ll see. I don’t work there. They may even be going after the IoT device or car business.

They may not even do anything at all with this new OS. They may even just take the new runtime and pop it on top of Linux. In fact I think that’s pretty likely.

You’re focused on the micro kernel part of it because QNX, but the more interesting parts of the Google project are everything else about it, like Dart. The improvements in runtimes etc may be the next Android and may well just end up running on Linux. Or they may take some ideas from it and bring them back to Android.

It’s an R&D project. Big companies do that as a way of experimenting and pushing innovation. They can afford that. BB cannot. Google isn’t betting their company on fuschia.

[Dunt Dunt Dunt]

I thought to add that as far as I know licensees of SDP 7.0 can get OS source code. QNX has licensing specifically targeting Academic and similar institutions. So, having an independent auditor(s) of the software shouldn't cost more than doing the same for Linux and other OSS.

Most Academic uses have ended..... QNX is no longer "open". While BlackBerry does offer special "free" licencing for Academic usages, no one wants to developed something that might become a commercial product one day, that would then have to be licensed. And no one wants to learn a system that doesn't have a commercial foot print. Outside of the BlackBerry zone in Canada, I doubt you'll find much QNX usage in collages today.

Even QNX used in IoT products like traffic lights, Industrial manufacturing equipment, control panels..... and other areas has fallen under the BlackBerry yoke. QNX is 95% an automotive product at this point... which scares many investors.

Look at Cylance.... there needs to be some level of corporation between companies and auditors, but BlackBerry no longer allows corporation by Cylance developers. BlackBerry as a company doesn't have a reputation of welcoming audits of their software.

[DonHB]

What discourages uptake QNX? Is it the licensing or that there are not enough developers familiar with it?

[DonHB]

You’re focused on the micro kernel part of it because QNX, but the more interesting parts of the Google project are everything else about it, like Dart. The improvements in runtimes etc may be the next Android and may well just end up running on Linux. Or they may take some ideas from it and bring them back to Android.

It’s an R&D project. Big companies do that as a way of experimenting and pushing innovation. They can afford that. BB cannot. Google isn’t betting their company on fuschia.

What I find interesting about QNX is the modularity of the system. That it can be easily reconfigured, stripped down, etc. Not necessarily something every microkernel based OS provides. What do you think is the cause of the lack of uptake beyond auto? I think it's lack of availability. If, for example, it was freely available on the Raspberry Pi 4, more developers would have experience with it and building drivers for it.

[app_Developer]

What I find interesting about QNX is the modularity of the system. That it can be easily reconfigured, stripped down, etc. Not necessarily something every microkernel based OS provides. What do you think is the cause of the lack of uptake beyond auto? I think it's lack of availability. If, for example, it was freely available on the Raspberry Pi 4, more developers would have experience with it and building drivers for it.

Linux can be configured in so many different ways. Even the scheduler is pluggable. Look at the way it’s used in so many billions of devices in so many different configurations, from Raspberry Pi’s to fleets running billions of containers at Google, and everything in between (including the vast majority of phones)

The reason QNX isn’t used outside of auto is because everyone is quite happy with Linux and QNX doesn’t bring anything new to the table. Even in auto, people who are building ground up are using Linux (see Tesla)

[conite]

What I find interesting about QNX is the modularity of the system. That it can be easily reconfigured, stripped down, etc. Not necessarily something every microkernel based OS provides. What do you think is the cause of the lack of uptake beyond auto? I think it's lack of availability. If, for example, it was freely available on the Raspberry Pi 4, more developers would have experience with it and building drivers for it.

QNX offers nothing more than existing, popular, better supported offerings. You want to pour millions of dollars into reinventing the wheel for some mind-boggling reason.

[app_Developer]

QNX offers nothing more than existing, popular, better supported offerings. You want to pour millions of dollars into reinventing the wheel for some mind-boggling reason.

I should add that we who use QNX in cars are mostly using it because we’ve invested so much in it already. It’s the incumbent choice. It’s what we know and it’s what works.

But again, new car companies like Tesla have no reason to use QNX at all.