Fascinating stuff, but not entirely unexpected. As most of you will know, BlackBerry maker Research In Motion is working on a tablet dubbed the BlackPad, and according to several sources reporting to Bloomberg, the device will ship with software written by QNs Software Systems, which RIM purchased earlier this year.
Bloomberg is citing several sources close to the matter (you know, those guys who know everything), and considering it’s Bloomberg, it should be pretty reliable stuff. The BlackPad will be pretty similar to the iPad, except that there won’t be a cellular version. Without WiFi, you’re supposed to hook it up to a mobile phone over BlueTooth (Foleo, anyone?)
It doesn’t require a megahuge brain to figure out that QNX’ software would find its way onto RIM devices in some form or another, and a tablet seems like a good place to start since it wouldn’t be fit for BlackBerry OS anyway. According to the sources, there’s simply too much legacy code in there.
Who would’ve thought that one of my favourite operating systems – in wide use in what I call the real world (medical equipment, weapons systems, nuclear power plants, etc., where toys like Windows NT and Linux have no place) – would eventually find its way onto portable device from a large mobile phone maker?
In any case, the tablet is said to arrive in November.
Well, I didn’t see *this* coming.
I’m sure tremendous amounts of crack cocaine were smoked while making the decision, if this is true (which I doubt). Does anyone develop for QNX?
“Does anyone develop for QNX?” Many people I am one of them.
Pictures or it didn’t happen…actually I’m just joking about that. QNX actually has a much better chance in this mobile arena where no one system utterly dominates yet. Of course I have zero clue about QNX, I’ve always developed on desktop systems and QNX never made it there and wasn’t designed for that anyways.
I’d like to see some comparison between QNX, webos and meego for tablet development. neither android nor ipad excite me much from the development side.
Or rather close minded people just can’t comprehend the possibility of porting high level APIs to another kernel as seen for example with Win9x -> NT -> Linux (the latter via Wine).
If RIM makes BlackBerry APIs available on a QNX kernel, developers won’t have much trouble adapting.
Nokia is also driving a dual kernel strategy with Qt on both Symbian and Linux (MeeGo).
OTOH, if they ported the API’s to Linux, they might still have developers come 2013.
It seems like a decision made to rationalize spending that money on QNX (possibly spiced with a desire to retain “proprietary” secret sauce factor in the overall system). Nothing wrong with that, especially if you work in HR instead of engineering.
Yeah, because all mobile developers write kernel modules and totally care about the used kernel because the kernel matters so much to them….
No offense, but do you know anything about BB development? RIM uses J2ME with a bit of custom stuff underneath (not much really). Its java. The NATIVE api is java…
The kernel they run is about as relevant to a developer as the color of the case.
I’m not talking just about the kernel (neither QNX or Linux are just about the kernel). You could layer BB API’s over a full Linux system. I’ve said it before, but RIM should have implemented the full proprietary BB stack over the MeeGo core.
Ok, but to what end? I realize this blackpad thing may be an entirely different beast, but RIM devices up to now have one and only one public API – J2ME (with their UIApplication framework bolted on to make things feel a bit more native for those that don’t mind loosing portability).
Do you actually think RIM would _want_ to enable developers to write native applications targeting an underlying OS (like MeeGo or QNX or whatever)??? Why on earth would they ever do that? Do you really think they would want to support such a thing? More importantly, do you really think they could support such a thing?
If they isn’t your goal with this whole use Linux thing, what is exactly? Is it just that you like the way Meego looks? I means don’t get me wrong, the underlying platform matters to some degree, but for a phone with a J2ME API the look and feel is pretty much arbitrary, and it certainly doesn’t have a platform dependency. If they wanted it to look like MeeGo, or an iPhone, or whatever, they could – without a whole lot of work (relatively speaking) and without requiring a new stack.
RIM’s core competency is all tied up into BES deployments – they make email devices that can run a few apps as a bonus. They are primarily corporate devices for corporate users with corporate sensibilities.
I’m not saying they don’t need to compete with iOS and Android, they do and they better or they will eventually get their asses handed to them… But at the same time, they need to stay within their own core competency. I frankly don’t think RIM has the expertise to release a device that allows low level API access without blundering badly…
What they need imho, is simply better hardware. Their latest and fastest phone (Torch) has a wimpy 624Mhz Marvell processor, a relatively shitty screen, and just feels cheap and plastic compared to an iPhone and even some of the better Android phones. Better faster hardware is what they need, not a different software stack.
They could cut their RND expenses and just take the platform as is – refocusing their corporate spending on differentiation, management bonuses or whatever may strike their fancy.
Going with QNX, they are paying for everything by themselves.
What platform? Meego is far from production ready as a tablet platform, a Linux/Plasma solution is even further off. Even a Android based solution require quite a lot of RND to get out the door as a competitive tablet platform, as is why the tide of Android tablet has not hit the market yet.
Since QNX are more than self sustained and generates income, a large part of the RND gets paid by QNX customers. And a good part of any improvements and continued development of QNX will also be beneficially in QNXs current market segments, both for existing and potentially new customers.
And it’s not like they are starting from scratch, QNX has already been in various types of equipment with touch based GUIs. In the end their RND expenses should not become too bad, compared to the RND requirements of bringing any of the Linux based solutions to a shipping standard. Adding factors like time to market and the potential for increased income for their QNX product, the decision makes quite a lot of sense.
Edited 2010-08-20 19:39 UTC
Interesting how you twist your own words in retrospect in a desperate attempt to keep your argument (which was based on the alleged consumption of crack cocaine within RIM) valid.
First you claimed that RIM should use Linux to keep developers and after your argument has been debunked you now say that RIM should switch to Linux to cut their R&D costs….
I think with your “expertise” you should apply for the job of RIM boss. Judging from your comments the company would likely bankrupt within a year…
Possibly, esp. since I know nothing of running a company. What seems idiocy on first sight may make total business sense.
I’ve changed my opinion on the QNX tablet, even. It may be the right move for RIM, business-wise, even if it doesn’t make sense when you look at the technology as-such.
But let’s see the tablet come to market first. It would be the worst of all worlds – proprietary like iPad, but without applications.
Edited 2010-08-22 17:21 UTC
I did at one point. Like Thom said, its really widely used in the embeded/real time markets.
I happen to like QNX myself, but I find it hard to imagine it as the basis for a tablet-top where it’s all about the interface. Sure, it could be done, but it would building that interface and the APIs for user apps…. well it seems like a lot of work when other OSs–Android, Chrome, MeeGo, WebOS, etc.–already come with all the interface trimmings.
Steven
True, but most all of those other choices are directly (or indirectly) owned by their competitors. RIM owns QNX. I guess it makes more sense to them to use something they own.
I know that it runs medical equipment and weapons, haven’t heard about nuclear power plants, YET.
There are patches in the kernel to enable controlling lasers, that seems very future real world to me.
(I guess Thom knew such a comment was forthcoming, but I had to bite)
Well, it isn’t a *power* reactor, but QNX does control several General Atomics TRIGA Reactors. And has since the 1990s (when some IC-DOS code was ported to QNX 4.0).
OTOH, I am currently in the process of rewriting that software to run under Linux.
I think this rumor could very well be true! Let my add that I really hope it turns out to be true!
Those comments you read everywhere like “oh please, not another mobile OS”, “they have no chance anyway”, etc. I don’t buy them. RIM has a name, RIM has a lot of marketshare, RIM has an excellent infrastructure for distributing their products all over the world. I totally think they have and deserve a chance. The BlackPad will most probably have a decent WebKit based Browser and some PIM applications. By using an ARM chip it should by default have good battery life. Even only that should make a decent product, and there will sure be more suprises.
But most importantly: There can never be enough operating systems, wouldn’t you agree? 😉
Edited 2010-08-19 16:37 UTC
I would love RIM even more if WeboS was 100% supported in Blackberry. I still have a soft spot on WebOS. As a matter of fact, I don’t worry too much about RIM because they can buy MDG Computers for fixing the sales structures.
Edited 2010-08-19 17:44 UTC
This is cute and all, but will somebody PLEASE SHIP SOMETHING.
You cannot have a successful product selling millions like the iPad until you ruddy well ship something people can actually get their hands on and use.
Gosh, and I really wonder why anybody thinks the iPad is overrated. People are actually enjoying their iPads; there’s nothing more overrated than vapourware.
/rant.
I with you on this. For two years now we have been waiting for the big wave of tablets = ARM tablets, Andoid Tablets, etc. They are shown at Expos, Electronic Shows, and yet – only a dribble has come out. There are the Android toys (KMart), the Always Innovating Touchbook (more a hobby project), but no wave. I am SOOOOOOO sick of hearing about what is going to come out. I want to run DNF on my Ubuntu ARM tablet – now!
p.s., of course, I mean “except the iPad”.
… and I also know about the Streak (tablephone??) and Archos devices – still not the wave.
They are probably waiting for Android 3.0 to make a better initial impact. Therefore they also have time to polish their hardware.
(Be sure to tell this prospective iPad buyers as well – there is no need to get disillusioned and end up buing iPad in a weak moment)
This gap in time should cause a *real* wave eventually.
What’s funny is that the only devices that have managed to make the cut are some pseudo-illegal hackjobs. I’m slightly disappointed with the tech industry.
This is always the problem. It seems lots of companies can build the hell out of hardware. But when it comes to actually *operating* that hardware, especially where software is involved, the companies fall on their faces. The hardware industry really really wants the “build it and they will come” paradigm, but that never happens. So we get really good hardware that doesn’t go anywhere because the investment was made in the wrong place.
Which makes me think about the movie “Iron Man”. The movie concentrated heavily on the assembly of the suit, but the whole software side was by and large poo-poo’d (just upload the AI and everything works??).
Most companies still haven’t gotten up to speed with the proper valuing of the intangibles for the mass market. And looking at market capitalization…the companies that do this properly are the ones that go huge. And the chinese build all the hardware.
Except that in the tablet space a lot of phone maker heavyweights are in the ring, like Samsung, Motorola, Nokia, HP Palm, now RIM and of course Apple. This isn’t a suprise since the smartphone operating systems and tablet operating systems have so much in common, they are almost identical. And those phone makers do a good job of getting HW and SW work together. Just look how many Android phones are getting released. The staccato of new models is incredible and the phone makers develop some kind of routine that will help them a lot when releasing their tablets.
You just need to be a little more patient! 😉
Edited 2010-08-19 22:20 UTC
> medical equipment, weapons systems, nuclear power
> plants, etc., where toys like Windows NT and Linux
> have no place
Is it really so difficult to talk positively about something, without bashing something else?
For the record, that those OSs have no place there is just a lie. Many mission critical things are running those, like it or not, although they may not be suited for some tasks, they excel in others.
But I suppose you already know…
jaco
In Thom’s defense, I’m with him in thinking that Windows NT doesn’t really belong in mission-critical or even financial situations, though probably not for the same reasons. I don’t consider NT a toy and I’m not concerned that much about stability, but I am concerned about security as it has long been dropped from support by Microsoft. I for one don’t want to see missiles launched at our enemy-of-the-week because some rogue internet terrorist found a back door in a 15 year old unmaintained OS.
As for Linux, my personal opinion has always been that it’s just barely good enough for me as a daily desktop OS; I prefer OS X but I won’t turn my nose up at Slackware or Linux Mint or (fill in the blank). Where it shines (given a well-tweaked kernel) is in the embedded space. I also think Android is the best prospect for becoming the dominant mobile OS for years to come, Oracle lawsuit notwithstanding. That doesn’t help the “toy” accusation of course.
In the server space, I have always and will always prefer FreeBSD. It’s hard to get more rock-solid than that.
As for QNX, well I’m glad to see it potentially put to good use in a consumer device. Now if only RIM would start using it as the backend of the BlackBerry OS!
I would say you’re probably correct. There are a lot more instances of inappropriate OS choice, then we would really like to know. I have a nightmare that one day I’ll wake up in a hospital and have my breathing machine bluescreen.
Linux is more acceptable to me in some of those use cases, as another poster posted its making its way in there. But if you ask me what I want to run the breathing machine I’m hooked up to, I’d prefer it to run Qnx.
This one is a strange bird.
I’ve worked with a guy who deploys systems for hydro power, natural gas wellhead controls, diesel engine monitoring and is currently working on a vehicle mounted astral telescope system project. They’ve very successfully redeployed from rtos to linux several years ago. The biggest reason? Cost control. And it works. One customer forced them to use windows as the controller, at that time they had to rewrite the whole windows scheduler to get it stable for real time monitoring (9 month effort).
What I’d really love is Android api and frontend hosted on top of slimmed down QNX. That would be a bummer!
Do you have some reason to expect QNX kernel to perform better than Linux kernel? Or to have better OpenGL drivers than Linux?
QNX was designed to be an embedded OS. They tend to be as slim and quick as possible.
“Do you have some reason to expect QNX kernel to perform better than Linux kernel?”
The QNX kernel is a thing of genius and is a different type of kernel model than the linux kernel. The linix kernel is a Monolithic kernel, while QNX is a microkernel and I personally feel that it is better suited for this type of environment. I am slightly bias, but the only reason i fell in love with QNX is because i’ve tried sooooo soooo many OS’s and while they all have their merits, I keep coming back to QNX for many reasons.
in the real time space it does preform better, even than real time adaptations of the linux kernel (wind river’s and others)
knowing a little about the QNX history I wouldn’t mind see this happening – looks indeed like a good match.
Personally, I’ll jump on it anyone releases a 14″ (A4 sized) tablet that can run a browser, a pdf viewer, and a picture/cbr viewer.. doesn’t need anything else.
Make one – I’ll buy
You really don’t like QNX, do you? I’m genuinely curious, did you have a bad experience with it in the past? Do you hate someone who works/worked for them? Do you have a strange aversion to lesser-used consonants like Q and X?
It’s a proprietary operating system that overlaps completely with several open alternatives (Linux, BSD). I don’t see a reason to “hate” it anymore than I see reason to hate OS/2, but I don’t see a reason to boost it either.
Granted I haven’t touched it in years, but you used to be able to fit QNX4 plus the photon GUI with a nearly complete user space into about 3MB… And photon was nearly as good as a full XWindows stack (better in some ways). It was rigidly developed with modularity in mind, and proved to be extremely flexible as an embeded OS.
But most importantly (and this is why I think RIM bought it) it was capable of hard/soft realtime operation. That imo is a very valuable property for a smart phone if exploited properly.
Anyway, my point is while there is some overlap, it most certainly isn’t complete, and QNX brings some things to the table that Linux/BSD simple don’t.
RIM probably better things to do than “asking” Oracle to sue them.
I remember running QNX RTOS back in the day as my desktop OS. Photon was an incredibly slick GUI for its time. Does anybody else remember CDM? Whatever happened to him after he left QSSL?
Photon was pretty slick.
***DELETED***
Edited 2010-08-20 11:12 UTC