A few years ago, I was an avid QNX user. I used the non-commercial desktop version of this wonderfully clean and elegant pure-microkernel operating system for a long period of time, as a desktop operating system. I liked the whole style of this operating system, its Photon user interface, and its excellent package management system. I even wrote a three-page article about it. Sadly, QSS, the company behind QNX, lost all interest in the non-commercial desktop version, and ditched it, leaving only a hard-to-find 30-day evaluation version alive. Community interest dwindled, and so did mine. Despite my lost interest, it saddened me today to learn that QNXZone.com, a community portal for QNX, has been shut down. Read on for a few short thoughts.The note on the website reads: “Due to the significant decrease in the size of the QNX community, diminished interest in QNX for non-commercial projects, this site is no longer providing a community forum or news.” It may have been shutdown months ago, I really would not know. I stopped visiting QNX websites on a regular basis ages ago, simply because nobody was really putting any energy into the whole ‘QNX on the desktop’ idea. And with the parent company losing any interest in that idea, why would the community continue?
I have always had this dream that QSS would continue to ‘silently’ bless the enthusiasts working on the QNX desktop experience by continuing to release the non-commercial version of its operating system (the 30-day evaluation version is fairly similar, but hard to find and inaccessible). It really would not matter much to QSS’ cashflow, since the company’s crown jewels are not QNX itself, but the extended developer tools that run on top of it. Us OS enthusiasts using and extending QNX on the desktop pose no threat at all to QSS’ income. They are after the embedded market, powering in-car computers and nuclear power plants.
It is sad that an operating system with such potential is held back by its parent company. I am sure that were QSS to be more open about running QNX as a desktop, they could easily gain a group of enthusiasts willing to work on improving the whole experience, like they did during the 6.2 days, with ports of all major open source software packages, like Firefox, Thunderbird, an X server, AbiWord, various games, you name it. The beauty of it? It would not cost QSS a dime. Just post a nice download link to a free-for-non-commercial use version of QNX, exclude the development tools, but include the incredible PhotonUI. That is all.
I still receive the QNX Newsletter in my inbox every month, reminding me of what a joy it was to use it on my desktop.
sadly is all that remain for me also, i used QNX as my main desktop for about 6 months in i’d GUESS 2002ish
(http://killerstuff.net/ says 2003, but that really dont sound right), i use that as my point of reference, since i used his irc client.
every 4 – 6 months or so i’ll re-install my 6.x cd and play with qnx for a day or soo.. just the interface is sooo nice, and everything very simple..
same goes for Be, but at least i have haiku to use now..
killerIRC! Now that brings back memories
.
I remember fondly playing with the single-floppy version of QNX. Seems like an age ago, though.
Is there anything like Photon for X out there?
You can create FVWM them to look and behave like Photon:
http://www.deviantart.com/deviation/15518943
I still keep QNX Neutrino 6.2.1 ISO to run it from time to time just for fun:
http://www.deviantart.com/deviation/53541989
It would not cost QSS a dime. Just post a nice download link to a free-for-non-commercial use version of QNX, exclude the development tools, but include the incredible PhotonUI.
Wouldn’t that mean that the open source community would be unable to code for QNX, unless individuals bought a license for the dev tools? I think I’m missing something.
You’re not missing anything. QNX is a proprietary OS with proprietary development tools. That’s not necessarily a bad thing, especially for the embedded market. But it’s impossible to support a community this way. Not when there’s other alternatives like MINIX or Haiku that allow community participation.
Not all operating systems are meant to be developed into desktop platforms. QNX had surprising promise as a desktop, but there was a lot more money to be made elsewhere. In the x86 server market, desktop presence leads to valuable mindshare. But in the embedded real-time market, the desktop is virtually irrelevant.
Nothing new here. uKernels are meant to die.
Nothing new here. uKernels are meant to die.
Die? QNX is doing extremely well in its intended space: the embedded, mission critical environments. QSS is part of Harman International, a company that’s also doing very well, financially.
So, again, die…?
“””
QNX is doing extremely well in its intended space
“””
Granted. Microkernels can do well in niche areas.
It seems that whenever a general purpose OS developer tries to go microkernel, they end up backpedaling and end up with a retrofitted monolithic design.
Except for Minix which, despite what Tanenbaum claims during his combination dog and pony show / microkernel pep rallies, performs abysmally.
And, of course, for The HURD. That shining example of the power of the microkernel design concept.
Except for Minix which, despite what Tanenbaum claims during his combination dog and pony show / microkernel pep rallies, performs abysmally.
Proof? Benchmarks? Wow me, sbergman27.
Well, off the top of my head, how about this:
http://lwn.net/Articles/220255/
well, they aren’t just comparing the performance of the kernel but of the whole os. it seems to be relatively simple to write a fast second generation microkernel, but it’s a huge endeavor to build an usefull new os around this kernel. in most cases, an old os is bolted on top of the microkernel, and the additional layer costs about 5% performance in case of l4linux:
http://os.inf.tu-dresden.de/pubs/sosp97/
even the l4darwin/darbat project, which started as an attempt to swap mach with l4 in the xnu-kernel stopped trying and implements xnu on top of l4:
http://www.ertos.nicta.com.au/publications/papers/Lee_Gray_06.pdf
as such, microkernels are probably more usefull as hardware-abstraction layer than to build a full fledged os.
Who tried to “go” microkernel? 🙂 Do you know anything about QNX design? IIRC, their OS was microkernel from the very beginning. In fact, the kernel was very minimal and everything run in the user space, even drivers.
Amigans surely remember that design, right? IIRC even on Amiga, you could stop and start new driver, eventually without reboot. And that is the point I met with QNX – during 1998, Amiga under the wings of GW, was supposed to switch to new kernel. And QNX was chosen. I met Dan Dodge from QNX in Koeln, and he is really a smart guy.
Back at that time, when later Jim Collas left Amiga, there was switch suddenly announced – switch to Linux kernel. And in 1998 it was us, Amigans, who in fact were scared, because there were clear figures of responsiveness of those two kernel designs. And I even remember Linus joining QNX usenet just to be educated by some russian guys. QNX was heavily used in nuclear plants, automotive industry, or for applications like blood filtering during the surgery. At that time, you surely would not like to have it running under Windows or Linux, none of which was – realtime.
My kudoz to QNX and what it represents – very clever design.
Well, now back to world, which is overtaken by systems like Windows Vista 🙂
Cheers,
Petr
Seeing this news reminds me of the tragic story of Be. Great technology that died because of the short-sightedness of the suits in charge. Granted, BeOS had some great people, but Gassee and Co. got greedy thinking they were set with Apple. We all know how that ended and BeOS was relegated to being an appliance OS, which I find to be an insult for the technology that was behind the BeOS.
Regarding QNX, I thought this OS had potential. It had a very polished interface along with what seemed like a robust environment. QNX brought back some of that passion that I had when I used BeOS, but alas that has all but disappeared especially after the lack of support for those enthusiasts that were behind QNX on the desktop. What a shame that the suits at QSS didn’t see it that way.
So now they claim that the reason they closed the community site was due to a lack of interest. It doesn’t take a high school diploma to realize that if you don’t give your users the support they require to keep them on your platform, they will move on.
QNX does have potential, and is even further than you can imagine. A representative even gave a presentation at BSDCan, where he discussed porting the NetBSD IP stack (see http://www.bsdcan.org for more information on that). And as Thom pointed out to another, QNX is doing quite well, and it is fulfilling its purpose: a high performance real-time OS. Also, its kernel is one of the highest performing kernels, whether microkernel or monolithic, it is even comparable or faster than Linux and the BSDs.
I am aware that they are doing considerably well in the embedded space, but it would have been nice to see them try out for the desktop space (be it corporate or whatnot)….especially now that there seems to be more of an opportunity for alternative operating systems. Like I said, I loved QNX and I really wished that it would have made that leap to the desktop, even if under a closed license that was free for non-commercial use.
I loved QNX, clean code very fast when running on decent hardware. I even wrote a few simple apps for it. Using the freely available version of the time, 6.21 I think.
Maybe if they had been more “open” with their development tools their OS could have gotten more attention, more code hackers would have been interested.
Without community support it seems like they could go the way of the Plan9.. oops I mean Dodo.
“So, again, die…?”
Yup. Die Die Die.
i am not a full fledged developer, so i dont know all the technical terms, but qnx did use gcc, and it did have a photon gui builder. and photon did run on X (or X ran with photon or something along those lines).
@QuadSix50,
“Regarding QNX, I thought this OS had potential. It had a very polished interface along with what seemed like a robust environment. QNX brought back some of that passion that I had when I used BeOS, but alas that has all but disappeared especially after the lack of support for those enthusiasts that were behind QNX on the desktop. What a shame that the suits at QSS didn’t see it that way. ”
sounds to be written by someone whos never used qnx, IMO. Just want to mention beos a little more. Guess i shouldnt have mentioned Be, and direct this convo off topic.
You couldn’t be more wrong. I had actually used QNX back in the early 2000s and I thought the environment had potential on the desktop, especially compared to the state of open source DEs at the time. I had it dual booting with Windows on my work laptop, but since it didn’t support all of my hardware at the time, I removed it and installed Linux. I had even installed it on a computer at home and used it for a while. I had hoped that support would grow for it, but then I noticed that the non-commercial version was not available for download so I just gave a sigh and moved on to something else. I still even have the links to QNXzone and other related sites that I bookmarked from way back when. I guess I might as well get rid of them now as there’s no hope for it heading to the desktop
My analogy to BeOS was only to show the similar feelings I had for QNX. Nothing more. I thought both operating systems had the same potential but ironically they ended up sharing a similar fate.
Edited 2007-08-12 23:11
One thing really interesting about QNX was the OS on a floppy (1.44MB). I really think that developing for a strong limit like that really brings out the coolness factor. QNX so far the the most “complete” OS I’ve seen on a floppy, with a full GUI and web browser. MnuentOS (sic) is also showing promise. The only Linux implementation on a floppy with a GUI took two floppy discs (but managed to fit an entire X server).
Edited 2007-08-12 21:58
The Internet Archive tells that the site was last seen on Sept 6, 2006: http://web.archive.org/web/*/http://www.qnxzone.com
After that there was a DB Error displayed.
Then came the message about the lost interest. I guess it would take too much time to rebuild the database and the owner thought it was not worth it.
Sad to read this QNX site is going down.
Even through I haven’t used QNX in quitte a while,
I liked it, it was fast on some out-dated hardware.
But the software repositries were outdated for quiite a long time already so this doesn’t really surprise me,
still it’s sad to see this to happen.
The non commercial version was driven by only a few people at QSS; among them Chris McKillop (cdm). When they left QNX, the whole NC thing went down the tubes. With the time limited demo now, the steam went out of the community.
OpenQnx.com still goes on, though.
If you can find the ISO and a “valid” code you can install in VMWare. Just set the BIOS clock back to 04 time frame and all is well. Certain sites that like pirates have a torrent with everything you need to install. Runs great and lets you play with a great OS.
…I too had QNX running in a VM…once!
It was a great OS to use, I wish they had done more with the non-commercial desktop
Well, what do you expect.
When you choose a proprietary code then the owner can take it away from you anytime they please.
I don’t care what others use but my main OS will alway be Linux.
And for those admiring QNX so much. Well, it had no drivers and no real apps included with it so no wonder it fit on a floppy.
DSL or Puppy Linux, now that’s something to admire and respect. Not some crap from a proprietary company that only cares about it’s bottom line. Let’s see how they compete with embedded Linux in the years to come. I see a lawsuit brewing – oh, you Linux, you!, You stole our IP!
Must be slow news day to post articles about discontinued QNX
Give it a rest. Not every frigging piece of code written has to be GPLed or Open Sourced.
“””
Give it a rest. Not every frigging piece of code written has to be GPLed or Open Sourced.
“””
True. Open vs closed source is but one factor for the user to consider. Open-source has advantages for the developer, depending upon the situation. And it has advantages for the user. Also depending upon the situation.
But, as much of an advocate of FOSS as I am, I believe that it is doing a disservice to developers, users, and the cause of FOSS itself, to go beyond seeing the FOSS method as one factor, all the way to viewing it as some sort of overriding reason for its use. Or worse, as some sort of moral imperative.
FOSS *needs* competition from closed source, commercial software.
We’ve got the staples down. Browsers, office suites, etc.
But where is the point of sale? Where is the business accounting? Our alternatives don’t even compare well to QuickBooks. Let alone, real business accounting software.
We *need* proprietary software to show us where demand exists… in areas in which we either do not exist at all, or where we totally suck.
I sometimes wonder how the comments on sites like this one might be different if FOSS advocates were required to put in a year trying to advocate FOSS in business environments. Answering the general managers’ questions. And actually having to *sell* the customer on the product. (Parents, siblings, Grandparents, and Aunts and Uncles don’t count.)
I know that the 20 years that I have spent advocating Unix/Linux as a consultant have tempered my naive enthusiasm.
And Unix’s advantages in a DOS world were an easier sell than the much newer concept of “software freedom”.
That said, I ***do*** wish that some group with the requisite talents would write something to run rough-shod over the proprietary point of sale software that I have to support, called Counterpoint, and sold by a company called Radiant Software.
Edited 2007-08-13 02:19
@sbergman27
Heh, nice of you to contract a complex issue into a few paragraphs. Very well done though (really).
But I was just responding to Thom’s desktop gripe.
He chooses an obscure, badly supported desktop like QNX while there is already far superior and better supported Linux desktop (just pick a distro).
BTW, I didn’t say everything has to be OSS.
But in the case of an Operating system, I’d always choose an open OS.
See, the thing is running your computer. If you chose proprietary OS, they can shut you down easy.
It’s a different thing with apps, you can always go buy, download, etc. a different app if the one you’re using is no longer working. Your computer is still usable as a computer for other apps.
No so with the OS.
Example, Vista reduced functionality mode.
Should that happen to you, your computer is kind of useless at that point.
Yeah, you can pop in a Knoppix cd and get on with your work. But I’m talking in general (applicable to most users).
Dude, that would require these people to actually have real jobs
Hmm…I cant recall the name of the software but one of the big supermarkets here (Philippines, Shopwise) is using a *nix based system for their POS.
I don’t care what others use but my main OS will always be Linux.
Linux is absolutely my main OS right now too, and I’m very pleased with it. But I could never have predicted that years ago, and I make no predictions about the future. The OS competition is a wild ride which I enjoy.
Not some crap from a proprietary company that only cares about it’s bottom line.
There is absolutely nothing wrong with producing software for profit. The for-profit model has produced most of the excellent stuff we have today. In fact, software is extremely unique it’s ability to support any kind of free or open source model.
But FOSS needs competition, especially from for-profit software. Imagine if one GNU/Linux vendor had 95% of the desktops. You can download most of their product for free, but they make their income from support. So if they had no competition, even a free-software vendor would be tempted to make their product require a little more support to make sure more people paid them. But as it is now they are well motivated to make it as easy-to-use (i.e. excellent) as possible.
I wish they had continued QNX for the desktop. I thought it was a great little OS that had possibilities, just like BeOS.
Hard to find? “Take advantage of our free 30-day evaluation. Download today!”
It was one of the first things I saw on the front page..
… but the OS scene is really all about MS stuff, Mac OS X, and Linux. It’s a sign of consolidation – these products I mentioned have come too long for a newcomer to have a chance, no matter how clean and nice the design is. Maybe Haiku has some small chance of changing this, but still. And I know the crowd here likes new takes on things, but the *practical* world couldn’t care less.
“Hard to find” 30 day trial happens to be clearly visible on the homepage…
As for not costing QSS a dime, releasing a free version of QNX would cost QSS a fortune.
First they need to pay staff to prepare it for release, and then release it. Staff that could otherwise be working on profit making activities. after the release, they’d have to pay for bandwidth, and im sure there’ be quite a flutter. I don’t know the details of QSS’s financials, but they’re probably not in a position to afford these profitless extras.
With no financial benefit, there’s no real reason for them to release a desktop version. They’re a business, not a charity, regardless of nostalgic desires.
Right, but the thing is that the free desktop version was all about free advertising and creating value in the product.
Lots of people here like playing with OSs, and we also work at jobs. Sometimes we even get to make decisions, or the stuff we play around with gets noticed by other people. That’s viral word of mouth advertising, and mind share. ONX is fighting against embedded Linux, and it would help them if more people knew that they have a proven and stable POSIX compliant OS.
The value part comes into play when there are people in the job market who are familiar with the OS, and there is enough of a community to make software and drivers for it.
Bottom line is that the desktop provided exposure for the company and their product which would translate into brand recognition.
When the non-commercial versions of QNX 6.0 were released QNXStart was the place to be. It was the first independent QNX community. The Quantum / QSS site was good but for a fast answer to a forum post you went to QNXStart. Many of the QSS engineers hung out at QNXStart and could post more freely there without as much concern that their quick post would be interpreted as company policy.
OpenQNX and QNXZone sprang up as latecommers after QNXStart shutdown. There was a Russian site also but not many others.
There were two versions of the first 6.0. One was a small (<50M) desktop-only installer that would put the QNX filesystem into a file on a FAT32 partition. The second version was a ‘full’ system that could be installed on a partition with a QNX filesystem.
QNX Photo had a lot of appeal as it was fast and light to use and easy to install. The whole installation process took less than 3 minutes for the lite version on a Pentium 90 with 32M of RAM. Many saw QNX as a real alternative to other OSs’ of that time.
But QSS was focusing on ‘Internet Appliance’, like Be, and wasted resources. Real-time Linux got better. The lite installer wouldn’t work on NTFS. X got better fonts, QNX didn’t. The 6.1 release was basically a bug fix. 6.2 was delayed. 6.3 had time-limited development tools that turn-off many potential experimenters. In the end the non-commercial version didn’t return anything to QSS so it faded.
PhearBear, cdm where are you now?
Thats the difference between OpenSource and ClosedSource.
ClosedSource Software (including Operating Systems) can die.
OpenSource Software will be alive, if there is at minimum one developer who think, that the software is good and needed to be updated.
You like the desktop version of QNX, there existed people, who have ported GTK+ to QNX and so on.
And all ist lost, because the company don’t want to improve the desktop-QNX and people outside QNX haven’t the source and the rights to doing that.
Thom, you wrote:
It is sad that an operating system with such potential is held back by its parent company.
You have no idea how true that is — or for how long it’s been true. As an undergraduate, I became enamored with QNX’s architecture, and I went to work for QSSL for two subsequent summers. I loved the operating system, and I loved working on it. In particular, the second summer I was there (summer of 1995), SMP boards were starting to become mainstream, so I brought QNX 4 up on SMP by running a separate copy of the operating system on each CPU, writing a little fake-o networking layer to string them together, and then let QNX’s elegant distributed system primitives do the rest. It was really very cool — and a great tribute to the architecture of QNX. And even though it could have shipped with a small amount of additional effort, QSSL didn’t ship it because it would have harmed a partner that was selling multiboard computers. And this to me has always been QNX in a nutshell: great technology, but so hell-bent on playing small ball that it never really amounted to much. Most people have no idea what a great operating system QNX is/was — it is the ultimate counter-argument to Torvalds’ ill-informed and dogmatic position on microkernels — but most people never will have any idea because it has been restrained to the point of irrelevance.
I will always be grateful to Dan and Gord for giving me my first really interesting technical work — and I still very much miss Dan Hildebrand (who was responsible for giving me that opportunity and who passed away in 1998) — but I have never regretted my decision to come to Sun in 1996 instead of QSSL. Sun has plenty of faults, but not being sufficiently aggressive with great technology doesn’t tend to be one of them.
You have no idea how true that is — or for how long it’s been true.
I installed QNX 6.3.0 today, and updated it to 6.3.2a. It’s still great stuff, and recently, all the latest Mozilla packages have been ported to it too. It’s still a joy to use, despite the fact its performance obviously isn’t optimised for this kind of computing.
QNX is how I see my ideal operating system: completely modular.The entire OS is built as a microkernel, including the Photon graphical user interface. Sure, this means a performance penalty, but you get so much in return (stability, clean design, you name it).
The current situation is a shame, but hey, it is the way it is. People parrot Linus Torvalds in his anti-muK attitude, but people fail to realise that Linus actually has an agenda (hint: he needs to sell a monolithic kernel). On top of that, QNX/Harman does what is best for their company (logically!) and a desktop version of QNX doesn’t fit in.
Sad, but hey.
QNX Neutrino is an excellent kernel, it would have been a fine foundation for AmigaOS like it was once planned to be.
After several plan changes made by Gateway this mounted mostly to nothing, although AmigaOS 3.5 did get released (and an Amiga emulation package based on QNX RTP). The reasons are largely unknown, but ex-Gateway employees and execs suggest Microsoft involvement (a Gateway exec also testified against Microsoft, for pressuring the company with Windows pricing).
QSSL, released the non-commercial version and the CEO partipated in Amiga community efforts I was also part of (Phoenix Consortium). I talked to him at one of the meetings, they showed Quake 3 at the meeting, but Dan Dodge’s performance lacked charisma, not at all someone I would expect to build up a desktop user community and he didn’t seem to care enough and so I eventually lost interest. A shame, their developers have done a fine job, creating something with more potential.
Edited 2007-08-13 18:36
boy that takes me back. A floppy version of QNX was my first taste of an alternative OS several years ago. That was back when I was printing out daily the emails from Be during the x86 development cycle.