An EE Times straw poll of embedded device makers in Europe found that while last year 50% said they planned to use Linux, that number had dropped to 33%. 26% said they planned to use a commercial OS, up from 15% a year ago.
An EE Times straw poll of embedded device makers in Europe found that while last year 50% said they planned to use Linux, that number had dropped to 33%. 26% said they planned to use a commercial OS, up from 15% a year ago.
15% of 1000000 is a lot more than 15% of 10
Without that the numbers mean nothing
Now, did I understood it right from the website but is this really nothing more than a website “‘straw poll’ with a self-selecting audience”?
http://www.eetimes.eu/quickPolls/archive/showPoll.jhtml?surveyno=26…
A poll on OSAlert would be as informative and scientific then?
Having worked in the embedded industry, a commercial OS is definitely the way to go (even if it is linux
). My friend works at Windriver doing their linux distro. The skillset is just not there in every company to have their own linux guru to handle everything from device modules, file systems… how to optimize things…
Not to mention a good development environment…
And then if you’re going for a commerical build, I don’t think the price difference are that huge between them so you’re really just going on the best embedded OS to suit your needs.
Without knowing how big the sample was and how the questions was asked any statistic is rather useless.
It is very easy to bias a result any way you want by carefully selecting your sample and and take care how you design your questions.
my sentiments
I always considered QNX to be a viable platform. They’ve been doing the embedded thing since the beginning and I wouldn’t be surprised if that’s where most manufacturers will go.
As others have pointed out, this may be meaningless. The article points out that the sample was self selecting. Even if it was a randomized sample, it is difficult to tell what the uncertainties are.
But now to my main point.
If this shift is real, the question is why? Without knowing why, Linux developers cannot fix the problem, if there is even a problem to fix.
Fixable problems may fall in several categories. The two that I can think of off the top of my head: maybe the tried implementing an embedded device using Linux, but found that it was too difficult to do. If they can elaborate on those difficulties, it may be possible for external developers to find solutions. Of course, that is a huge thing to hope for since embedded systems cover an enormous range of devices. The other problem may be the reception of their customers. A case in point may be illustrated by those netbook return rates, which isn’t really embedded but is approaching the realm of embedded systems. If the client finds it too unfamiliar or somehow unreliable, it may be possible to make the user interface more familiar or more underlying software more reliable.
Non-fixable problems may be suitability to task. If the market is shifting away from complex multi-function devices and towards less expensive single-function devices, then Linux is probably not suitable. Cost constrained applications may use a microprocessor that Linux won’t even run on, or Linux may increase the demands on memory or the CPU to unrealistic levels. After all, some stuff can get away with an 8-bit microcontroller. Other applications may require a more powerful CPU, but you would save money if you could use 4 MB RAM instead of 8 MB of RAM.
Alas, surveys like this don’t offer any insight into those sorts of issues.
Not surprising. But it’s not Linux’s fault. It’s the GPL’s fault. Many embedded system makers are scared away by the GPL.
It all depends on the number of units you’re expecting to sell. For instance, in Germany, a Windows CE licence costs about 10 ~Ac^A`^Anot, now do the simple math.. If you’re expecting to sell 10k units, that’s A LOT OF MONEY.
For the money you save with the licences, you can easily hire some skilled Linux/*BSD people.
WinCE is actualy a very well done and incredibly customizable embeded OS (and desktop os if you have enough skills with platform builder and the right board suport pachage). It is fast, reliable, mostly shared source, and above all it is insanly customizable. if I had to pick an embeded OS (and i did as i develop apps for them and mess around with embeded OS’s for fun) I pick WinCE and QNX. I am anxiously awaiting WinCE 7 and QNX 6.4. (I am even more excited about QNX 6.4.1 since it will have CDT 5.0 and Eclipse 3.4 integrated in Qnx SDP 6.4.1)
Why did my comment score get lowered from 2 to 0? Cause people don’t like to hear the truth?
The GPL really does scare embedded device makers away. Because oftentimes embedded devices have industry trade secrets in the hardware. The fact that they would have to GPL any mods they made to the kernel to improve performance on that hardware, or support certain new features of that hardware, scares them away. It’s the same type of scenario that prevents nvidia from open sourcing their drivers. Cause that would give away trade secrets about their hardware.
Embedded system vendors really are scared away by the GPL. Many of them simply don’t want to be at the mercy of a license that would force them to open source any changes they made to the kernel.
The GPL really is hurting Linux in the embedded systems market. Which is sad, because Linux would make a great OS for embedded systems.
If you think I am wrong, then tell me why I am wrong. Don’t just lower my comment score cause you don’t agree with me. Tell me why you don’t agree with my analysis.
Fine, I’ll tell you why you are wrong. First lets start with Nvidia just to get it out of the way. Nvidia CAN’T open up its drivers as it licenses some of the tech from other companies. Just like Sun licensed some Java libraries from other companies. But that is besides the point.
First, most embedded devices are probably using one of several standard CPU’s which already have Linux support. Most companies are NOT producing their own chips, so there isn’t any trade secrets there for them to give away.
I think the main reason why this is happening, if it is a problem at all, is due to money. There are a few OS’s which will run in less memory than linux. That means less memory on the board and less cost. However, in at least the one instance I know of, the substitute OS was not as robust as linux. It may also be the case that these other chips support other OS’s better than linux. That may change as Intel ramps up their new low power chip line. Again, the survey didn’t really get specific so who knows what the reasons are unless we ask.
nvidia has specifically stated that their drivers contain hints about their own intellectual property, and that’s part of the reason why they won’t open them. So at least some of the reason is protection of their own intellectual property.
When it comes to embedded systems, several companies I know of have custom chips built for them, and make kernel tweaks to support those custom chips better. Open sourcing those changes would give away a competitive advantage they have based on some of the chip designs they have had done. And that’s why they aren’t going to use Linux.
There is also another issue, which is that they want someone they can reliably count on for support. As of right now, there are no Linux vendors specializing in embedded Linux. No commercial vendors anyway. If that changes, their might be more interest from hardware vendors in embedded Linux.
Edited 2008-10-08 01:13 UTC
Wind River, Timesys, MontaVista, Intel… There are quite a few people doing embedded linux. Maybe one possible explanation for this article is that Europe is a small embedded Linux market??
Or because you didn’t put any proof to your assertions.
Which explains the big success of *BSD in embedded devices?
That was a sarcasm, in case you had a doubt..
Since when is proving proof required in osnews (or any discussion site) comments? It’s certainly *helpful*, but a publication this is not.
On the company and their resources. Many startups also use netBSD as an embedded OS for their devices also. This is unencumbered by the GPL. GPL is not compatible with some corporations b/c of what they deem as secret. Whether it’s the pact the made with the chip or device makers for drivers, or a left wing belief in security by obscurity… they all have their reasons whether valid or not.
from reading the blurb. 50% said they might, and out of these ultimately about 70% did – it was foreseeable that not all would adopt and I think 70% is actually not bad..?
so what? It is better in every respect. If they do not want to contribute, then they could buy an exiting solution with support. But …. this results to more expensive products.
“Better in every respect”? I believe even Linux developers will disagree with that sentiment.
http://www.linuxdevices.com/news/NS3856352298.html