Linux in Embedded OSs 75
Carnage4Life writes "ZDNet has an article on the viability of Linux as the future belle of embedded OSes. It quotes Linus as mentioning the fact that since license fees are free and developer support is relatively abundant, Linux is a prime candidate for startups creating Web appliances and the like. It lists Sony's, tiVo, Lineo, Transmeta, Intel and national Semiconductor as major industry players who are embracing embedded Linux. "
Re:ZDNot (Score:1)
You only have a single judge in a case. After the facts are laid then, the judge decides one way or the other. I agree with your point of having two sides to a story, but that usually is an analogy of having two lawyers who represent each side. As you say It's opinions that bias something - not facts, but it's bias to choose what facts to point out. Having one bias to one side and another bias to anther, will help show the most facts of both sides, and then you can get a better picture. Of course this doesn't always work. If one lawyer is much better than the other, then you don't get an equal showing of facts. That's probably the problem with our (US) legal system. But that's another story.
Again, I agree that the press should always have reporters that favor each side. But opinions are hard to subdue. Let the reporters fight, and then we can get the best news. My news-paper has reporters that are both Democrats and Republicans, and I love to read the difference in opinions. I enjoy reading Pro-MS as much or even better than I do for Pro-Linux. Being Pro-Linux myself, I find that I too am bias, and like to find the ways to make Linux better. I read that Linux has trouble with configuring the Video, and I say to myself "yes it does, let's fix that".
Steven Rostedt
Re:Here's what the GPL says: (Score:2)
How should we go about contacting Stallman to see if he is in agreement with this and hopefully get either a clarification or modification to the document?
On a different note, as I understand it, the BSD license doesn't suffer from this problem due to its 'non-viral' nature. Is this correct? Are there any BSD-based embedded solutions out there? If so, is source available?
What does "Embedded" mean these days? (Score:4)
Traditionally, the term "embedded" denoted a system with minimal (or no) UI. Code for such a system could not be developed natively so development was done on a host and 'installed' on the target via an emulator, debug port, PROMs, etc. Typical examples are heating/cooling systems, flight control systems and microwave ovens. In my case, they are traffic controllers and other associated equipment.
Things like Web Pads and Palm Pilots seem to blur the distinction between embedded and non-embedded systems. While code is still developed on a host (for the most part), many of these devices have quite sophisticated UI's (even GUI's). What do you think? Should they still be considered embedded or should there be some new term to describe them? Maybe I'm just old-fashioned, but I have a hard time thinking of anything with a GUI as embedded.
Re:BSD vs GPL for embedded systems (Score:1)
Application Appliances (Score:2)
The word people are using for a bunch of this stuff is "appliance". This generally refers to the "instant-on" category, where you don't see an OS "booting", and then you don't see a shell either, except to implement what is generally considered an Application. Problem is, appliance by itself is not distinguished from washing machines, so there's the urge to say information appliance which works great for palms and webpads but not for "gameboys" or who knows what. How about application appliances, or "app-apps". Catchy?
Of course... (Score:1)
I'll shut up now.
The GPL makes Linux the wrong choice. (Score:2)
As an embedded system developer, you cannot afford to use anything GPLed. PicoBSD, QNX, and other OSes without "poison pill" licenses are better choices in this realm.
--Brett Glass
Re:Too much OS? (Score:2)
My take on it isn't actually that at all - more that Palms succeeded mostly due to having just enough functionality while being cheaper. WinCE boxes are more powerful but the benefits aren't always clear to the user comparing the two in Dixons, while the Palm costs £100 less. So which do they buy?
They're nice machines, sure, but I don't think the interface is actually as nice as some people seem to think.
Greg
OK, but why is it REALLY being chosen? (Score:2)
I remember moans about covermounted applications 5-6 years ago with the Amiga magazines. The main problem from the developers was that, when something was essentially free, it doesn't have to be better, just good enough. And good enough is a pretty low threshold - people will put up with quite a bit if it saves them money.
I'm glad to see Linux expanding into new markets. It's not for me right now, but it's good to see community development having such a huge impact. But is it chosen because it's better or because it's good enough and free?
Sure, there's the cost of porting (well, sometimes) and the cost of striping it down to what you need (well, sometimes again) but after that it's free. You give the code back - not a problem, it was cheap to modify anyway in the corporate scheme of things - and it's suddenly zero unit cost as well as short development time. Plus, sites like slashdot give them free publicity! Now, how many bean counters are going to work against that one?
I'm sure there are companies who've chosen Linux purely thanks to technical superiority. Really, I am. But I suspect most are simply using a cheap shortcut that generates them good publicity via the Linux bandwagon.
Greg
Re:ZDNot (Score:1)
I believe we're arguing over what the persuasive writing style is. I agree with you over this style - it is intended to persuade someone to your side. Typically facts supporting the other side are omitted, opinions are intermixed with facts, ad nauseum. Editorials are almost always persuasive.
However, an informative writing style is more akin to reading a report about a gas leak downtown that levelled a 3 block radius. There won't be much more than facts: there will certainly be no blame assigned. Quotes will typically be from both sides (if one side cannot/would not comment, it is usually noted). Reading the police reports on the back pages is an example of this (albeit dry!) style of writing, as is a technical "white paper".
As you say It's opinions that bias something - not facts, but it's bias to choose what facts to point out.
This is true enough. However, unbiased reporting, in contrast, would be reporting all facts one could find at time of publication regardless of whether which position they support.
Re:BSD vs GPL for embedded systems (Score:2)
Windows CE (Score:5)
See the Linux CE Project [linuxce.org].
How about "Minis" (Score:1)
Just an idea, since it is getting hard to distinguish "embedded" with actual normal software. Heck, soon a Palm may be as powerful as today's laptops.
We really need a new term to be able to remove this confusion. At my work, we really do write "embedded" and RT systems. The software runs aircrafts and that is probably the most RT and embedded as you can get. Most of the software needs to fit in less than a Meg, since the hardware needs to take tremendous evironment abuses and must only produce minimal heat, we can't use the stuff you get at Comp USA, and thus we are limited.
Steven Rostedt
Re:Too much OS? (Score:1)
Anyway, there is a reason that PalmOS beat WinCE, and it wasn't necessarily the normal Microsoft bugginess. It was the fact that PalmOS just does what is needed, and doesn't have a bunch of extra fluff brought down from bigger machines. Would an embedded-linux be any different?
Part of the reason that WinCE does so poorly against the PalmOS has to do with its interface. A Start menu and desktop metafor simply takes up too much space to be useful on a small PDA screen. This is a separate (and IMHO more important) apart from bugginess or code bloat. Since Linux is at its heart a CLI OS, most processor cycles AFAIK are used up the X Window system and processes and services which would not be needed on a PDA and therefore would not need to be compiled.
I'm no Linux expert but my guess is that you can pretty much put up any interface you want on top of a slimmed down Linux kernel. Palm Computing for example is working with Symbian [symbian.com] to run the PalmOS GUI on top of the Epoc32 kernel. There should be no reason, other than perhaps licensing, why someone couldn't do the same thing with Linux.
Similarly, it should be possible to create a webpad type device that would simply run Mozilla on top of a Linux kernel recompiled with only enough support for whatever was needed to run a browser plus wireless IR or radio support.
Multi-Jesse Theory (Score:2)
poison pill vs. vitamins (Score:2)
So, the question in this case is: is there any inherent problem with the GPL? It doesn't seem that there is because Transmeta has crossed this bridge. If there is no problem, that needs to be made clear by... ? by advocates who wish to see either linux or the GPL adopted.
But, Embedded Guy, have you looked at BSD licensed stuff at all? Also, who are you embed with? ha ha.
Re:How about "Minis" (Score:1)
Main Entry: minicomputer
Function: noun
Date: 1968
: a small computer that is intermediate between a microcomputer and a mainframe in size, speed, and capacity, that can support time-sharing, and that is often dedicated to a single application
Re:BSD vs GPL for embedded systems (Score:1)
Re:ZDNot (Score:2)
I submitted the story and I disagree with your post. Yours is a case of ignoring the message because you do not like the messenger. The primary reason you give for disliking ZDNet seems to be Jesse Berst who isn't even a reporter but an opinion columnist this is similar to disregarding news from CNN (which is the premier news service in the world and watched in over 100 countries) because they have a show with Johnny Cochran as a host. I primarily read ZDNet because they report more Windows bugs and security leaks faster than any other mainstream Tech news provider (not before Bugtraq but they aren't mainstream), they are good at reporting industry trends and Dvorak's opinions are interesting. Up until i started reading Slashdot I had never heard of Jesse Berst and after I read one of his articles I realized he didn't know his head from a hole in the ground and stopped reading his opinion pieces (Linus as hardware man of the year?!?). So dismissing ZDNet because of Jesse Berst is not only an illogical decision but also robs you of hearing different views on the technology industry beside linux r00lz, MSFT sUx that fills the threads on Slashdot.
The main reason I submitted the story was because I though it was neat that Playstation 2 would run linux as well as web appliances from Intel and National Semiconductor. On slashdot I've read discussions on Lineo and Transmeta several times but didn't know about Sony, Intel or National Semiconductor and their work on using embedded linux. Therefore I thought it would be nice to hear the views of informed hardware people on these developments.
Thanks for your time
Re:Here's what the GPL says: (Score:1)
Some of the BSD licenses (there are variations) only require acknowledgement in the product and/or advertising. The BSD license does not place further restrictions on derivative works.
The eCos license is interesting in that it requires modifications to the "covered" (i.e. the OS) code be published in source form but accomodates the combination of non-covered (i.e. proprietary) code and covered code in commercial products. Way to go Cygnus!
So it looks like BSD and Cygnus sources are fertile ground for embedded systems work. More so than GPL. And do I even have to say, IANAL.
Re:BSD vs GPL for embedded systems (Score:1)
btw, thanks for looking into the BSD & eCos licenses. I'm curious how Cygnus gets away with licensing eCos that way since it's based on Linux. Seems like the viral nature of the GPL would prevent that.
I was considering opening an "Ask Slashdot" thread to get more folks involved in this discussion. Do you think that would be worthwile?
Re:BSD vs GPL for embedded systems (Score:1)
I agree with your point that the intention of the GPL is probably not to exclude use in embedded systems but its wording makes it unsuitable. On the other hand, the GPL was intended to draw developers in to the open source game so maybe we would be breaking the intent by using GPL'd code in new products without giving something back to the community. But what I've just realized in the last couple days is that the other open source licenses (BSD, eCos, and some others) aren't as restrictive as the GPL.
As for eCos, even though it is put out by Cygnus, I don't think it is based on Linux or other GPL code. I think those guys built their embedded OS from scratch. That's why they aren't bound by the GPL.
Redundant (Score:1)
Sure, the Transmeta processor opens lots of doors to get rid of the current incarnation of the computer -- soon enough, the vast majority of home computer users won't be using a beige box, they'll be using webpads for their browsing, and bitchin' phones for their e-mail, and not this "all-in-one" machine.
But advances in technology happen on a daily basis. What does this article teach us that we didn't know already?
ZDNot (Score:3)
It may be flamebait, but I think ZDNet has become see-through in the lip-service it's paying to linux. I point you no further than Jesse Berst, who in his "berst alerts" went from "linux sucks - it'll never compete with windows!" to "I always said linux could be a contender" to "linux rulez" in a span of 5 months. Something ZDNet should try someday: balanced reporting. It's a novel concept I urge any reporter (slashdot included) to employ - *represent both sides equally and without bias*.
An important point was missed (Score:4)
Startrekking? (Score:1)
Imaging you just bringing such a pad to school/work, checking in on a main-frame, and doing your work. With an internet connection, these things will need very little harddisk space (or none).
Also, technology will make them smaller, compare the calculator of 20 years ago with the ones we see now! Imagine now such a pad, not thicker than 2 mm, and at a very low weight. If you need anything typed you can connect a keyboard to it (or the cool "wireless keyboard without the keyboard" thingie).
And now for the best part: this all under Linux! Now I understand why those Startrek computers never crash...
The Jesse Berst Syndrome is Spreading... (Score:1)
Is it just me, or? (Score:1)
HDTV's and such (Score:1)
Apart from this being the one thing that could stop the world population from watching tele (would you boot a tv to windows?) they also now have competition.
The real shame here is that whichever 'side' gets the lions share of the market they will have made a GUI the de-facto standard whether its X11 style, palm style or even PSX style; you won't be able to use your TV unless you have this GUI on it. (A bit like Betamax dying and VHS prevailing)
What I would like to see is this: a range of television sets which can be configured with whatever you like (ie: O/S / GUI) so you don't have to have it forced down your throat.
And first and foremost *all* the GUI's should be made compatable to a _fair_ industry standard so everyone can compete on equal grounds.
Lastly why isn't England going to get decent HDTV's until last?
Re:Too much OS? (Score:1)
One difficulty which I think you may have missed is that PDAs are almost always graphical devices. The Palm certainly is. And obviously X Windows is far too large to fit on a PDA. (And has far too much unneeded functionality.
A true CLI is a royal pain in the ass with only a stylus.
I'm sure it could be done. I just suspect that by the time you through out everything that wasn't needed on a PDA, and added in extra things that were needed. (Some sort of Grafitti version, a simple graphics interface, etc.) you'd end up with something that wasn't particularly compatible with mainstream Linux, and was bigger than something written from the ground up.
It probably depends on what sort of embedded device you are looking at. If you are looking for a real computing platform, it might be a good idea. But do you really need multitasking on something like a palm-pilot? Users? A file-system, even? If not, then why Linux?
Re:ZDNot (Score:1)
But as for your statement
*represent both sides equally and without bias*
I would like to see how could you do that. Can someone represent a side with being bias? Isn't that kind of an oxymoron? Or do you mean just mean have a bias MS person and a bias Linux/free software person. Of course I don't know anyone that would like to write on
Steven Rostedt
Re:Linux is still missing a few pieces... (Score:1)
5) Streaming video codecs/player
Looking at the list of vendors working on Linux-based "info appliances" surely at least some of these companies realize that support for streaming media is a must for a consumer-oriented information or Internet appliances. The Missing Broadcasting Link...
Re:Windows CE (Score:3)
Same for NetBSD:
See NetBSD/hpcmips [netbsd.org] for MIPS based PDAs.
There's of course a SH3 port as well: NetBSD/SH3 [netbsd.org], but they currently don't support any SH3-handhelds (AFAIK).
Here's what the GPL says: (Score:2)
2. You may modify your copy or copies of the Program or any portion of it, thus forming a work based on the Program, and copy and distribute such
modifications or work under the terms of Section 1 above, provided that you also meet all of these conditions:
a) You must cause the modified files to carry prominent notices stating that you changed the files and the date of any change.
b) You must cause any work that you distribute or publish, that in whole or in part contains or is derived from the Program or any part thereof, to be licensed as a whole at no charge to all third parties under the terms of this License.
c) If the modified program normally reads commands interactively when run, you must cause it, when started running for such interactive use in the most ordinary way, to print or display an announcement including an appropriate copyright notice and a notice that there is no warranty (or else, saying that you provide a warranty) and that users may redistribute the program under these conditions, and telling the user how to view a copy of this License. (Exception: if the Program itself is interactive but does not normally print such an announcement, your work based on the Program is not required to print an announcement.)
So far, so good. Now here's the troublesome part:
These requirements apply to the modified work as a whole. If identifiable sections of that work are not derived from the Program, and can be
reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you
distribute them as separate works. But when you distribute the same sections as part of a whole which is a work based on the Program, the distribution of the whole must be on the terms of this License, whose permissions for other licensees extend to the entire whole, and thus to each and every part regardless of who wrote it.
Now I would argue that my traffic controller software can be "considered [an] independent and separate work" and therefore should not fall under the GPL. But the last sentence seems to imply the opposite since it is distributed as part of a whole (Linux and my app are burned into the same PROMs). But if you look carefully you see that it only applies if it is a "work based on the program". The question is if a program is compiled and linked and loaded in the same set of PROMs as GPL'd code, is it considered a "work based on the program" or a "separate work"? I would argue the latter, but it's not entirely clear.
Clearly, if my app is non-embedded and I distribute it separately in binary-only form (Netscape, eg), I'm not bound by the GPL. I'm sure FSF never considered embedded code specifically but I really don't think they intended embedded s/w to be treated any differently than non-embedded just because it is distributed differently.
What I would like to see is for the license to be modified so that it explicitly deals with the case of embedded code. RMS, are you out there? What was your intention?
Re:Linux is still missing a few pieces... (Score:1)
with the support of more and more IPO Linux companies.
Support is expanding from the server strength that Linux has to embedded and desktop systems.
Let's also not forget that the strength that Linux will receive will be from the developing
countries of the world like S.America, India, China and Russia.
These places are already pillars of strength for Linux development
(e.g GNOME's Icaza from Mexico) but we have only seen the tip of the iceberg.
Re:Too much OS? (Score:2)
I agree about WinCE being a hog, but take a look at uClinux [uclinux.org]. It is a modified 2.0.38 kernel with virtual memory stripped out. It fits in 2MB EPROMs with lots of room left over for your apps. The difference is that WinCE is a big monolithic program (just like desktop windows), whereas Linux (or any Unix) is modular. You only include the pieces you need.
You wouldn't have to add it to the kernel, you could implement something like this as another program that runs on the (more or less) standard Linux kernel. Sorta like an X replacement. It could even be the PalmOS API if you wanted it to be.
But do you really need multitasking on something like a palm-pilot?
Probably not, but there are lots of embedded applications that would benefit from multitasking.
Users?
Sure I can see some applications for multiple users in embedded devices. At the very least, it would be useful to have a normal user account and a root account (for debugging, system maintenance, etc.).
A file-system, even?
Absolutely! A RAMdisk, for example. Or the ability to mount an NFS volume.
If not, then why Linux?
Well, here are a few reasons. With Linux you get a stable kernel, with a good scheduler and very good networking. You get the source code. You can develop (and debug) nearly all of your application on your desktop (also running Linux, of course) and only port to the target when you're done. And you get portability. Pretty good reasons, imo.
embedded linux (Score:2)
I find this article interesting, as I've watched the attention paid to embedded linux grow since the summer, a time when it was a far less mature market. The splitting off of Lineo from Caldera was one of the more interesting developments that I've seen, although the purchase of cygnus by Redhat also added to the legitimacy of the market. With the recent CPU announcement by the Silicon Valley 'wunderkind' Transmeta, embedded linux has really entered a more mainstream commercial phase.
The article on ZDNet helped to point out some of the majors issues in the industry. Who wants to pay for the 250 licenses needed for even a smallscale embedded application. It may mean less to the typical consumer who might pay $90 for a copy of windows, but I really think the cost issue associated with embedded linux applications is way more relevant than in the server or desktop market. It doesn't make a big difference if you pay $250 for software on a $16,000 server compared to buying a Software license for say $30-40 on a $200 portable web device.
Another good point they brought up is ease of use. It doesn't matter to the endconsumer what OS the product uses, as long as it performs as expected, and without high incidence of failure (high being relative) Many of the companies in the embedded linux software side of the market provide technical support as a primary source of income (linuxcare, montavista, etc) Although some companies such as Lineo appear to be pursuing a more license based approach (although still providing plenty of technical support) Hopefully the end result will still be a savings of cost, but the high amount of technical support provided by these companies should help to offset any difficulty in actually using linux as the core of these applications.
Other benefits not explored in the article include portability, the 'open source' ability to take an already extant solution and modify to suit your individual projects, as well as the ability to advance or modify a project that may have been discontinued, or to correct bugs without waiting for the original programmers to take action.
Embedded projects are often so unique that a non-custom solution is often useless. The ability to customize linux combined with its low cost of use really makes this appropriate for this industry. I don't think the ZDNet article was worthless, and it in fact brought to light some points that people might not realize about embedded applications. It may seem like its just repeating stuff that people have always been talking about, but I think it was worth repeating and highlighting for this specific market. I see the success of embedded linux continuing to grow, and perhaps to exceed the growth that linux is seeing in other markets.
If you would like to learn more about embedded linux and whats going on in the industry, please check out Linuxdevices.com [linuxdevices.com]
Re:Linux is still missing a few pieces... (Score:1)
Linux needs -
1) Flash drivers
2) a viable Flash File System
3) Bootloader from flash/ROM
Many embedded systems (PC104 comes to mind) present the flash chips as an IDE interface, so the standard linux disk drivers work fine. The trouble with flash chips is the ~100K rewrite limit, but you can usually just keep all your dymanic data in a ramdrive to solve this problem. Check out the initrd kernel options.
Re:Here's what the GPL says: (Score:2)
In our product (part of which functions as a router) we would like to use components of the IP stack found in Linux distributions. Doing so would only require minor modifications (if any) to the GPL'd code to adapt it to our system. The rest of our system would not be derived from any GPL'd code. But the hitch is that our software is distributed in ROM in a piece of physical hardware. It is not possible to distribute our system software without the GPL components as "separate works". Hence the terms of the GPL seem to extend to the entire work and the poison pill goes to work. If the commercial embedded systems market is going to work with the GPL, I think the GPL needs to change the way it defines what is considered a separate work.
Incidentally, I think the embedded systems world could benefit from some open source platform and component developments (like eCos) but the GPL is an inappropriate license for the reasons describe above.
Re:ZDNot (Score:3)
OK, at the risk of being flame-bait... may I point this out: has it ever occurred to you that Jesse Berst might have gone through a "conversion"? Is it so inconceivable for a person to believe in popular FUD against Linux and to speak out his belief? Is it so inconceivable that this same person may have found out eventually that there is more to Linux than the FUD would have people believe? You have to realize that reporters usually do NOT have the means nor the time to do a 100% accurate research about their subject. They go by what they judge to be an accurate picture based on the majority of information they collect -- if this majority happens to be tainted with FUD, it should not be surprising that their views show this too.
But after 5 months, if the reporter is worth his salt at all, he'd have dug deeper and perhaps discovered that Linux really isn't what the FUD depicts it as, and that there is a glimmer of truth to the claims made by Linux supporters.
I hate to say this, but Slashdot seems to be home to a lot of paranoid people who believes that popular media is a Big Satan that is totally clueless and always inaccurate. While it *might* be true that popular media is usually inaccurate, that doesn't justify the conclusion that *anything* from the media is not reliable. So please, people, before flaming this article to death, let's do some research and let's show some hard evidence of why this article is so lousy, as it's claimed to be. Pointing fingers at a reporter's reputation is not sufficient grounds to dismiss an article.
Just to be sure (Score:1)
new embedded apps, the focus of the Linux development ffort is still
the server: am I right?
EmbeddedLinux.com (Score:1)
I know the guy who owns the domain embeddedlinux.com
He's not sure what to do with it (just a link page so far). Any suggestions?
BSD vs GPL for embedded systems (Score:4)
It seems that economic model for open source that makes a lot of sense in the embedded market is for the chipmakers, who are not software companies particularly, to port an open OS and then give it to their OEMs to make a more attractive packaged offering for their chips. But in that case, wouldn't some embedded systems makers who are in hotly competitive areas prefer to have an OS without the "gotta hand over the source to our modifications" feature? I.e, isn't there room for the BSD license here? Linux and the *BSDs are essentially the same from the chipmakers standpoint. Hire a small staff, do/maintain a port, give it away with the chips... what do their customers want?
I prefer the GPL myself, but it does have to compete against the BSD license and I want to explore/understand the implications, since the embedded systems area is going to be so big.
Open Source has been there for a while (Score:2)
Too much OS? (Score:4)
Is it wrong to wonder if, perhaps, we might be better off having different OSes that serve different purposes? Do we have to have only one, jack-of-all-trades OS?
Anyway, there is a reason that PalmOS beat WinCE, and it wasn't necessarily the normal Microsoft bugginess. It was the fact that PalmOS just does what is needed, and doesn't have a bunch of extra fluff brought down from bigger machines. Would an embedded-linux be any different?
Re:What does "Embedded" mean these days? (Score:1)
I'm late in starting a new project which will use the ARM Thumb-based AT91F40416, a 32-bit chip with 2MB of embedded flash memory. One of the sticking points is the expense of specialized development systems - I've had quotes between $4K and $10K, which are over my limit at this time.
I've looked at several open-source options, like RTLinux, eCOS, ucLinux and so forth, but user-friendliness (in the strict sense) isn't a feature... most of them seem to assume that you're developing for the x86 architecture, or necessarily want to use a Linux or BSD machine as a development system (which I definitely don't).
Look at the Palm development system based on CodeWarrior: you get a standard IDE, debugger, and a Palm simulator box to run things in. The linker grabs the PalmOS as a library, or if one had the source to that, one could build stuff directly into the PalmOS. This is the ideal way to write an embedded system, IMHO. Until somebody develops a similar way for me to write my own stuff to, say RTLinux' API, but will all the facilites I'm used to, and then just compile/link the kernel with that into a ROM image, I won't be a convert to embedded Linux. Sorry...
Re:HDTV's and such (Score:1)
"Embedded" means... (Score:1)
This problem also extends to calculators. They have a cpu and minimal UI. Are they a computer? Technically yes, but since they have custom input keys they are classified as a calculator.
Heck, my 10 year old HP48 has 256K of ROM for its "OS" with a built in filesystem including hidden directories !
A new term is probably the most un-ambigious way to go. Otherwise everytime someone mentions embedded, they won't know, just 'how minimal' is the system.
Following after the Real-Time OS's termonology, I propose the terms "hard-embedded", and "soft-embedded."
Meaning, the hard-embedded system has no GUI, and the soft-embedded system has some sort of interface.
Feel free to suggest better termonology.
Cheers
Yahoo news re Red Hat (Score:1)
Bias in Reporting (Score:2)
Re:BSD vs GPL for embedded systems (Score:1)
...
Do the developers simply have to observe a "chinese-wall" model for developing, keeping the OS in one set of directories and the "app" in the other, and linking/stitching them together only as the last step before ROMming?
Now, IANAL, but if I recall, the viral nature of the GPL does not extend outside the actual OS. Therefore, if an embedded systems developer makes a modification to the kernel, source must be made available. However, if other OS components are developed ground-up without GPL'd code, the rest of the system need not be placed under the GPL.
Also, the GPL doesn't require that source be provided with the product, but that it must be made available for no more than the cost of conveying the media. This means online for free, or via CD for cost of media and shipping.
If an embedded systems developer were smart, they would use an embedded-linux kernel and build closed source tools around it (if they wanted to obfuscate the source). OTH, if there were a hypothetical LinuxPilot, the developer could give away the source, but need not disclose proprietary chip design used to build the device. So the GPL is not that large a hurdle to leap to the manufacture of Embedded systems using Linux.
-- Never underestimate the power of very stupid people in large groups
Re:BSD vs GPL for embedded systems (Score:1)
So can anyone 'splain to me whether a proprietary device driver that is implemented as a loadable module would or would not fall under the GPL?
-- Ryan Waldron
Re:"Embedded" means... (Score:1)
Still kind of a grey area where certain devices are concerned. The criterium I used to use was if you could develop code natively on it, then it wasn't embedded. Now in theory with Pocket C you can write code on a Palm Pilot, compile it (interpret actually I think) on the Palm Pilot and run it on the Palm Pilot. To me this makes it unambiguously not embedded. But what about the Web Pad thingies? Where do they fit in? Can you run more than one application on it? If so, I'd say it's not embedded too, but if the only thing it will do is run Mozilla, I could make a case for classifying it as embedded. (Just with a more complex UI than most embedded stuff)
Calculators? You can write code on them (the programmable ones anyway) and you can run more than one app if there is sufficient storage space. So I dunno about that one.
Application and customer preference determine OS. (Score:2)
A project which has gone through several stages orignally used QNX as its base OS because of a real-time OS was required by the contract. In the second version, the client has demanded affordability and we have switched over to a Redhat5 release, and have ported over most of the code (not as fun as it sounds).
Even the front-end of a robotics project which is now going into a production model, is being moved from a quickly written VB app to a Linux based solution. This is being done for (surprise) reliability reasons.
Lastly we are using PC104's with a slimmed version of linux (fits on a floppy - pardon my ignorance as to what dist) as an application specific translator (goes from serial communications to IP).
Is Linux being used in all our embedded systems? No, but it is being used with greater frequency, as it continually is being recognized for its price and versatility.
Re:ZDNot (Score:1)
It's possible, but until I see an article about the (albeit typical) troubles he went through installing linux, I won't believe him. I need something verifiable before I change my opinion of somebody. Also keep in mind that Jesse and his ilk tend to bend with the breeze. That is to say they are sounding boards for the popular opinion at the time of publication.
But after 5 months, if the reporter is worth his salt at all, he'd have dug deeper and perhaps discovered that Linux really isn't what the FUD depicts it as
Got any articles indicating this alleged "digging" he's been doing? I'm reasonably certain that if he's a reporter worth his salt at all he would have posted a factual article indicating the results of his research. Thus far, I have found nothing to indicate this. Perhaps there is a URL I have overlooked, and if somebody would show it to me I may look on Jesse more favorably.
While it *might* be true that popular media is usually inaccurate, that doesn't justify the conclusion that *anything* from the media is not reliable.
True, but in a court of law a person's credibity is tested by asking questions of those who know him, as well as past actions. Shall we hold Jesse to a different standard just because he's a reporter and naive?
Linux is still missing a few pieces... (Score:2)
Re:ZDNot (Score:1)
Not really. So long as the person doesn't have a vested interest in the outcome (it doesn't affect him) you can ensure a certain level of impartiality. The legal system uses this to ensure the judge is impartial. Another thing that is impartial is the facts. If you write a report saying "Drive xyzzy is better than Drive ABC" and can back it up by saying the aureal density and maximum seek times of drive xyzzy are superior to abc, that's a fact - so long as you state it as such. It's opinions that bias something - not facts.
Re:BSD vs GPL for embedded systems (Score:2)
Yes, this is my understanding, but IANAL. If somebody out there knows otherwise, I'd sure appreciate it if you'd let me know. I develop s/w for traffic control equipment and plan to use some form of real-time Linux (eCos, RTLinux, eg.) for my next product. My PHB would be very upset to know that we had to release the sources to our apps. Clearly we are required to make available any changes we make to the kernel (or any other GPL'd code), but as I understand it, we don't have to release anything that we write that lives in user space.
Re:HDTV's and such (Score:1)
Can a WinCE box go BSOD?
According to a Microsoft leak ages back, the code is there; and whats more it happened to me, only once on a educational notebook & no I didn't get a photo, sorry
Question 2 - will we be able to upgrade a TV's code for free with linux or will we have to pay for the operation from our manufacturers web server?
Food for though (or eating infront of the telly)