Slashdot Mirror
Linux In Robots, Windows in Handhelds
savuporo writes "Robots.net is reporting that Linux-based robots are far more common than Windows-based robotics. Especially various Asian robot builders are increasingly selecting Linux and other open-source software as a basis for robot products and research. Linux is also gaining ground in other embedded applications like PDAs and mobile phones." That said, prostoalex writes "50% of all the PDAs sold in 2003 had Palm OS, while Windows family accounted for 37.7% of PDA market. In 2004 Microsoft is the leader of handheld OS market with 43% market share, followed by Palm OS with 36.3%."
Since years we've been reading the PDA is dying, and unlike all the "BSD is dying" crap this actually means the market is shrinking. As long as Windows isn't a big player in the mobile phone market, that's nothing to boast about. And their mobile phone products suck - they've even crashed. That is something mobile users aren't to accept, because other key players seem to have it worked out better.
Linux gets slowly but steadily adopted into more and more mobiles, same with carrier grade Linux with the telcos.
Add this to robotics, which is associated with the biggest increases in productivity, there seems to be a bright future for embedded Linux, which is really contending with stuff like vxWorks or Symbian, not so much Windows.
Linux and free / Open Source software are used more heavily than commercial software for research and development projects.
Who would have thunk it?
This is the same as saying "Users choose Windows, Hackers choose GNU". It's not something specific to the handheld or the robotics market. It's the same that happends in the Servers Vs. Desktops dept. In areas where there is a Hacker in charge, for example, sysadmins, developers, etc. a Unix like OS will most certainly be choosen, and GNU is in most cases the best choice, because of many reasons, including ethical and comercial ones.
It's not easy to reach the end user. Specially because it's expensive. Some companys spend more on publicity than in development, why?, because that's the way to reach the end-user market.
ALMAFUERTE
WTF am I doing replying to an AC at 5 A.M on a Friday night?
depending on design of the robot, it may have more than one running CPU (one system to deal with movement, another to AI) which could increase the licensing cost of each robot... proprietary OS' that charge per CPU would eat more of your budget...
There are tons more Microsoft Smartphones than Linux based ones, and there are a ton more in the queue. Microsoft has gone from 0% to the majority in the PDA/handheld world and will eventually do the same in the cell phones. You guys don't understand how powerful it is to own the most common desktop API's on the planet. Many developers already use the Windows toolset (Visual Studio) and Win32/.NET API's and thus can start developing for these devices quickly, in C/C++/C#/VB plus the Open Source .NET Compact Framework (yes, there is one maintained by the community). The same cannot be said about Linux/Qt (or whatever UI that particular device is using) and whatever API Symbian is pushing this year.
.NET is making large inroads in that market as well.
Unification is key. One single toolset is key. One common API is key. One common UI is key. Without it you won't be able to keep up with Microsoft.
As for the robotics market, yes it is a growing one. However the number of robotic devices produced is very very small. Microsoft CE
Microsoft 2, Linux 0
They don't have the market share YET. Microsoft didn't have the game console market share when they started. Microsoft didn't have the PDA/handheld device market share when they started. Now they do.
You guys don't get it! MS is looking ahead, not behind.
However in the embedded market, these things are either not the case or don't really matter. Please note that I exclude PDAs here.
So in the long term, Windows-devices will have a hard time because while royalties make up just a small amount at the beginning of the lifetime (paying the developers is more expensive), the longer the product (or the product-line) is sold, the less new developments are needed and the royalties become more and more important. Also market pressure usually forces the sales price down which also causes that the royalties make up a larger share compared to revenue.
Also, Linux offers a rich software library which is readily available and just needs to be recompiled.
So while some WinCE-solutions might have some small success, they are pretty much doomed in the long term because they just can't compete in a matureing market.
It took them 5 years from those early awful Windows CE based handhelds to beat PalmOS. I'm sure that the PDA market will probably end up 45/45/10 Windows/Palm/Other though, before disappearing into obscurity.
But their Smartphone product sucks, people hate using it. People like Symbian though. Symbian being something resurrected from the dying EPOC PDA operating system.
Thing is, the first PDAs did work, albeit slowly. People don't mind that. But they will remember that Microsoft based phones crash, and their friend's one didn't, and they'll get a non-Microsoft one next time around. Windows gained ground in the PDA arena because PalmOS stood still for so long. Symbian and PalmOS aren't standing still now.
Depends on what one chooses to mean by 'intelligent', doesn't it?
If Windows were free (or near so), then I can answer which one would be chosen more often. The same one chosen more often today (when it isn't). Windows.
Why?
Because it it ubiquitous. Because it is, for all its faults, easy for the non-geek to use. Because it has the most applications and tools that most people want and currently use. Because its what they use at work. Because it's the easiest for which to get software under-the-table.
Are these decision points (and I'm sure I've missed a few) signs of intelligence or defect?
So far, all bigger tries of Windows-based smartphones have either failed due to unreliability (both T-Mobile and Orange have discontinued their Windows-smartphones over a year ago IIRC) and there is no positive trend in sight.
Symbian is established but costs royalties.
Linux is free but isn't established.
Windows is neither established nor free.
Those who want to play it save go for Symbian. Those you are willing to take a risk go for Linux (which offers being free as a reward for that risk).
Why the hell should anybody take the risk of using Windows on a smartphone? It just doesn't make any sense.
Linux is much cheaper, you can develop a Linux based product with no need to pay for a commercial license for the source. Of course you need to make your modifications available if they link to the kernel.
Linux can be made to respond a heck of a lot quicker too, due to the ease at which you can strip out the bulk and compile for embedded systems (2.6 has such a kernel option). You stand more chance of getting Linux to a near real time state than you do with Windows.
The main reason that Pocket PCs sell as well as they do is simply that they natively support Office and Outlook. You don't get any goofy unsupported meeting reccurance patterns, and you can open attachments in MS office format. Most corporate (l)users run Windows and Office, and thus this is what they are familiar with.
Of course the attachment item is possible on a Palm, but it requires additional software. Again, out of the box functionality is king...
I think you hit it on the head when you say the five year old palm does everything you need in a PDA. But I think you're wrong when you say the solution is to deploy more advanced technolgy. Unless there is a killer app for this tecnology, it will likely only hasten the move away from stand alone PDAs. The original Palm experience was about stripping stuff to its essentials. Adding to this minimality without a killer app is only detracting from it. This is why the PocketPC user experience still lags the PalmOS experience of five years ago. Not to say it doesn't have some cool stuff in it, it's just awkward and irritating to use -- take this from somebody who as a developer has multiple PDAs.
If anything, the reason the PDA market is dying is that people don't need palmOS 5, much less palmOS 6. There is nothing compelling in the PDA form factor to drive new sales.
Convergence is not some brave new world where people will be watching movies on their cell phone, its really a contraction and subsumption of the old world into to the phone handset. People are rejecting having more capabilities stuffed into their PDAs, and voting with their feet by either going with plain old cell phones, or smart phones, or devices like the blackberry, which is frankly pretty rudimentary from a technology standpoint.
It's an emotional thing. The developers of PDA technology have lost touch with the user. There is only one company that understands this well enough that it could really revitalize the PDA market: Apple.
Post may contain irony: discontinue use if experiencing mood swings, nausea or elevated blood pressure.
People, when given the choice, prefer to pay nothing.
This is especially true in areas where "support" isn't an issue. For example, robotics is a very special application. Microsoft isn't going to be of much help when it comes to such an application... at least not in the general sense. The best they could offer is the base OS... and that's pretty simple -- if you're a technical guy and can't troubleshoot that little bit, then you probably don't need to be building robots in the first place... if you can, then why do you need Microsoft "support"? And since you don't need that, then what's the benefit of BUYING an OS when you can get one for free?
There's no escaping free in special applications.
Man, you Linux fanboys are uncouth. The Xbox went from 0% to 27% in ONE GENERATION of the console. XBox Live is raking in revenue. The XBox outsells NINTENDO in most markets.
Forget about the PDA market. Microsoft has already won that, they are focused on the Smartphone market.
This is a long term plan. So they lose money for a few years, they have the cash. It will pay off in the end. Their last quarter proved that.
Keep your blinders on. Sony and Nintendo are going to lose.
Because it has the most applications and tools that most people want and currently use. Because its what they use at work. Because it's the easiest for which to get software under-the-table.
And, unfortunately for Windows, once Linux gets a respectable market share these are all reasons why there could be a sudden flip from the vast majority of people using Windows to the vast majority using Linux. Once Linux has enough users that software companies and individuals start releasing Linux versions of their software by default and people realise that Microsoft is not the only choice there could be a snowball effect. Once that happens Microsoft would have to work incredibly hard to regain anything more than minor market share.
Of course this might not happen but I believe Microsoft is more at risk from Linux than mere market share would indicate,
Jeremy
Linux gets slowly but steadily adopted into more and more mobiles...
That's quite an assumption to how things will play out. I'm not so certain the first statement leads to the second.
While I understand that some companies (Nokia, due to its ownership stake in Symbian, being the most significant) have a vested interest in Microsoft not being the OS of choice in a phone or smart phone, I wasn't aware that the consumer had much choice in what ends up in the phone. My understanding is that the relationship between the software supplier and the phone maker (and the phone maker and the carrier) is more significant than what the user is interested in. The challenge is that the consumer criteria for purchasing a phone are the brand name of the phone, the design (straight vs. clam shell), the camera (or lack thereof), cost, ringtones, SMS capability, games, and other features; the OS is mostly (if not completely) transparent to those decision criteria [remember Marketing 102: people buy solutions to problems, needs & wants; they do not buy products]. If I got a new phone, I would ask what OS the phone is running; however, I bet most people don't care. As a side note, I don't actually know if Microsoft-based phones display a MS logo on boot; however, you should consider that people might associate failure (e.g. crashing) to the brand name of the phone as much as the OS it is running.
There may be long-term damage if the systems do not work properly, but it will take a long time to play out (The replacement time for phones is 18+ months in the US last I checked). This (along with the lack of major press on the issue) is probably enough of a reprieve that Microsoft can fix its problems. This is a much better place (from their point of view) for Microsoft to get itself entrenched - because it only needs to maintain the corporate relationships with the manufacturers (and to a lesser degree the carriers)... Then, with "good enough" products, they can survive.
The same goes for Microsoft's push into IPTV and its deals with SBC and others. There isn't a need for a consumer to make a choice - if you subscribe, you're using Microsoft's products; your only non-Microsoft choice is to not receive the service. While some staunch anti-Microsoft individuals may be willing to take that step, many others (I would argue most people) would just as well have the service, even if it means dealing with a Microsoft product. If Microsoft wins any cable companies, some consumers may have no choice at all if they want to have on-demand services.
It is, in truth, a brilliant play by Microsoft into areas where it is harder to make a consumer choice to remove a specific type of software. I highly doubt we will see the day where the software has to be independent of the phone or set top box, as was the case with mainframe computers when IBM got itself into anti-trust problems. So Microsoft is here to stay, even if they have to share the desktop.
Palm doesn't lose to PPC on app availability, quite to the contrary, Palm still wins out there.
PPC wins out in media compatibility and multitasking.
PalmOne/PalmSource have been royally fucking the goat on making progress with their OS and their hardware over the last few years. The Treo 600 was good, and the Treo 650 is basically what the Treo 600 was supposed to be (a decent screen and working camera, and it's a fucking 600 dollar upgrade, AND they went ahead and made the software buggier and their hardware quality control worse in the process).
Why haven't they moved to Palm OS 6? Why isn't there real multitasking? I want to be able to switch back and forth between my email and phone app and web browser without losing my place and without UI delays. Why does this device only have 32 megs of RAM onboard? Why the atrocious GUI delays that didn't exist in the Treo 600 (like switching to the phone app - takes about 1-2 seconds, crazy)?
I still like the overall Palm offering better than the PocketPC offering, and there are no PocketPC smartphones that offer the same form factor and usability as the Treo 650, but I'm sick of watching the idiocy that goes on with the Palm platform.
Personal robotics today doesnt exist outside of the hobbist techie and academic realms. But M*ft wont ignore an opportunity once the turnover is worth more then a few hundred million dollars and growing. As they did with the Internet, Java, handhelds, games, phones, settop-boxes etc. they may eventually take over the personal robotics market too. :-(
LS http://robosavvy.com/
Yes, and the market for phones without PDAs is going away. The two are merging into a single device, typically called a smartphone. These days you can get a pretty nice one for just a couple hundred bucks even as a prepay phone, I think I've seen this Motorola V300 for that. J2ME/MIDP 1.0, and it folds up so you don't beat the shit out of it like a Nokia or something :) Some of the Nokias have a killer feature set, though. Anyway the point is that the phone market is going away too. But, PDAs are not vanishing, they're just merging with phones. Sooner or later there will ONLY be PDA phones (you can often get touch screen organizers for five bucks at wal-mart now, they're not much of a PDA but they're getting there) and this whole conversation will be revealed as entirely silly and unnecessary.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Um...the car you drive was created by a robot. In fact, many consumer items (used by adults) are made by robots.
How many handhelds are used by adults? Millions.
If you are just counting handhelds, (not PDAs) than children win this category (Gameboys have outsold every other handheld made by a large margin.)
Any questions?
Sure. What was the point of your post?
Open Source Sushi
Don't worry about crashes; worry about script kiddies (because these robotic butlers are going to be plugged into the Internet for convenience's sake). If a script kiddie 0wns your computer, you are FUBARed figuratively; if he 0wns your android butler, you are FUBARed literally.
I would highly recommend keeping Windows away from any mobile automaton, especially from those with sharp blades (lawnmower). Far away.
Forget magic. Any technology distinguishable from divine power is insufficiently advanced.