The future is mobile. That much we know for sure. But it seems that the operating system world in this market is being rapidly taken over by –again– Microsoft. The new smart phones are are using WinCE, Symbian or Palm. Linux has barely 1% of this new, smartphone market.If Linux visionaries had seen the writing on the wall in early 2000 (e.g. MontaVista, Red Hat’s embedded division and even Familiar) and contribute towards a united, powerful, usable and most importantly, properly marketed Linux phone/PDA edition, things might have been different now.
The desktop operating system door was already closed with the release of Windows 95 in 1995. Alternative OSes can work hard to capture 4-5% of the global market share but they will never manage to overcome the legacy of Windows and its monopoly — for well-known reasons that have been discussed everywhere and by everyone.
But the phone OS door was open. It was wide open and it was waiting for its next big predator. And Microsoft jumped into the opportunity taking over PalmOS year after year and Symbian is next.
Where was Linux all this time? Is the fact that Linux is driven by hobbyists had a role to play to this strict embedded market that has its own rules and needs? Or is it the moderately high requirements of Linux (at least 208 Mhz, 32 MB RAM) put a dead stop to integrators while PalmOS/Symbian/WinCE can run on 104 Mhz (or less) in full graphics mode with minimum memory requirements at 8-16 MB?
Here are a few new features coming next month with Windows Mobile 2005 according to MSMobiles:
* Powered by Windows CE 5.x not by Windows CE 4.x as the second edition
* .NET Compact Framework 2.0 pre-installed
* No distinction between smartphone and Pocket PC – Windows Mobile will run both Pocket PC 2003 and Smartphone 2003 programs
* direct support for Qwerty keyboards, so that new products from HTC, Motorola, HP and Samsung will be better usable just through the Qwerty keyboard
* support for new push functionalities of Exchange, but not only for e-mails but also for contacts and calendar events
* Wi-Fi support for smartphones (devices without touch screen)
* “Storage” – a place in memory that doesn’t dissapear even without battery – now will be available also for Pocket PC
* improved Pocket office applications like Excel and Word; Pocket PowerPoint added
* improved Windows Media Player 10
Symbian 9.0 comes with a slew of new features too, and PalmOS has always version 6.1 under its sleeve, just in case. Where is Linux (or QNX for that matter) right now? Why has no company stepped up to really help Familiar optimize/develop mobile Linux further and make it a better contender? Linux is more powerful than any of these other phone OSes, what it needs is polishing and real backing behind it with a serious-enough company to drive the way ahead by creating a powerful mobile distro.
What I divine from the comments here is that the cellphone OS market is still not profitable for Microsoft, and it might not be profitable for Palm. Symbios is the leader.
A Linux customizer would have to compete with Symbios, and be willing to plow money into the project just like MS is doing. Though the existing codebase gives Linux and edge, WinCE and Palm already have their own existing codebases, so the edge is narrowed.
Exposing the code is not an issue with the LGPL.
Linux is very far from being ready for the cellphone, because it’s weak in providing the necessary foundation of multimedia. (Again, this is derived from reading this thread.) X is not right – something like SDL or SVGALib is better. That means writing your own font rendering, and GUI. Video and music playback may or may not be pre-integrated. You need to write your own drivers for the camera, sound, keypad, etc.
Even the file system might not be right for a celphone. The data structures the users experience are more like an OO db, or a flat file DB than a file system. So you’d need to integrated a tiny db, probably proprietary.
Last, I think that the interest in the celphone OS market is due to the fact that you can more easily get people to spend small amounts of money via celphone. The increased revenues in the market will enable OS vendors to charge more for their product, if it works. Again, there might be a lot of investment in the market that’s precluding Linux from entering.
Linux will be the number one OS in least than 1 year on Palm Device. Curently, Palm still own a big 59% marketshare. (I however agree than in fact there are less than 59% real Palm owner, mainly because Sony is out of the market, but, unlike computer, they’re really no need to update this kind of device, a Palm Vx is still fine for most people, and few Tungsten user will ever upgrade their hardware.)
The big marketshare of Palm will propably not drop because of their adoption of Linux. They just have to keep their simplicity (It’s much more faster do to anything on a Plam than both PPC or Gnu/Linux, especially anything related to personal notes or Calendar). Even with fewer people upgrading their Palm, all new Palm based on Gnu/Linux will progressevily take up to about 60% of the market. But to do so, they must use Palm OS GUI interface: Qtopia or Agenda are not quite to the point yes.
I do not consider the new version of PAlmOS very Linux-y. Yes, it will use the Linux kernel, but when I am talking about Linux on PDAs, I am talking about the whole package, an *original new* solution. PalmOS under Linux does not offer that. It’s still PalmOS running the same PalmOS apps as it used to before. What I am talking about here is a solution like GPE, Opie or Qtopia. Something original, something that a company would invest to develop new stuff based on Linux kernel.
Eugenia,
(who had dinner with a palmsource engineer just last night).
Eugenia,
(who had dinner with a palmsource engineer just last night).
Apart from bragging purposes, this holds no real purpose/value whatsoever without some sort of additional info/background on why you mentioned it and how it relates to this story. For all we know you talked about the weather.
You’ve also not given one single reason why, in your opinion, Linux (in whatever shape/form) would be the best/most appropriate OS to run on these devices, just that ‘Linux’ (as a whole?!) missed an opportunity. I understand it’s an editorial, but some well-founded/researched arguments would definitely help.
My point is that menuetOS might be ported to the other platform being an open source project,maybe it’s not possible,but linux started out being for X86 too
No, porting menuetOS to another processor architecture is literally creating another OS from scratch.
It will not happen unless another team comes out of nowhere to develop for another architecture, or the current team completely stops working on x86 to BEGIN work on a different platform
Menuet OS was programmed in x86 assembly. This means that the source code is only good for x86 assembly machines, and can not be ported.
Assembly Language is 1 step up from binary opcodes which your CPU actually uses to process data. There are no cross-assemblers, only cross-compilers for languages such as C or Java.
Advanced mobile phones are relativly new. It’s too early to tell. 10 years from now we will probably buy a generic cell phone (preinstalled with a OS), but with the option to install whatever OS you want just like PCs?
The reason behind this is simple. When the hardware becomes more advanced, the phone producers will buy more and more commodity hardware because the manufacturing will be too expensive to do alone.
New competitors will also come into the market and their only chance of surviving will be to give the users real choice.
This is likely the most clueless article I’ve read here, and the comments match.
In year 2000, Linux was a lot more arch-specific, uClibc and busybox did not exist, and powermanagment was abysmal. Just look at how new thing dyn-tick is.
Yes, you can squeeze Linux into 2MB of flash and 8MB of ram, but you couldn’t do that in 2000. Also you will lose all the advantages (standard libraries and tools) if you do that.
With multimedia features coming standardplace in mobiles (sorry american luddites, it is reality _everywhere_ else in the planet – even in africa), the amount of CPU power and memory will sustain running Linux using OSS libs and a X gui – With apps less than 30MB.
At that point propiertary RTOS’s are too expensive, and MS can’t run Lonhorn on the systems, while winCE is too restricted in features and ties you to MS and to compete with taiwanese vendors with same software selling the same hardware with a thinner margin…
Where would you go, if you where a hardware vendor? when ALL your part’s manufacturers provide Linux reference drivers? The last part is already mostly true.
Read the mobile group presentations from here to get more upto track:
http://tree.celinuxforum.org/pubwiki/moin.cgi/TechConference2005Doc…
…Looks like most vendors are already keenly looking at that way. Ofcourse legal departments are scared of GPL, but they hate all other software/IP licensing as well.
Correct. Going with MS and Symbian is what happened until today because putting beefier hardware on with decent power usage made the phone too expensive. But phones have proven not to be a static target(otherwise they’d all be using cheap microcontrollers) so as better hardware gets cheaper and smaller it’ll end up in the phone.
And when that happens the “Linux is too big” complaint will disappear. Instead, the current leaders will start falling behind – their structure will constrain further development. Linux (under the guise of whatever vendor offers the best solution) will be ahead of the game. It didn’t “miss” anything – its best days are ahead of it.
Linux runs on Motorola E680, E680i, A780, A768, etc…
Just take a look at:
http://linuxdevices.com
I agree with the author. Palm is a perfect example of a company that was there first, but not innovative enough to stay on top.
I dislike Microsoft as much, if not more, than others but one thing I always hand it to them is that they refuse to just go away and die. I just wished they stayed on top with decent products and not by keeping the FUD and marketing machine going.
My wife loves Palm pilots and I have found it interesting watching them the past few years. Before, say, two years ago their lineup of Pilots was pathetic. Lack of good color displays, low memory, little software, etc… Then enters Microsoft with the Pocket PC and all the bells and whistles. Thankfully it was as flaky as Windows because it kept the Palm faithfuls from switching. But new users, used to Windows, went no problem. Then Palm comes out with a really nice line of PDA’s…only because they had to, but then it was too late. They have lost tons of market share simply because they got complacient.
The author of this article is saying that Linux had a chance to enter a very new market and dominate. I believe Linux is the best thing out there. If it entered a fresh market and dug in its trenches it would be there to stay.
Those who do not think small multimedia devices are the wave of the future. Go look at the Palm (even Pocket PC’s) isle and count how many kinds there are. Then go count the MP3 and Phones and then tell me what consumers are buying.
BTW… I use only Mac and Linux and Palm devices…no trolling intended here .