Six of the world’s largest telecommunications companies have officially launched a non-profit organization dedicated to creating a common mobile Linux software platform. Founded by Motorola, NEC, NTT DoCoMo, Panasonic Mobile Communications, Samsung Electronics, and Vodafone, the LiMo Foundation is inviting membership and participation from application and middleware developers.
Everybody knows Eugenia likes ultra-mobile computing and despises the lack of standardization that hinders its progress.
TFA makes LiMo look like an acceptably open project with an impressive list of giants behind it. But there’s always something wrong with consortia of giants. And whatever it is, Eugenia will find out and give us the scoop. Until then, I’ll think happy thoughts.
The problem is that this is about the 985345987th consortium that’s going to standardise Linux on mobiles.
“the 985345987th consortium”
Please , there is not even that many GNU/Linux distribution , I also like to point out that before that Alliance , partnership and projects where being discussed witch don’t come with a 800 000$ entry fee.
No, Eugenia is not happy (I just asked her
As Thom said above, there are too many consortiums and APIs and crap. Instead of ALL of them coming together to create one super-distro that will blow everyone else away, they re-do the same desktop-distro mistakes, but in an even worse scale, because Linux phones from different consortiums won’t be binary compatible at all.
Linux has time and time again proven to “suck” on phones. I’m with you on the lack of standardization, and I’ll take it even further a step.
I’ve not seen a single speedily-operating well-designed UI on a linux-based phone yet. They are either sluggish, ugly/unusual, or both. Actually, always both on the phones I’ve used.
Phones aren’t good devices to toy with cramming this or that *insert random OS* on. They are meant to do one (or a couple, for PDA phones) functions, and do them well. Linux is great and all, but that’s not what Linux’s strong point is. Linux is the opposite, it does everything “ok”, but it works on just about anything. Phones need very specific implementations to operate well.
Even WM (from my old experiences) and now my new experiences with S60 haven’t been too impressive. They are slow, have annoying bugs, and in general irritate me. However, at least the UIs are *usable*. I won’t go so far as to say nice. :p
At least they are going to standarise the GUI since all are using GTK+.
No, this is not standardization. The fact that they are all using GTK as their basis does NOT mean that you can easily port your graphical application from one phone to another. GTK is just the basis for their own custom stuff and libs, not the end point where every dev can just compile once and run everywhere.
But they are trying to standardize APIs for this particular platform. So in theory phones from any of the member companies could all run the same apps.
edit: Also, may I ask: What other consortiums are there aside from this one, that have this level of backing? All I can think of is Qtopia, Maemo and Access, but each of those is pretty much limited to one company (Trolltech, Nokia, and Palm(?), respectively).
edit2: Okay, I read further down in the article about the Linux Foundation’s Mobile Linux Initiative and the Linux Phone Standards Forum. Still seems to me like the one with the most handset makers backing it wins–i.e. this new one.
Edited 2007-01-27 13:29
false some use qtopia, who have a better look than GTK+
Collinm, only 1 uses qtopia (Trolltech’s own implementation). The rest use Qt Embedded, and this is not the same as Qtopia. Not only that, but they use their completely own look, so you can’t say that any of the GTK or Qt implementations will look like GTK or Qt. They won’t. They might “feel” as such because of the way they resize or the way they refresh the screen, but they won’t look as such. Example: Motorola’s EZX phones which use Qt Embedded as their base, but they look nothing like Qt.
D-Link V-Click and some samsung phone use qtopia
As I said, they are using Trolltech’s implementation. But most others don’t. The rest use only the Qt Embedded part.
normal, qtopia is not so old
Read the pdf, they are all using GTK+ for this.
“there are too many consortiums and APIs and crap.”
Its normal its a new market , But you need to make and understand the difference between projects , Initiatives , discussion , proposals , consortium alliance and partnership VS Foundations.
Foundation usually create or protect something and its members actually fund it in order to accomplish there goals. With an entry Fee of 800 000 , 400 000 , 40 000. I seriously doubt that it is going to be making the same impact as all those other projects and discussion you have in mind.
You have this obsession with one solution for all , where in fact there is no such thing in any market. If everyone wanted all the same there would be only one solution for cellphone already.
Now on top of that your making a comparison between two different market with two different need and two different situation , sorry but when GNU/Linux desktop have has just one OEM brand name supporting them fully and making real offers then there will be a possibility to compare.
So far I can only disagree with you that this is a bad thing.
I think it’s a good thing. Not every conortium, group or whatever will eventually succeed. However, if you take a closer look at WM2002/2003/5 systems, you see that there is a single platform, with different hardware and if something’s not really compatible, well, it’s exactly that kind of stuff.
We all could argue about a standard and in the long run, there may be one; at least it’s being tried!
I hope they can do something better than the horrendous Windows Mobile 5 Persistante Storage.
I’ve got a new O2 XDA Exec, and Contacts/Messages are just plain slow to update. I’ve used to have TMobile MDA (PocketPC 2002) and it was really fast on it.
Just some lessons to learn….
Here is the “Why Persistance Storage is a Good Thing”
http://blogs.msdn.com/windowsmobile/archive/2005/07/14/438991.aspx
I don’t agree. WM5 is pretty fast. Faster than my Linux Motorola phones for example.
Yeah, this means there is gonna be more virus on mobile! With one common platform, flaws found in common parts of the OS will affect all phone built with the Linux firmware. Guess what will hapen next…
Edited 2007-01-27 02:24
“””
Yeah, this means there is gonna be more virus on mobile!
“””
Let’s not start believing our own propaganda.
A monoculture can be bad if the platform is built upon an inadequate security model.
There are lots of benefits to standardization. And lots of benefits to a good security model.
Combining the two is the Holy Grail.
That implies a few restrictions on the design. You can’t base the whole thing on a programming language that is an open invitation to buffer overflows, for example.
But there is no reason that standardization has to imply viruses.
Not that this development *can* really be called standardization.
Linux performance sucks on cellphones both with respect to battery life and with respect to responsiveness. Both of these are difficult to fix design flaws stemming from Linux originating as a PC operating system.
The mainstream Linux community is interested in a power management ala laptop and that model doesn’t work well with ARM devices because it doesn’t cope well with all of the power complexity. An appropriate model for embedded devices, from Montavista and IBM was described at the power management summit, but has been rejected in practice.
Meanwhile, ARM9 is a horrible architecture for embedded Linux because of the nature of the caches and TLB. ARM11 and Moore’s law will eventually overcome this, but it won’t be for 3 to 5 more years.
A herd of consortia, all doing the same thing in slightly different ways, none concentrating on these issues, will only make Linux acceptance take longer.
And that’s before you get to the actively bad design of the UIs on offer.
while battery life and responsiveness may suck with linux — I don’t know, cannot test this; have you ever looked at WM2005, WM5?
That platform is specifically written for such hardware and if you want to know what battery drain is and lack of responsiveness is, you should try that first…
…is Nokia. And they could really contribute, since they’ve already got a working GTK+ based mobile platform (Maemo).
Of course, them joining would basically be like them admitting that they’re giving up on Symbian. Which they may eventually do, even though right now they’d never admit to it…. Time will tell.
Anyway, assuming Nokia doesn’t magically decide to join the coalition, I doubt that Maemo will be the basis for this thing. I expect the member companies will want to have full control over the development, and are thinking, “Hey, if Nokia could develop a new GTK+ platform in just a couple of years, so can we”.
Edit: Also worth noting is that this announcement is a major blow to Trolltech and their Qtopia platform. It’s really a shame, since it looks like by far the most polished thing out there at the moment.
Edited 2007-01-27 13:14
I’ve never been particularly interested in cell phones, and so don’t pay a lot of attention to this kind of news.
However, from what I do know (and I hate to say it), it looks to me that at least in this important market, Linux has *already* “Fragmented Like Unix Did(tm)”.
I thought GPL was supposed to make this kind of silliness impossible, or something?
It makes all kinds of sense for these companies to standardize on one API. It makes absolutely no sense *not* to. Why don’t they just do it instead of playing the Keystone Cops like this?
It makes all kinds of sense for these companies to standardize on one API. It makes absolutely no sense *not* to. Why don’t they just do it instead of playing the Keystone Cops like this?
Because, as of yet, no one really knows what that one API should look like and the founders of the various consortia are all competitors hoping their version will win.
Welcome to “OSF versus Sun/AT&T, the cellphone years”.
“””
Welcome to “OSF versus Sun/AT&T, the cellphone years”.
“””
I was afraid you’d say something like that. Damn you! 😉
-Steve
Alas, we are failing to learn from history.
I was deep into the OSF/AT&T debacle, as I sat on POSIX and was was research liason to OSF for seven years.
The last three years or so I’ve been deep into mobile devices and watching the Linux community fall into the same trap has made me sad.
Perhaps we have a slight different situation now?
At that time, Unix was synonymous of $$ per se.
I think they all realized that the OS and the basic software infrastructure is just too expensive and don’t increase their profit at all, it is just something that needs to be there. They could have thought “lets have a party, share the cost of the foundation and compete on topping and service” (i.e. hardware quality, functionalities, etc). After all this is what is generating money now and can even generate a surge on services market for mobiles (what will happen soon or later).
Also, why should them give this market on a gold tray to MS and become as chained as most of the PC industry is? I don’t think they would, they want to be at command.
Makes a lot of sense to me.
Perhaps we have a slight different situation now?
Yes. It’s worse. The fragmented Unix camp might have been able to fend off the onslaught of the Microsoft upstart had the players figured out the lessons sooner and played together rather than fight for market share. We were, in total, after all, the OS of choice when the mess started.
Now, Linux is a dark horse against the powerhouses, and the fragmentation is keeping it from even getting off the ground in any significant way.
As a software developer hoping to write applications for Linux-based phones, I certainly hope the light at the end of the tunnel comes sooner rather than later. I don’t even know which platform / environment to develop for.
It irritates me that these various consortiums feel like they’re trying to browbeat one another into submission so that only their standard remains. I feel like a movie studio caught between releasing on BluRay or HD-DVD. It’s either I conserve what resource I have and choose one, or spend more to release on all. And there will always be the fear that should the standard I choose get dropped or changed dramatically, that would be more work porting to whatever new system.
That said, I’m glad that the FIC Neo1973 based on OpenMoko has come around. They may not have the best hardware nor is their implementation based on any of the standards raised by any of the consortiums, but there is that … feeling of freedom and innovation and creativity about it. It makes developing apps feel like fun again.
Edited 2007-01-27 19:08