Slashdot Mirror
Launch of OSS For Mobile Phones
Linux Mobile Phone Guy writes "Members of the open source GPE project (GPE Palmtop Environment) today announced a new offspring project to create a fully open source software stack for mobile phones, GPE Phone Edition.
GPE Phone Edition is a fully open source project based on developments from
the GPE project adding necessary components for mobile phone usage. Based on
standards defined by the LiPS Forum a complete application software stack is built. The current implementation is based on code contributed to the
LiPS Forum by Orange/France Telecom's research and development lab located
in Beijing China in collaboration with GPE project members.
The result is now an open sourced software stack which can handle a GSM
compliant mobile modem for making voice calls, handling the SIM address
book and sending and receiving SMS. Also some additional application
exists e.g. for media playback, instant messaging and email.
They have some screenshots there and even a downloadable VMware image
using which you can try the whole thing in a virtual phone on your PC — if you connect a GSM Modem (like an existing phone) to /dev/GSM-Modem
you should probably even be able to use the full phone functionality
(access SIM card, send/receive SMS, make a call!)."
I wonder if this is just going to be some "tech experiment" or maybe something that people will be actually able to use. In real life, I mean
Pumbaa! I don't wonder; I know.
Is there a list yet
So where is the DRM in this? All mobile phone stacks are rushing down the "Trusted Computing" (or ARM's TrustZone) route with locked down software and DRM built on top of that. So where in this stack is the rights management stuff (or the placeholder for it)... since all of this software will be digitally signed and locked so it cannot be modified or changed by users.
It uses GStreamer, and since Fluendo (a company that touts itself as a Free software company) is committed to bringing the wonderful ness of DRM and "trusted" packages to Gstreamer, GNOME and Linux in general. I assume it will be based on Sun's DReaM and Gstreamer DRM.
How does this compare?
_ Page
http://www.opencellphone.org/index.php?title=Main
Also, what about the network? Seems to me that as long as the carriers don't open their network up it's going to be difficult. Adding a SIM chip might let one make calls, but it hardly strikes me as something where it's easy to push out new innnovative services. What does it matter if the software and hardware is completely open if the network isn't? Even "wired" broadband comes with enough pain as it is. It's as if the "Internet" is becoming more of an abstract idealistic medium with its own philosophy than an actual existing entity.
Regardless, here's hoping that the mobile world eventually becomes more open. What is the state of the art in "open, ubiquitous wireless broadband"?
Greenphone from Trolltech already has a development device available, although it is a bit pricey for common usage ($695 USD). Does anyone have any information about the differences in the project?
I'm looking forward to the day that I can buy a phone and install an open source OS on it, customize everything to be just the way I want it, etc. This projects brings that day one step closer. Congratulations, and good luck!
Please correct me if I got my facts wrong.
Coming soon: software to turn a bunch of mobile phones into a Beowulf cluster.
Please correct me if I got my facts wrong.
Being able to run a virtual phone on a PC is what entices me. It means anyone can sit at home and play around, write software, etc, even without owning a phone. While this is all "for fun", by the time these phones become cheap and accessible there will be a mass of kids who already have the expertise to use and program these devices.
I'll definately check this out on my PC.
... Didn't we already have a fully open source development stack for cell phones in QPE? I might be reading it wrong, but the summary makes it sound like they think it's something new?
A clunky suite of software to emulate a phone, fax, and answering machine through a voice modem came with my first Pentium computer in 1995. It used the mic and speakers, and was surprisingly good for what it was. I think it stored messages in the cranky old .voc format, and of course each fax received was a huge TIFF.
Slashdot Burying Stories About Slashdot Media Owned
This is a good step because now we (the global linux community) know how it (a GPS device) works in software,
before the user>UI>Kernel team couldnt fully interface with a phone as it was all closed source (so none of the many eyes/ideas philosophy which is what counts) all that was hidden away from a lot of people being non FOSS, plus moretime the cell developers dont get connected to the user in the same way as we do (the user/ui/linux community)
now with a standard protocol all open and visible in place, we will soon have full messaging/TCP anywhere in the world there is a cell signal, and these days thats pretty much anywhere, combine this with GPS and we are well on our way to having a intuitive reliable global communication device, web/internet/voice/video its all the same thing.
A.S
Which Windows mobile phones are the easiest to install this stack, which have the most of their features supported by the OS for apps? That can then run (recompiled) existing Linux apps? Mapping a touchscreen, joystick or keypad combo to the mouse, full Bluetooth and other radio control for voice/SMS/MMS/data...
Not just because I want my phone to run the same apps on the same shared data as my desktop. But because the limited phone UI will force new paradigms in using these little mobile devices which will run on my desktop. Linux has been a reworking of Apple/Windows too long. Jumping ahead on the desktop, with an eye on Apple's parallel iPhone evolution, will not only speed development of simpler, more powerful "phones", but also rejuice the networked desktop that's stuck in the 1990s.
Which current, cheap phones are the best targets for trying this new environment?
--
make install -not war
He refuses to own a cell phone until a completely Free (you know the drill) one is available
The LiMo Foundation
OpenMoko
The GreenPhone
GNU radio should be able to be the modem, in software, given merely a transmitter/receiver/DAC/ADC setup.
:-) Maybe that too can be faked in software.
How do you get a SIM card into a PC?
Now, why this is all labled under the term "Tuxphone" is beyond me. It's misleading at best. But it's probably still the most open effort around, as there simply is NO completely open solution here. There are lots and lots of companies which claim to offer an Open Source cellphone, but they all lie. When it gets right down to it, there's at least one part which is closed off and locked up. This includes Trolltech's Greenphone.
Note there are better replacement libraries for the Tuxphone, which are more robust, generalized and secure. Here's one: libgsmc.sourceforge.net .
Regarding the Carriers, this entire project doesn't appear to deal with the protocol that goes out over the airwaves. That's still locked up. What this DOES deal with is standardizing the interface to the chips which DO handle the actual airwave protocol.
GSM chips offer an interface which is just like a modem. ATDT..., but taken well beyond what Hayes originally intended. Yet it still works. So this effort doesn't seem to be dealing with the airwave protocols at all.
Or, in short, no, you won't be able to hack the cellphone network.
Regarding DRM, that remains to be seen. It's unlikely that any DRM will be put in for GSM. GSM is a well-defined international standard that anyone can use. But for non-GSM networks, forget it. There's not a chance in the world that the Carriers will open up their non-GSM networks. They like it locked down, and strongly so. Otherwise, they can reem you for all the rediculous charges on your cellphone.
Or, in otherwords, it's not your phone. It's theirs.
If anybody can actually find this mythical vmware image can you post a link please.
The situation with phone DRM is terrible. Phone that we buy actually belongs to the operator. It betrades our interests for the operator profit. In addition to your example of applications copy protection, I can give few others.
l eItem&ida=154 should be possible on all phones. But not occasionally, it's impossible to read Cell ID in java. Another example can be the strange (but very useful for operators) difference in WAP and TCPIP price. Here in Israel, there is a GPRS plan which includes unlimited WAP and only 5MB of TCPIP traffic. It's because of the closed platform nobody can tunnel all traffic through the WAP. Imagine if ISP would set different prices for the web and for the email access.
For example, there is no justification for the price of an SMS. Data wise it's equal to less than a second of the conversation, yet it costs unproportionally more. Another one: the phone always tracks its Cell ID. But it's impossible to develop an application which will access this data for the benefit of the owner, for example reminds you something when you return home. This http://www.psiloc.com/index.html?action=ShowArtic
I wish the best luck to any effort to stop this nonsense. Cellular operators should not make the most of their profit from giving us back what was possible from the beginning.
I'm not really sure what you are trying to say here but I can tell you a little about GPE. You demand:
and shall receive. GPE already runs recompiled applications with good mapping of input from all available buttons and the touch screen. Xstroke is on of the best graffiti handwriting recognitions I've ever used. The average distribution, like OpenZaurus, has a repository with many useful programs that work well. All the buttons work when it's useful and so on.
I want my phone to run the same apps on the same shared data as my desktop
That one is a little tricky still, but the applications all recognize normal formats like vcf. Projects like kandy are supposed to deal with that and will.
the limited phone UI will force new paradigms in using these little mobile devices which will run on my desktop. Linux has been a reworking of Apple/Windows too long. Jumping ahead on the desktop, with an eye on Apple's parallel iPhone evolution, will not only speed development of simpler, more powerful "phones", but also rejuice the networked desktop that's stuck in the 1990s.
Apparently you are not very familiar with GNU/Linux desktops, but that's beside the point. GPE does contain a collection of applications designed for PDA display constraints, but you should not confuse application and desktop design issues. X has proved itself admirably in GPE. Don't like your PDA's constraints? X forward the display to your laptop through secure shell to get both terminal and GUI interfaces complete with keyboard, mouse and screen real estate. On it's own GPE is somewhere between GNOME and Palm and is simple to understand and use. Work with Enlightenment as a Window manager promisses virtual desktops to help overcome small screen limitations and make multitasking easier. Right now, you can run Inkscape on GPE but you are better off using the simple GPE drawing tool.
GPE rocks because free software rocks.
Friends don't help friends install M$ junk.
This isn't like getting a PC onto the internet, where any software can be installed and anything that talks TCP/IP can connect. Like it or not (and I don't), many of the more recent phones, music players, etc. are not open platforms in that sense. And the service providers are more than a tad concerned about keeping control of what connects to their networks. (I don't know whether their worries are about a compromised device crashing the network, introducing some unreliability, or just bypassing one of their many ways to profit from every byte that gets transferred.)
Do we have any reason to think that this stack will be treated any differently?
Ceterum censeo subscriptionem esse delendam.
The Neo1973, will be released with its own software, OpenMoko, but is designed to be completely open, so I assume you could strip OpenMoko off and put GPE/LiMO/Slackware/Whatever on there if you wanted to (well, maybe not Vista).
Also, as this comment points out, service providers can't control what people use to connect to their network -- if you've got a T-Mobile SIM card, you can pop it into any GSM phone and dial away.
Hopefully once the software is there, handset manufacturers will start realizing that this is a low-cost way to sell some powerful, hassle-free handsets and we'll start seeing more PC-style handsets on the market soon.
No jokes, please
Most of these 'stacks' are application stacks with the actual protocol/modem part that controls the RF hardware locked away in a chip elsewhere. So in terms of breaking the network, the risk is very low, I'd liken it to the damage that someone could do with a PC connected to the Internet via dialup to a premium rate number. Also remember here that the network operator can disconnect you if they want (think of how pre-pay SIMs are managed, or the black lists for stolen phones that identify device IMEIs). A spam zombie PC connected to the Internet via flat rate broadband is probably much more harmful prospect, yet fixed line ISPs manage with this fine.
Therefore I think the problem is that the networks don't want competing services. The faster they get over this and realise that they are stifling innovation and usage of their connections the better.
It is not a matter of service provider or operator, it's the law (often - not in every country). In many countries _type approval_ is required, if the device does not pass the type approval tests, it is illegal to sell or use it. A cell phone is really very different from a PC. System upgrade images made available by some phone manufacturer (Nokia, Sony Ericsson) do pass through type approval.
Critics need to know these details, if they are true. If not, they need to refute them. Either way, this post was informative to me and would be to others if it weren't below most people's thresholds.
Have you driven a fnord... lately?
You must wait a little bit before using this resource; please try again later.
There is also the Silicon Valley Homebrew Mobile Phone club at http://hbmobile.org/ where the focus is to take the open phone concept all the way to open up the hardware. i.e. use only hardware components that have a spec that anyone can see. Many of the parts used in phones need a manufacturer NDA to get the spec, and this applies to some of the parts used in linux based phones. SVHMPC is using Gumstix CPU's which come with a Linux 2.6 kernel, and building up the software from there.
This isn't always true. Many high end phones have an apps processor dedicated to applications and UI with a separate modem processor. In some systems the interface between these can be as basic as a serial line with AT commands (yes, really). It is the modem part that requires rigorous Type Approval since that could easily degrade network performance by transmitting needlessly or out of spec.
Because of the type approval requirements of the the modem, it is often quite sealed and isolated from the apps processor. Changing the software stack on the apps processor should therefore not be a risk to operators, unless it is competition to their apps and services they fear...
You can run your own software on quite a few GSM phones, which typically don't have any memory protection in order to save on hardware costs. You can pretty much do anything with a Symbian based phone that you want and there haven't been any massive service outages yet.
The control issue is only about selling you extras. They want you to have to purchase $2 wallpapers, ringtones, etc. from the carrier.
Sadly, that's why FOSS always fails: there is no incentive to power through that last 5% which means really everything. If it's not fully functional, large corporations are going to ignore it- they have no patience to play with tech toys, unlike the lunix community, which exists solely to tinker with a tech toy and dream of the day it can accomplish more than Windows 95. Sadly, 1995 is getting farther and farther away every day, as are Microsoft's tail lights.
Coming soon: software to turn a bunch of mobile phones into a Beowulf cluster.
Ah, hell, my poor little Treo 650 can't even manage to run all day on a single battery with its dumb little single-threaded OS checking my e-mail and you want it to start folding proteins? The poor dear would burn itself to a crisp in 23 minutes.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
But whats the point of this when you can't go gown to your local phone store and buy these things off the shelf. Now when I first heard that people were releasing devices that run Linux I was very, very interested in purchasing one or at least testing one out. Much to my surprise these devices were no where to be found. Not only are they not at the local phone store I have emailed countless people about getting a linux based phone and no one has even replied, are these phones vaporware or what? Does anyone have one? Also I only know of one of these phones that has 3G, now really in this day and age 3G just doesn't cut it and does anyone even want GSM phone? I'm surprised they aren't going directly to landfill. I know that I sure don't want a technology that is EXTREMELY out dated. Where are the "Open" HSDPA phones and even 3G phones?
... the same way they have not subsidised mobile phones with DECT [1] capability.As a result mobile phones with DECT never made it to the shops.
d less_Telecommunications
Martin
[1] http://en.wikipedia.org/wiki/Digital_Enhanced_Cor
Maybe you want to have a look at Symbian Signed so see what kind of loops and hops you have to go thrue to run your own software on a smart phone.
t Summary
It's the same for Windows Mobile and will be the same for the iPhone.
And, of course: a network provider can ban the use of specific API's and or unsigned software on subsidised/branded phones.
Martin
[1] https://www.symbiansigned.com/app/page/dev/devcer