RMS / FSF / OsmocomBB / Free Software GSM phone
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi! Some months ago, Richard Stallman (rms) has contacted me regarding my work with OsmocomBB. As it is not a big surprise to us, he is really interested in the project. After all, it would be the first mobile phone that he himself would be able to use, given that there is no proprietary software on the baseband anymore. And of course it would enable loads of other freedom-loving users to finally have an alternative to the proprietary telephony world. Here in Morocco, I've had some further discussion on the topic face to face with him, and he is sort of unhappy with the fact that nobody is working on making an actual self-contained phone (no matter how simple or limited in features) that can be used by a regular user as 'just a phone'. I explained to him that our motivation is mostly a different one (research, security, GSM-attached PBX, etc.) and thus there is no intrinsic motivation to work towards a more user-friendly version. Furthermore, we are system level hackers with an interest in communications, not particularly people who like to work on a UI or usability. So we agreed to make a public call for volunteers to wokr on that aspect of the phone. I understand this will likely cause some effort on our side (fencing off poeple who don't have the neccessary skills, integrating such code, finally deciding on a RTOS to use, etc.). However, I more or less see it as my (and our?) duty to realize the potential of our protocol stack and baseband firmware. Next to all our own self-motivated personal interestes, there is a bigger cause that we can help along: Free Software based telephony. So the least we can do is to try to find somebody who can work on that part, and help developer[s] to interact with our code. It's the question of whether we are just hacking away on our personal little pet, or if we try to achieve something bigger. I've already drafted a version of the 'job description' and with some luck the FSF will soon publish it, reaching out beyond our existing small Osmocom community. e will try to draft a similar job description related to MS-side GPRS support (L1, RLC/MAC, ..). That would be yet another area where we would appreciate some contribution, and which eventually be important beyond our existing voice telephony capabilities. Regards, Harald - -- - - Harald Welte <laforge@gnumonks.org> http://laforge.gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iD8DBQFNvs0JXaXGVTD0i/8RArkDAKCONleE9HJ2i1jp7L/XnMY0YweIjACbB8Lz puH7X4lkqY8AqzU0SAlOAxM= =Cmo8 -----END PGP SIGNATURE-----
I think this is awesome. I would try to put in some help if I could, as long as someone else told me what to do (:-). I have experience with fairly low-level C programming and I did have a job writing laboratory control software (in Forth) when I was in college (1980s). I have several Osmocom-class phones I have purchased over the last year. I did run the hello world app on a phone, just to make sure I could. I compiled the tool chain myself with a script I found and customized that made a really clean install (let me know if anyone is interested, but there are better scripts out there now). Most of the cell stuff talked about on this list is above my head, though I have gained a basic understanding of what's going on. I found the c3 lectures fascinating, and watched quite a few. I'm definitely interested in doing something for this project if I can. Harald, I know you and the other people contributing to this project are interested in this more from a research and understanding perspective. But I also agree with RMS about the need for a fully open phone. You guys have revealed just how weak the GSM network is, with its closed stack assuming that all participants were noble knights who always followed the rules (:-). It actually concerns me just what the telcos might be capable of doing, especially after seeing the talks on silent SMS messaging. There's no good reason for this software to be so closed. This is the big one too. There are lots of other areas where proprietary software is dominant, but few affect as many people directly as cellular phones. I also have no doubt that a usable phone would get a LOT of attention. Could help encourage efforts in other areas. Color laser printers are another area of concern, because of the embedded ID on every printed page. I am not surprised in the least that RMS is interested, even very interested. Scott On Mon, May 2, 2011 at 6:26 PM, Harald Welte <laforge@gnumonks.org> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Hi!
Some months ago, Richard Stallman (rms) has contacted me regarding my work with OsmocomBB. As it is not a big surprise to us, he is really interested in the project. After all, it would be the first mobile phone that he himself would be able to use, given that there is no proprietary software on the baseband anymore. And of course it would enable loads of other freedom-loving users to finally have an alternative to the proprietary telephony world.
Here in Morocco, I've had some further discussion on the topic face to face with him, and he is sort of unhappy with the fact that nobody is working on making an actual self-contained phone (no matter how simple or limited in features) that can be used by a regular user as 'just a phone'.
I explained to him that our motivation is mostly a different one (research, security, GSM-attached PBX, etc.) and thus there is no intrinsic motivation to work towards a more user-friendly version. Furthermore, we are system level hackers with an interest in communications, not particularly people who like to work on a UI or usability.
So we agreed to make a public call for volunteers to wokr on that aspect of the phone. I understand this will likely cause some effort on our side (fencing off poeple who don't have the neccessary skills, integrating such code, finally deciding on a RTOS to use, etc.).
However, I more or less see it as my (and our?) duty to realize the potential of our protocol stack and baseband firmware. Next to all our own self-motivated personal interestes, there is a bigger cause that we can help along: Free Software based telephony. So the least we can do is to try to find somebody who can work on that part, and help developer[s] to interact with our code.
It's the question of whether we are just hacking away on our personal little pet, or if we try to achieve something bigger.
I've already drafted a version of the 'job description' and with some luck the FSF will soon publish it, reaching out beyond our existing small Osmocom community.
e will try to draft a similar job description related to MS-side GPRS support (L1, RLC/MAC, ..). That would be yet another area where we would appreciate some contribution, and which eventually be important beyond our existing voice telephony capabilities.
Regards, Harald - -- - - Harald Welte <laforge@gnumonks.org> http://laforge.gnumonks.org/
============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6) -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux)
iD8DBQFNvs0JXaXGVTD0i/8RArkDAKCONleE9HJ2i1jp7L/XnMY0YweIjACbB8Lz puH7X4lkqY8AqzU0SAlOAxM= =Cmo8 -----END PGP SIGNATURE-----
Hi,
After all, it would be the first mobile phone that he himself would be able to use, given that there is no proprietary software on the baseband anymore.
Well, if you're extremist about it, there is still the DSP. Not much we can do about it without making our own ASIC tough :p
Here in Morocco, I've had some further discussion on the topic face to face with him, and he is sort of unhappy with the fact that nobody is working on making an actual self-contained phone (no matter how simple or limited in features) that can be used by a regular user as 'just a phone'.
Well, even if we have reached major milestones (IMHO) and are able to make actual phone calls, there is still a bunch of things missing even for voice / SMS and reliability problems in some cases. Besides the UI there is also a lot of things that need to be done to be usable as a phone, like being able to charge the battery, or at least not drain it in a few hours by having all chips fully powered on all the time.
e will try to draft a similar job description related to MS-side GPRS support (L1, RLC/MAC, ..). That would be yet another area where we would appreciate some contribution, and which eventually be important beyond our existing voice telephony capabilities.
The RLC/MAC layer can certainly be useful for a lot of reasons (both MS and BTS side), but is anyone seriously gonna use GPRS non-edge ? Cheers, Sylvain
Sylvain Munaut wrote:
it would be the first mobile phone that he himself would be able to use
Well, if you're extremist about it, there is still the DSP.
Yes why I wanted to have a go at making modem. Unfortunately not top priority, so it's going real slow. (But tomorrow I'll make some RF tests with real simple trx and soft modems. Fun!)
Not much we can do about it without making our own ASIC tough :p
Getting a modem to work at all would be the first step anyway, then optimize for size and cost, ideally of course reusing cheap hardware already on the market. //Peter
Hi,
Not much we can do about it without making our own ASIC tough :p
Getting a modem to work at all would be the first step anyway, then optimize for size and cost, ideally of course reusing cheap hardware already on the market.
I don't think it's that hard given we have working examples. TX is easy. RX is already implemented in RX The main part is having properly timed TX/RX. Sylvain
Hi Sylvain, On Mon, May 02, 2011 at 08:13:50PM +0200, Sylvain Munaut wrote:
After all, it would be the first mobile phone that he himself would be able to use, given that there is no proprietary software on the baseband anymore.
Well, if you're extremist about it, there is still the DSP. Not much we can do about it without making our own ASIC tough :p
I do not think the DSP is an issue, as it is programmed by mask ROM. As far as I understand RMS' position, any mask ROM code on a separate processor is tolerable for him. Just code that resides in FLASH or other r/w memory is required as Free Software.
Well, even if we have reached major milestones (IMHO) and are able to make actual phone calls,
of course, I didn't want to say that the project has not reached milestones or is not making progress. I just wanted to say that the speed has slowed down and recent developments go in other areas than to make a 'stupid phone': sniffing, running phone as BTS, fuzzing/security stuff, general reasearch, lcr-integration, etc.
there is still a bunch of things missing even for voice / SMS and reliability problems in some cases.
Sure. RMS and I were not talking about a 'end user production ready' phone, but something that -even though initially unreliable and very limited in functionality- can be used as a phone. Progress in stabilization, testing in real networks etc. can be made from that point on.
Besides the UI there is also a lot of things that need to be done to be usable as a phone, like being able to charge the battery,
roh has done some battery charging work in the past, but i guess it was never merged.
or at least not drain it in a few hours by having all chips fully powered on all the time.
i'm not so sure if that absolutely has to come first. A Free Software phone that runs for a couple of hours is better than none at all. The power management can be improved gradually.
The RLC/MAC layer can certainly be useful for a lot of reasons (both MS and BTS side), but is anyone seriously gonna use GPRS non-edge ?
It depends for what. You have to keep in mind the screen real estate and processing power. So for e.g. updating your identi.ca, or using som instant messenger instead of SMS it would be sufficient. Also, the RLC/MAC will be independent of the L1 (like our GSM L2/L3 is), and it could be used on EDGE capable devices if other baseband processors are supported later on. I seriously hope the Calypso is not the last chip we support. Cheers, Harald -- - Harald Welte <laforge@gnumonks.org> http://laforge.gnumonks.org/ ============================================================================ "Privacy in residential applications is a desirable marketing option." (ETSI EN 300 175-7 Ch. A6)
Hello, On Mon, May 2, 2011 at 11:26 AM, Harald Welte <laforge@gnumonks.org> wrote:
Some months ago, Richard Stallman (rms) has contacted me regarding my work with OsmocomBB. As it is not a big surprise to us, he is really interested in the project. After all, it would be the first mobile phone that he himself would be able to use, given that there is no proprietary software on the baseband anymore. And of course it would enable loads of other freedom-loving users to finally have an alternative to the proprietary telephony world.
For what it's worth, I work for an embedded systems shop doing microcontroller-level C, and would be happy to lend a small hand if the need arises. I don't have very much experience, but this is a project that I hope succeeds and I'll do what I can to see it happen. Regards, Ryan
participants (5)
-
Harald Welte -
Peter Stuge -
Ryan Mullen -
Scott Weisman -
Sylvain Munaut