: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /home/twopigs/surjpatel.com/includes/common.inc on line 1782.
: preg_replace(): The /e modifier is deprecated, use preg_replace_callback instead in /home/twopigs/surjpatel.com/includes/common.inc on line 1782.
If you can do it, my hats off to you | Surj

If you can do it, my hats off to you

It would be cool to design your own cell phone. But I hope you guys have thought it through. Hackers are generally know software. Not hardware. Are you guys designing everything from scratch ? Do you know how to build the RF components of a cell phone ? Do you have a PHD or worked with a hardware developer before? Building a cell phone is one thing, but building one that works is totally different. I would think it would be easier to reverse engineer a current phone. May not be as much fun, but at least you're starting off with something that works.

But with enough effort and smarts, I'm sure anything is possible. Forget about the screens and the OS for now. None of that is important to even make a simple phone call. I don't know the original intent for this project. You want to design cool applications like VoIP applications that would be compatible with Skype and such ? There are thousands of people working in the wireless industry. They've thought about this stuff before. Reason why it hasn't come out yet ? #1 reason is because the technology isn't here yet. GSM is very crappy. Data rates are very low, and network latency is even worse. Even the best stuff out right now like EVDO Rev0 isn't good enough. #2 reason is that people don't need this stuff. And if they wanted it, they couldn't afford it. Anyone can make a super phone, but nobody will pay $2K for one. Doesn't make business sense. Reason for not having a super cool O/S like Linux on a phone is because Linux is not designed for cell phones. Cell phones are simple devices which are getting more complicated as time goes by. OSes have to be optimized for the instructions it requires. One of the major problems with cell phones today is battery life. You simply cannot have a super phone because it will literally work for an hour or two at most. Now, if you don't care how big your phone is, I'm sure you can make one from the parts of your laptop. The first cell phone required a backpack afterall. Much of the development of phones are based in unix systems in the development stages. But once they figured all that out, they will move onto a more pratical system. I think it would be more fruitful to reverse engineer a phone. Figure out how to call hardware functions and create a new layer to interact with this. I'm prefer CDMA, so I won't comment on GSM :)

good luck guys!
-K

>>There are thousands of

>>There are thousands of people working in the wireless industry. >>They've thought about this stuff before

Yes i was one of them. They also have a modus operandi of fear rather than innovation. Quite rational when you invest 10 billion quid in a network. Me? I just want a phone i can dial numbers with, get data and SMS/MMS through an API in Python without banging my head against a wall. Windows vs Linux for developers - same argument.

Thanks Ken!!! :-)

Okay - Your right in most of what you say but also let me briefly give you oyr P.O.V and some answers.

>>Are you guys designing everything from scratch ?

No, if you read through the posts you'll see that the key enabler is the Telit GM862 module. Basically a black box GSM phone with serial line control for call set up and data layer. Add a mic and speaker and some Hayes AT commands and you are away. You need millions to design this stuff from scratch.

>>Do you have a PHD or worked with a hardware developer before?

No neither I or Deva have. I want to go for one when my kids are older perhaps. We both are masters grads from the MIT Media Laboratory. http://www.media.mit.edu/

>>I would think it would be easier to reverse engineer a current >>phone. May not be as much fun, but at least you're starting off >>with something that works.

I'd actually say the reverse is harder in our case. Starting with components you know about means that you know how subsystems function. Black boxing will not be complete information about the phone unless the manufacturer agrees to divulge that. In most cases due to carrier pressure they wont.

>> None of that is important to even make a simple phone call.

Yep. Casey Halverson has done so already. Google him. Next we have to :-)

>>You want to design cool applications like VoIP applications that >>would be compatible with Skype and such ?

Not quite that ambitious with GSM GPRS - data rates max at about 24kbps. I want to work on a demos where Plaxo or Linked in is your adress book - simple stuff like that. One click dial without two presses for confirmation as it is now in J2ME ( Malware protection)

>>Anyone can make a super phone, but nobody will pay $2K for one. >>Doesn't make business sense.

Your dead right. My mum wouldn't be able to hack linux kernels :-) This is a phone for developers as Linux boxes tend to be in the PC world.

>> One of the major problems with cell phones today is battery life.

If i can get twenty minutes out of Rev 1 and not burn my hands I'll be overjoyed. Acknowledged.

>>Now, if you don't care how big your phone is, I'm sure you can >>make one from the parts of your laptop.

You can. Also Zaurus, Ipaqs and ARM boards.

>> I think it would be more fruitful to reverse engineer a phone. >>Figure out how to call hardware functions and create a new layer >>to interact with this. I'm prefer CDMA, so I won't comment on >>GSM :)

Check out what Haralds up to...

http://ganesha.gnumonks.org/~laforge/weblog/linux/a780/

Ken - if you start on a CDMA version let us know - it will be important.

Thanks for your post mate,

Surj