> On Mar 16, 2015, at 4:49 AM, Dr. H. Nikolaus Schaller <h...@goldelico.com> 
> wrote:


> I am just curious to understand how people think and decide such things in 
> March 2015.

Here's my take on this:

My particular vertical market need is for a very small computer that:

-Boots up upon receiving an external power (charging) signal.
-Customizable boot code
-Runs Linux and can run whatever full-blown Linux tool/app we need at 
startup/boot without any user intervention
-Can acquire and use a GPS signal at 4Hz or better.
-Has an audio input (headset/mic jack)
-Has Wi-Fi
-Has accelerometers
-Has at least 8GB storage

The GTA02 does a fine job at the above, the external GPS antenna capability is 
a bonus. That's why we bought hundreds of them (everything Openmoko had left in 
2010), for one particular vertical market use.

What (else) would we need in a GTA05 ?

-Better GPS (10Hz, Glonass, Galileo, BeiDou, offline A-GPS)
-LTE or at least "4G-ish" capability
-GSM Certification to operate in major countries 
-Dual SIM card capability
-Dual internal MicroSD cards (RAID)
-Retain external GPS antenna port
-Support Invensense IMU chip for advanced motion processing (9 DOF)
-Barometer/Temp sensor
-Infrared blaster
-RFID/NFC capability
-Fingerprint scanner
-Built-in GPIOs with externally-accessible connector
-USB OTG, USB 3.0 ?
-Built-In Ethernet capability
-Built-in A/D converter with externally accessible connector
-Built-In RS-232/RS-422 ports, externally accessible
-Built-in camera with an external port to add a remote cameras (say up to 3 
meters from device). Basically multiple USB 2.0 ports at a minimum.
-A decent-sized multi-touch or Glove-Friendly screen (two versions ?), the 
GTA02 screen is much too small by modern standards (we only use the screen for 
stuff like Wi-Fi password data entry by the end user)
-HDMI outputs (support two external screens)
-Speakers (like all phones)
-Offer various battery sizes (thicker backs).

For the external connector for serial and GPIO, flat round contact pads with a 
docking station on the back like Garmin uses on many units.

I believe such an open device would be highly successful in a Kickstarter 
campaign. The M2M (Machine-to-Machine) market is huge. Such a device would have 
broad appeal. Size is not very important, the capability/connectivity: Yes!

Ubuntu phone, but for true hardware hackers and companies with vertical needs 
that are not met with current smartphones or by Raspberry Pi or Arduino 

GTA04's limited 1Hz GPS was a killer for us... didn't bother finding USB camera 
options for the USB OTG on the GTA04 as a result, nor the mysterious built-in 
camera option (not ideal for us, remote camera is what we need).

The M2M folks want low-level hardware support (block diagrams, schematics, 
etc...) and open software drivers (abandonware is the problem here from 
commercial vendors).

Offer a version that's a smartphone and a bigger version that's basically a 
small computer with a built-in very smart UPS, with the ability to add 
expansion cards for non-mobile uses.

Personally I am not a purist, I don't mind inevitable closed aspects of some 
hardware/firmware that don't have open equivalents (the Invensense IMU stuff or 
LTE modem for example).  Nothing on the market offers what I'm looking for... 
and I'm sure I'm not alone seeking the Holy Grail of connected small mobile 
computer that's not iOS or Android.

The current proposed Ubuntu smartphones all have serious flaws (no microSD on 
the Meizu and on the EQ 4.5 they can't even bother mentioning which exact 
Mediatek processor they are using !!!) and pretty much none of the 
expandability and interface capability that I'd like to see in the device. 

When you can't even publish a proper detailed hardware spec sheet for your 
Ubuntu phone, you know lower level support is going to be an absolute 
nightmare. That's what Ubuntu Phone is now.

Better yet, the phone could have an internal small expansion slot with routing 
to the external connector pads. Need a bizarre interface ? Build a board and 
you are done, no internal soldering mods required.

I would expect demand for such a device from the Drone market alone to be 

More realistic than the current Phoneblocs-type project.  An open phone with a 
bunch of expansion ports. How hard can that be ?

What would it take ?  $5M ? $10M ?  $30M ? 


Openmoko community mailing list

Reply via email to