Lipstick on a fetus

Harald Welte laforge at
Tue Sep 4 18:02:34 CEST 2007

On Wed, Aug 22, 2007 at 12:56:00AM -0400, Ken Young wrote:
>    I've really enjoyed playing with my neo1973, and I'm very glad I bought
> it.   I've written a little PDA application for it which I hope other
> people will eventually have fun with.   But I must say I'm a bit worried
> by the implications of the effort that has been put into the new 2007.2
> interface.
>    I have made perhaps 6 successful phone calls with my neo.
> I have *never* been able to use the data service.   I have *never*
> successfully received a phone call.   I have *always* had to reboot the
> neo after placing a phone call, before I could make a second phone call.
> After the reboot, I must run alsactl.  I have an AT&T SIM which
> occasionally works, and a known-good T-Mobile SIM which has never worked
> once in the neo.   Perhaps a few more hours looking through the Wiki
> would allow me to find solutions to all of these problems.   Perhaps
> a few well-timed questions on the IRC channel would enlighten me.   I
> have not really cared, because the fact that I can occasionally make
> a call means my hardware is almost certainly OK, and I had assumed that
> the high-level developers would fairly quickly (since the "Mass Market"
> version is scheduled for October) fix up the problems with the phone
> application.   I'm primarily interested in the neo as a PDA and GPS
> device.   But as far as I can tell, the usability of the phone
> software has not improved at all in the few weeks I've had the phone
> (I've tried several different builds of the code).

We are very well aware of this situation, but please let me emphasize
once again that we are a _really_ small team that builds everything,
from hardware design through hardware implementation, testing,
production process, logistics, UI design, framework, applications,
distribution management, software packaging, hardware packaging,
webshop, sales, ...

I don't really know if those numbers are public, but the last time I've
seen our list of employees in the Taipei office, there are something
like 25 people on the openmoko side, and about the same number on the
FIC Mobility hardware side.  Of both the hardware and software R&D team,
only half of them are actually working on GTA01/GTA02.

So any progress will inevitably be slow.  And everybody has many
different tasks/jobs inside the company.   Since the whole idea of GTA02
was introduced very late in the project progress, it meant that a
significant amount of resources had to be put into the GTA02 hardware
design, verification, production process, system level software.

Those are all resources that were taken away from finishing the software
side, especially on the GSM part.  The issue is like this:  The more
effort we put into the GTA02 hardware, the higher the chance is that we
won't have any serious hardware problems.  The GSM software side (and
actually hardware side) of GTA01 and GTA02 are identical.  We have
verified the GSM part up to a point where we are confident that there
are no hardware bugs.  This means there is no urgent need to work on the
GSM side.  

However, if we reduce our GTA02 hardware design and verification
efforts, then we might get the GSM side running faster.  But at what
expense?  At a delayed GTA02 hardware, potential liabilities, or god
beware even re-calls or lots of failing units.

This combined with the fact that openmoko is being spun off into a
separate company (actually a whole series of companies in various
countries), the entire company moving into new offices - which again
drew significant R&D resources away.

So please exercise patience.  When we started selling phase1, I made an
internal comment that I don't think the software stack will mature a lot
during the first 2-3 months after the hardware release.  Compared to
that, there was actually quite a lot of progress, mainly thanks to
Mickey, Stefan, Daniel and O-Hand.

- Harald Welte <laforge at>         
Software for the world's first truly open Free Software mobile phone

More information about the community mailing list