FreeCalypso funding and direction

Mychaela Falconia mychaela.falconia at gmail.com
Sun Jun 10 06:20:13 UTC 2018


An addendum to my previous post: if no one steps forth to fund FC hw
projects while I am taking care of my medical/gender needs and
therefore these FC hw projects have to wait a few years until I am
done paying for my needs, I will probably do some work on the software
side of FreeCalypso in the meantime.

I don't think that I will be able to produce end-user-usable FC fw for
C139 or other pre-existing Calypso phones ahead of my desired (and
currently financially out of reach) FreeCalypso handset: the work
required to put together a usable UI for Motorola's 96x64 pixel screen
size while being denied funding for a proper 176x220 pixel UI platform
would be too unpleasant for me to do it on an unpaid volunteer basis.
However, there is one other FC software direction which also needs
some work and which I am more willing to take on: producing a modem fw
version (for modems, not phone handsets) that combines the freedom
qualities of FC Citrine (compilation with gcc from full source without
any blobs) with the functional completeness of FC Magnetite (GPRS,
CSD, audio services and other frills).

The to-be-created FC fw project that will combine the good qualities
of Citrine and Magnetite will be named Selenite, and the plan to get
there is as follows:

1: Exercise the hybrid config in Magnetite more heavily (it is the
config that is fully deblobbed except Nucleus, GPF and TI's proprietary
compiler), and fix any regressions relative to classic TCS211 or
l1reconst.

2: Create a hybrid-gpf config in Magnetite in which GPF will be fully
rebuilt from source.  We have the original corresponding source for
all of GPF except OSL and OSX components; for the latter I will need
to take my earlier source reconstruction from 2014-2015 (used in
Citrine) and polish it a little better.  Test and exercise it.

3: Start the Selenite source tree from Magnetite hybrid-gpf, and
replace the TCS211 binary libs version of Nucleus with the source
version from Comrade XVilka (again already used in Citrine); fix any
issues that arise at this step.

4: Implement a gcc-based build system in Selenite (as an option
available in parallel with the TMS470 compiler from Magnetite), and
get all of the code to pass compilation.

The plan outlined above is all software work and does not require any
money, so I am going to work on it while waiting for donations or for
my medical/gender expenses detour to be complete, whichever happens
first.  I will use one of my existing FCDEV3B boards with sleep
disabled for this work, plus the one-and-only TI-made D-Sample handset
connected to this FCDEV3B for the loudspeaker and mic for voice calls.
(Producing a new loudspeaker of our own to replace this historical
TI-made one will require money to hire an appropriate expert, hence
won't be possible without funding.)

My hope is that having modem firmware with full functionality (CSD,
GPRS, audio frills etc) that is completely free from blobs or
proprietary compiler requirements will help bring forth the funding to
make hardware for this awesome fw to run on.

Hasta la Victoria, Siempre,
Mychaela aka The Mother


More information about the Community mailing list