A crowdfunding campaign to build a free baseband

Paul Sokolovsky pmiscml at gmail.com
Wed Apr 8 12:49:00 UTC 2015


Hello,

On Tue, 7 Apr 2015 23:03:45 GMT
falcon at ivan.Harhan.ORG (Spacefalcon the Outlaw) wrote:

> Paul Sokolovsky <pmiscml at gmail.com> wrote:
> 
> > Good project. It would be great if you worked on current, not
> > historical, hardware, but I agree it should be done step by step.
> 
> How would you propose I (or any other free community member for that
> matter, outside of the closed circle of NDA-bound Qualcomm employees)
> could work on current as opposed to historical hardware when there is

Just to clarify, when I'm talking about "current hardware", I mean
hardware on which current off-the-shelf products are made, not 4G and
other stuff completely uninteresting to (freedom?) starving people in
Africa.

> absolutely NO documentation publicly available for any baseband chips
> that are newer than the "ancient" ones from TI I'm working on?

Wait, dude, so you have documentation and you still do that "reverse
engineering"? Then you either wasting your time or availability of
documentation is overrated.

> Do you have your own personal armed forces who could invade USA with

Speaking of army, you appear to play one-man army with this
reverse-engineering effort. I was astonished to see contents of
https://bitbucket.org/falconian/freecalypso-reveng repo, where there're
thorough notes on various topics, and most importantly, bunch of tools
written - in raw C - all adhoc to the specific project. It goes up to
following: there're surely ARM/Thumb disassemblers widely available,
but shiver me timbers, freecalypso project needs its own personal
implementations!

And that's what bothers me with all this stuff - "reverse engineer to
make open-source" community doesn't learn. Year over year it happens in
following patterns:

1. Zealots write every piece from scratch, down to otherwise widely
available assemblers/disassemblers. Of course, all that effort goes to
toilet, because their tools are written for specific purpose in
non-reusable manner, and concealed in unrelated (to reusability) repos
in strange VCSs on funny sites.
2. Everyone else uses cracked IDA in shame.
3. After some time both 1&2  swear never to waste their time on
that again, then in few years some new interesting piece of hardware
comes up with alzheimerish vendor forgetting to release a datasheet and
it all happens again.


> troops, take control of the city of San Diego, storm Qualcomm's
> headquarters with a Spetsnaz unit and seize their chipset
> documentation and source code?  If not, how do you propose we could
> obtain the necessary documentation and reference source which they
> are unwilling to release freely?

If Qualcomm is that bad, let them be. World is big, there're other
"villains" to look at. Mediatek churns out mobile chipsets like sh%t,
and don't shy away from cents' profit on an item, so they have low-end
offerings which are cheap, widely available, and largely compatible from
generation to generation.

[]
> Some of those laid-off employees surely didn't want to see their work
> perish, hence the source code and documentation got leaked.

In China, leaking is something like a normal part of business process -
vendors are not yet ready to just release their stuff, but hurry to
leak before their competitors did. Because otherwise shanzhai will
churn out others' stuff first, and then markets (bazaars) are suddenly
full of talks about X instead of yesterday's favorite Y. And even if big
customers don't care about leaks, they surely care about what's in
critical acclaim now.

> I am not aware of any even remotely comparable source+documentation
> leak being available for any newer chipset, hence I don't see any
> realistic prospects of how we could run free modem/baseband/radio
> firmware on anything newer.

I can't believe you missed the fun of
http://www.bunniestudios.com/blog/?p=4297 . You're the specialist, you
can see what's available for yourself. But available a lot (but then
not all, but that's apparently the case with TI stuff either).

> But here's the thing, there is still a *huge* unsold surplus left of
> these old chips, enough to build tens of thousands of phones or modems
> - so I really don't see what's the big deal about these chips being
> "historical" as you said.  And if we ever do exhaust that surplus,
> hiring some Chinese "pirate" chip fab to make verbatim clones of these
> old chips 

Ah, piece of cake! Except that Chinese did that 20 years ago, and for
some 15 years offer their own stuff - nowadays $3 a chip, $8 a module,
$15 a phone, $25 a smartwatch, selection and availability is great,
leaked sources included in the overall offering.

> > Also please consider having a mirror of your codebase on github to
> > popularize the project and improve access to it.
> 
> It is currently on bitbucket.org - how is it worse than github?  And
> the source repository is Mercurial, not git - can github support Hg?

Bitbucket is more rough around the edges, e.g. seeing it now offers
ability to edit file online (pull request created automagically behind
the scenes), I tried to fix a typo in one of your docs, 1st time it
gave me error, 2nd time hanged for 5+ mins. Turned out it created 2
fork repos, and still didn't manage to submit a PR, which I had to do
manually.

But point is not in that - git is more popular VCS and github is more
popular hosting service. If you're interested in popularizing your
project (to bring freedom to more users), you should be interested it
making it more accessible. It's also a backup and redundancy - when
FBI, NSA, KGB, and FSB will seize you, your code will live a little bit
longer than without such mirroring.

Also, let me know if you plan to act on suggestion to add more
diversified perks - I'd be happy to forward your announcement to more
lists, but most people will only look at the campaign once, so if they
don't see something right away to make them act, they never look back.

> 
> VLR,
> SF



-- 
Best regards,
 Paul                          mailto:pmiscml at gmail.com


More information about the Replicant mailing list