MonkeyofDoom has quit [Read error: Connection reset by peer]
knttl has quit [Ping timeout: 255 seconds]
knttl has joined #neo900
knttl has quit [*.net *.split]
x29a has quit [*.net *.split]
mzki has quit [*.net *.split]
mzki has joined #neo900
x29a has joined #neo900
knttl has joined #neo900
MonkeyofDoom has joined #neo900
chomwitt1 has quit [Ping timeout: 240 seconds]
xmn has quit [Quit: Leaving.]
DocScrutinizer05 has quit [Disconnected by services]
DocScrutinizer05 has joined #neo900
neo900 has joined #neo900
neo900 is now known as Joerg-Neo900
Joerg-Neo900 has quit [Killed (verne.freenode.net (Nickname regained by services))]
neo900 has joined #neo900
neo900 is now known as Joerg-Neo900
DocScrutinizer05 has quit [Disconnected by services]
DocScrutinizer05 has joined #neo900
MonkeyofDoom has quit [Ping timeout: 255 seconds]
arcean has joined #neo900
Guest30720 has joined #neo900
<Guest30720>
hey guys
<Guest30720>
Allah is doing
<Guest30720>
sun is not doing allah is doing
<Guest30720>
to accept Islam say that i bear witness that there is no deity worthy of worship except Allah and Muhammad peace be upon him is his slave and messenger
<Joerg-Neo900>
you sell hardware that you feel meets the criteria.<< HMMMMMMMMM
<Joerg-Neo900>
*I* *feel* we meet the criteria zhe FSF *should* have set up for RYF
tsuggs has joined #neo900
<ravelo>
hmmm
<ravelo>
true
drrz has quit [Ping timeout: 255 seconds]
<atk>
Any chance of making a Neo900 which works without the baseband?
<atk>
(nobody will buy it, but it would qualify the project for RYF)
<Joerg-Neo900>
sure, no problem
<Joerg-Neo900>
the 'baseband' aka modem is a module like an USB stick. we simply can 'unplug' it
<Joerg-Neo900>
but, that's pretty much bizarre nonsense to get a RYF for such neutered device
<Joerg-Neo900>
the true problem with RYF is: FSF has no clue what they're really talking about. they *assume* that some chips have firmware, some have updateable firmware, other have none. In fact they never know
<Joerg-Neo900>
the whole concept isn't sustainable
<Joerg-Neo900>
RYF needs a better definition of what'
<Joerg-Neo900>
s breaking the rules and what doesn't
<Joerg-Neo900>
aka. simply "better rules"
<Joerg-Neo900>
they set up riles for RYF in best intention, but those rules need improvement
<Joerg-Neo900>
s/riles/rules/
jonsger has quit [Ping timeout: 248 seconds]
<Joerg-Neo900>
>>What we want most is for manufacturers to publish designs for hardware under free licenses. But the minimum is to publish key technical specifications sufficient to write free drivers for their hardware<< sounds pretty sane
<Joerg-Neo900>
alas that's not what I recall RYF is all about
<Joerg-Neo900>
if it was, then Neo900 would be "the reference implementation" for RYF
<Joerg-Neo900>
((sounds pretty sane)) given those drivers are LINUX drivers
drrz has joined #neo900
Airwave has joined #neo900
<Joerg-Neo900>
which implicitly defines outer perimeter of what FSF considers their domain of expertise and control
<Joerg-Neo900>
in that sense it's irrelevant if my RYF device with GNU/linux running on it sends a blob of data via USB to a network dongle that connects to internet, or via WiFi to internet, or via USB to a modem module to update some flash storage there
jonsger has joined #neo900
<Joerg-Neo900>
it's also irrelevant if that USB has a part and plugs or is hardwired
<Joerg-Neo900>
s/part/port/
<Joerg-Neo900>
it's even irrelevant when we use SPI instead USB for transmitting that blob, and if the module on the other end of this "data cable" is a modem module or a wifi module or a BT module
<Joerg-Neo900>
I know it would be a bery-very-VERY-nice-to-have to be able to provide FOSS firmware for all those modules, but honestly this shouldn't be a K.O. criterion for RYF
<Joerg-Neo900>
for sure a "it must not be able to receive binaries that look like CPU code (as long as we suspect it stores those binaries and usses them for a suspected CPU inside the blackbox" is a very troublesome and not helpful argument
<Joerg-Neo900>
this is clearly an argument born from free SOFTWARE foundation, which isn't really helpful in hardware context
<Joerg-Neo900>
RYF never made a fuzz about firmware in HDD or CD/DVD drives, not even about the AMT aka ME in intel chipsets
<Joerg-Neo900>
for me RYF means: user must always be in full control (at least potentially, even when they don't exercise it) of every single byte that gets executed on the CPU(s) that run the GNU/linux system
<Joerg-Neo900>
this must also be true for P-machines (CPU emulators), and microcode sequences
<ravelo>
someone should tell ryf to stop this
chomwitt3 has joined #neo900
<Joerg-Neo900>
this definition sounds sufficiently precise and comprehensive for me, and could replace all the "you must not have an update path for modules that run closed firmware" voodoo in the recent RYF
chomwitt4 has joined #neo900
<Joerg-Neo900>
or even more terse: the system behavior must always be 100% deterministic from a user's POV (note that even intentional randomness is deterministic, while an unsolicited update pushed by manufacturer isn't, from user's POV)
chomwitt3 has quit [Ping timeout: 252 seconds]
<Joerg-Neo900>
this however implies we have already established what 'the system' comprises of. Otherwise a CD with a read error in drive would break determinism
<Joerg-Neo900>
so we qualify for RYF a 100% when we simply don't ship _any_ software with the device
<cc___>
Joerg-Neo900: well a brick can qualify for RYF, if sold as such
<Joerg-Neo900>
then everything fine, we explicitly never offered software
<cc___>
Joerg-Neo900: why are you so obsessed by this RYF thing ? if the GTA04 never got it, why do you think the Neo900 could get it ?
<Joerg-Neo900>
except for a factory-test and BSP package
<Joerg-Neo900>
it simply annoys me to no end
<cc___>
Joerg-Neo900: no phone ever got the RYF label and no one will get it soon, you're just running a project that gets as close as possible from freedom, why bother ?
<atk>
cc___: I think it might be good publicity to get FSF to endorse a project as ambitious as this.
<cc___>
you've got two options then : sell a version of the device without the modem and make a dedicated shop for it, or wait for the OsmocomBB project to be finished, working, usable, used, people put in jail for it, years of legal battles to recognise the right for people to run their own software on a GSM chip, sell the device fully liberated, and get validated as RYF :)
<Joerg-Neo900>
our device clearly meets all requirements lusted there, as far as I can tell
<Joerg-Neo900>
listed, even
<Joerg-Neo900>
...and then a lot more on top
<cc___>
I dunno, I just assumed it would be the major problem :)
<cc___>
the other one being the GPU but if you can make a free driver even barely run on it it would be ok I guess
<Joerg-Neo900>
there is such free driver since ages
<Joerg-Neo900>
honestly, FSF has a very nice general salvatory clause in their RYF that allows them to decide based on whatever arbitrary ideas they may have this moment or the next one: >>However, each certified product carries with it contractual obligations that may differ from those listed on this page.<<