[Arm-netbook] CC3000 Wi-Fi for MCU

[lkcl luke]

On Wed, Jan 25, 2012 at 6:58 PM, Gordan Bobic <gordan at bobich.net> wrote:
> On 01/25/2012 06:53 PM, lkcl luke wrote:
>> On Wed, Jan 25, 2012 at 6:48 PM, Gordan Bobic<gordan at bobich.net>  wrote:
>>> On 01/25/2012 06:21 PM, lkcl luke wrote:
>>>> 2012/1/25 Henrik Nordström<henrik at henriknordstrom.net>:
>>>>
>>>>> I draw the border slightly different but similar. If the device would function the same with romed firmware as with a firmware blob where the cpu i run my free OS on is only uploading to the device and not executing it in any manner then it's acceptable by aggregation imho.
>>>>
>>>>    lots of people take this view (myself included).  and then it turns
>>>> out that the proprietary firmware monitors, eavesdrops and informs on
>>>> your location, or allows for the arbitrary execution of code that is
>>>> downloaded over-the-air.
>>>>
>>>>    whoops.
>>>
>>> Just like it would if it was in ROM rather than RAM. The point is moot.
>>
>>   yes i know.  it's not ideal.  the absolute ideal is to write the
>> entire firmware - program the device from scratch.  then you know it
>> can be trusted.
>
> Which you cannot do unless the hardware takes a firmware blob from the
> driver.

 ... which is fine, now, because the source code for that firmware is
available, can be audited, digitally-signed as audited, and the hash
published so that people who trust those people who carried out the
audit do not themselves have to spend the time learning about security
and programming.

 anyway - we're kinda off-topic, but it's still relevant as it's
background to what's behind this initiative.

 heck, with bari's idea to push the power usage up to 10 watts it'd
even be possible for people to run entire proprietary OSes and i will
_still_ happily sell them an EOMA card (just not a proprietary OS),
will thank them for the money and say "thank you for enabling me to
pursue my goals of increasing software freedom". etc. etc.

 but that is just me.

l.