[ltp] Re: Intel 2100b WiFi Card

[wes schreiner]

Frank Schaeckermann wrote:

>
>
> wes schreiner wes-at-infosink.com |Linux on ThinkPads| wrote:
>
>> Martin Gramatke wrote:
>>
>>> Frank Schaeckermann wrote:
>>>
>>>
>>>
>>>> [...] really a pitty and shows how far the
>>>> Intel commitment to Linux REALLY goes...
>>>
>>>
>>>
>>> To double the pressure on Intel, I wrote them a similar mail. :-)
>>>
>>> I don't understand why they keep the technical info under disclosure. A
>>> distrustful mind might suspect a microsoft deal.
>>>
>>>
>>
>> No, the reason is the FCC (the US Federal Communications Commission). 
>> The latest generation of wlan cards have the radio controlled 
>> directly by software (the driver) not by firmware like the older 
>> cards. Since now the code to set frequency and trasmit power is in 
>> software, companies can't open-source their drivers (or give out the 
>> technical specs) because that would give an easy way to adjust power 
>> and frequency, and giving that kind of information away to people not 
>> licensed by the FCC violates FCC rules. Companies must abide by FCC 
>> rules in order to be certified in the US, and they need certification 
>> in order to sell in the US. So blame the FCC.
>>
>> The best we are going to see for now are drivers with closed-source 
>> binary lumps in them, like in the Atheros and ADMtek drivers. The 
>> binary lump is a hardware access layer and controls the frequency and 
>> transmit power settings. If you read the Atheros driver source you 
>> will see that several unexplained values are passed to this binary 
>> lump. Those values might have something to do with frequency or power 
>> but there is no easy way to tell.
>>
>> Eventually someone will reverse engineer the hardware and we will 
>> have entirely open-source drivers for at least one chipset, but I 
>> expect that by the time that happens nobody will care much about 
>> 802.11[abg].
>>
>> wes
>>
> Isn't that kind of stupid at the part of the FCC to believe they can 
> control that kind of knowledge? 


Sure, but they have been doing this kind of regulation for literally 
generations, so they aren't going to stop just because we think it is 
stupid.

> First, only the people with some kind of hightened curiousity and 
> technical knowledge would really care about playing around with the 
> frequencies and power, right? 


No, I might want to use an illegal channel for a point-to-point link in 
a crowded area but not have the required technical skills with the 
current products, though I might if I had nice documented source to the 
entire driver.

> Second, those are exactly the people smart enough to pick up the 
> Windows driver, disassemble it and find out what they want to know 
> anyway. 


Nope, that only gets them so far. It doesn't matter if they take apart 
the Windows driver or the closed-source binary Linux driver, all they 
see is that the driver pokes mysterious values into the hardware. 
Getting to the meaning of those values is not as easy as looking at the 
driver disassembly.

> And third, yes this would be outside the legal boundaries, but then... 
> playing with the frequencies and power probably is anyway. Go figure.
>
> Frank


It is quite legal for someone with the proper license to go fooling 
around with a radio, but they stand to lose their license if they use a 
radio illegally. Messing with a radio is not illegal, just using it 
incorrectly is. Go ask any ham radio operator. What the FCC is doing is 
trying to make it difficult for the non-technical to modify and use 
their radios illegally. Reverse engineering in and of itself is not 
regulated.

And really, there is nothing new here. The older wlan cards, like those 
with Prism chipsets, all have closed-source firmware and few people 
complain about that. Those chip makers have given out the technical info 
needed to talk to their firmware, but have never openly documented the 
interface between firmware and hardware (nor can they, due to the same 
FCC regulations). The only thing new with the latest generation of wlan 
chipsets is that the firmware is gone and now the driver talks directly 
to the hardware. So instead of the technical obsfucation being in the 
firmware, it now is in a binary-only driver. I don't like this either, 
but I can see where the FCC is coming from. Yes, ultimately one can get 
around this problem, but not many will try and that's the whole point.

wes