Eben Moglen's linux.conf.au keynote [LWN.net]

The issue in Japan

Posted Apr 25, 2005 21:05 UTC (Mon) by ncm (subscriber, #165) [Link]

Are we talking about radio receivers, and receiving signals they think we ought not to be equipped to hear? Or are they just worried about transmitters?

The issue in Japan

Posted Apr 25, 2005 21:47 UTC (Mon) by brouhaha (guest, #1698) [Link]

I don't know about Japan, but in the US, the rule established by the FCC is that it must not be easy for the end-user of a device to modify it to operate on frequencies or at power levels that are unauthorized. And US law requires that no products be sold that can intercept cellular phone conversations, except to law enforcement.

Atheros claims that the former problem is why they will not release a fully open-source driver for their 802.11g chipset; most of the driver is open-source (GPL'd, so it also qualifies as Free Software), but there is a "HAL" portion that is only supplied as a binary under a non-free license.

The actual FCC ruling is not that clear; many people believe Atheros is overreacting, since their chips are not a true SDR (Software-Defined Radio). Still, it is the job of corporate attorneys to minimize the corporation's exposure to legal liability, and to some extent not providing source code for the HAL does that.

If you're interested, you can read the actual FCC Report and Order (PDF format, 24 pages, ET Docket No. 00-47, released September 14, 2001).

The issue in Japan

Posted Apr 25, 2005 22:53 UTC (Mon) by Ross (subscriber, #4065) [Link]

And don't forget that all equipment must obey the broadcast flag starting
this summer. Not only that, but it must prevent users from bypassing it
with safeguards which are non-trivial to bypass. That's probably just not
possible in a tool which is distributed as source code. I don't know if
Japan has the equivalent of the broadcast flag but it is likely since it was
required to be legislated in international treaties.

The issue in Japan

Posted Apr 26, 2005 0:20 UTC (Tue) by dberkholz (subscriber, #23346) [Link]

It could be. If you need to crack a generated 2048-bit key or something crazy like that.

The issue in Japan

Posted Apr 26, 2005 1:06 UTC (Tue) by Ross (subscriber, #4065) [Link]

I don't see how. Just modify the code to skip the authentication routine.
It's the same problem as any client-side authentication. It doesn't actually
do anything. It's just a speedbump. It would be a bigger speedbump if it
was binary-only, but still not real security.

Broadcast Flag

Posted Apr 26, 2005 5:38 UTC (Tue) by Ross (subscriber, #4065) [Link]

Sorry to reply to myself, but here's an interesting interview with the
founder of pcHDTV (the only Linux-only HDTV capture card) which discusses a
few of the requirements of the broadcast flag:

http://digitalmedia.oreilly.com/2005/04/20/pctv.html

As you can see there is no way a software implementation, especially an
open source one, could meet the requirements.

Broadcast flag not /quite/ activated ... yet

Posted Apr 26, 2005 17:58 UTC (Tue) by Max.Hyre (subscriber, #1054) [Link]

Broadcast flag not /quite/ activated, yet

The broadcast flag is being challenged in court, mostly under the belief that the FCC has no damn' business regulating hardware. Their charter is to regulate use of the airwaves, and how one accesses such use is beyond their pale.

The suit is being driven (if I have it right) by Public Knowledge, a ``Washington DC based advocacy group working to defend your rights in the emerging digital culture''. Their report, in brief, is that the matter is currently in the U.S. District Court of Appeals for the District of Columbia, and the court requested a second round of argument (unusual, I believe). The additional briefs were presented early this month, and everyone's holding their breath.