Hacker News Reader

Comment by lexicalscope | Hacker News Reader

This is because this is entirely intended for software defined radio and modular radio - around which the entire point is the RF component does not contain as much signal processing logic, and that is done in software. Restricting software defined radios to specific frequencies in hardware would take away the entire point of modular/software defined radio.

View on HN

Did you miss the conversation about how this fucks over people who want to install custom firmware on their routers for legitimate administration and traffic-shaping? There are plenty of uses for radio software that don't involve going outside approved frequencies.

I agree - but this rules doesn't apply to anything except SDR and modular systems that are also consumer devices - which are not most consumer devices.

That said - I am not condoning the FCC policy at all - I'm actually against it - I just don't really care that much because it impacts nothing I do and so few use cases I care about.

>"I don't care about X, because I never X."

You sure have spent a lot of time ITT explaining to everyone, advocating, practically that the FCC ruling is unimportant because you think that your own interests will be unaffected. Well, congratulations, but if we were all so short-sighted we'd eventually be reduced to only the hobbies and activities that a majority of us approve. We get it. You don't care, because you don't think it affects you.

I've actually said a few times I think it's a bad ruling - I'm just not practically impacted by it :). That said though I think being alarmist about it does not help the cause - it makes people less likely to actually listen to you once they realize you were being an alarmist. I actually agree with the points in the article - I just disagree with how they were made.

In this thread my only point is the FCC hardware restricting instead of requiring software restriction neuters SDR to begin with so is a dead end. It's all or nothing insofar as "protections" go. My view is still it should be nothing, but I don't think it is as big a deal as its being made out to be.

> this rules doesn't apply to anything except SDR and modular systems that are also consumer devices - which are not most consumer devices.

Um, what? Pretty much every device that does wifi has a SDR in it.

Does it affect cellphones? From the description I would have assumed it did, but I haven't seen much talk about that.

Even if it does apply, the radios on modern phones are already black boxes and completely detached from the rest of the phone. Some consider them a significant security threat already.

Contrary to the common interpretation, I've yet to see where in the new rules it is required that manufacturers implementing the new 5GHz U-NII device software security requirements is required to forbid 3rd-party firmware. Yes, there is an administrative document that asks questions about how such updates are prevented, but if you read the full document in context it also asks a lot of other redundant questions, and the FCC have since responded to Ars questions stating that it was not their intention to ban alt firmwares - just that the administrative processes starting up at the moment probably assumed that it would be necessary for the host device to do this to meet the new requirements. And since when does answering a regulatory compliance question in the negative mean that your application will automatically be rejected? All of the responses are used to help an FCC assessor arrive at a proper conclusion, potentially with further clarification sought on each point - it is not a hard script that you must always answer every requirement in the positive (in fact in many cases this would be impossible).

The new regs themselves do not state this requirement. It lists several possibilities for manufacturers to guarantee conformant emissions from their device, several which will continue to allow 3rd-party OS firmware.

Admittedly, the brave new world looks like region-locked devices and cheaper routers that truly are locked down in the exact ways we don't want, but that is not a hard FCC requirement, just a side-effect of the new regs on APs that have poor separation between OS and radio module.

    There are plenty of uses for radio software that don't involve going outside approved frequencies.

Except unlike ISM bands, U-NII 5GHz spectrum has been carved up with consultation of the 5GHz primary (licensed) users in each country and granted exclusively for U-NII conformant devices.

Unlike 2.4GHz ISM, nothing gives you the right to transmit on 5GHz U-NII bands (well, there's a bit that overlaps with secondary amateur spectrum) than otherwise permitted through the same FCC approvals process every device manufacturer must undergo.

That was the case before the new rules. Now the new rules are imposing sucky requirements for U-NII device software security.

However, that's been largely misinterpreted in every discussion I've seen recently.

For some context, check out https://wirednot.wordpress.com/2014/01/07/what-else-is-in-th...

You really don't want U-NII devices configured for Japan to be stomping on licensed spectrum in the US; you also need all that power negotiation, radar/interference avoidance algorithms in your radio so we don't get the same 2.4GHz mess happening in 5GHz.

Reply on news.ycombinator.com