[pdonis]

> they're only talking about modular radio systems or software defined radio systems

Which, in practice, means any computer/electronic device that has a radio in it.

> this is targeting consumer equipment

Including "consumer" equipment that is being used for research, software coding, etc.

[csirac2]

    Which, in practice, means any computer/electronic device that has a radio in it.

No. FCC approvals apply to the modular transmitter. Another approval may apply to the overall device.

[pdonis]

> FCC approvals apply to the modular transmitter.

Including the software that controls it. Which, in practice, includes the OS of a laptop or the firmware of a router, as discussed in the article. Technically that might not be the "entire device", but it's the part that matters.

[csirac2]

> Which, in practice, includes the OS of a laptop or the firmware of a router, as discussed in the article.

The article is continuing to imply that the FCC is explicitly banning alt firmware for 5GHz WiFi devices. That is only the case if the module manufacturer fails to come up with any other way of getting their device approved. I don't see anywhere in the regs where banning alt. firmware is a goal of the regs.

Given that this also means a brave new world of region-locked/region-specific 5GHz WiFi devices, and the added compliance burden for integrators ("host device manufacturers") trying to use modular transmitters that rely on host device manufacturer controls to achieve FCC approval, I am hopeful that this will change the way that WiFi manufacturers lock down their radio firmware - and that is exactly what the FCC wants.

The situation where a modular transmitter places additional avoidable test/conformance/approval burden on the host device will also be a PITA for manufacturers, not just end-users.

Check my other comment here https://news.ycombinator.com/item?id=10256905

[pdonis]

> I don't see anywhere in the regs where banning alt. firmware is a goal of the regs.

It's not an explicit goal. But the article argues (and I tend to agree) that the practical result will be manufacturers locking down their devices to prevent alt firmware from being loaded, since that will be the easiest and cheapest way for them to demonstrate compliance.

> I am hopeful that this will change the way that WiFi manufacturers lock down their radio firmware - and that is exactly what the FCC wants.

Are you saying the manufacturers will come up with some way of locking down their radio firmware that still permits something like OpenWRT to be installed on a router (or Linux on a laptop, for that matter)? Why would they bother when they could just lock the device down completely?

[csirac2]

> Are you saying the manufacturers will come up with some way of locking down their radio firmware that still permits something like OpenWRT to be installed on a router (or Linux on a laptop, for that matter)? Why would they bother when they could just lock the device down completely?

Yes. In the actual regs (which nobody seems to read), they suggest a list ("including but not limited to") a number of mechanisms by which manufacturers may choose to control the portion of their radio software that would impact the validity on their RF testing/validation/compliance results. One of them is signed firmware blobs: to me, that's the easiest, cheapest non-invasive method for WiFi module makers that won't create a huge compliance burden on the host device manufacturer. You go from having to load an unsigned firmware blob anyway, to a signed one. Which is a huge step in the right direction for firmware security anyway. As for modules which don't have blobs but do have the means to create non-compliant emissions just through the driver: it doesn't seem like much of a stretch that they could run new region-locked revs of their modules, or at least move those previously adjustable RF parameters/behaviours over to a signed image in a $0.10 SPI EEPROM chip.

All of this isn't just a PITA for users, it's a PITA for the device manufactures as well.

Particularly for laptop manufacturers. They could save $5 locking down the entire laptop, something they've never been able to do even when they try, or they could spend the extra $5 and get the module that's got a stand-alone certification and only requires them to submit a reference to the module's own FCC approval and some demonstration that the gain of the antennas in their product are in-spec.

[pdonis]

> One of them is signed firmware blobs

I'm not sure I understand what you're describing here. Suppose I have a router and I want to run OpenWRT on it. Are you saying that the router will have basically two "firmwares" in it? One that controls the radio chip only, and is signed, and has some kind of defined driver interface; and another that controls the rest of the device, and has a driver that talks to the signed blob, so as long as OpenWRT has that driver, I'm good?

Or suppose I have a laptop and I want to load Linux on it. Are you saying the wifi chip inside the laptop will have a signed firmware blob that controls the radio, and has a Linux driver interface, so I can load Linux on the laptop and talk to the chip?

Assuming the above is correct, how different is it from the way these devices are designed now?