Weekly Edition
Return to the Kernel page
| |||||||||||||
Sam Leffler releases HAL source
Sam Leffler, maintainer of the Atheros hardware abstraction layer (HAL), has released the source under an ISC license. Previously, this piece of the MadWifi driver for Atheros wireless chipsets was only available in binary form. "In his announcement, Sam states:
Coincident with the release of this code I have concluded my agreement with Atheros whereby I had access to information about their devices. This means that in the future all fixes, updates for new chips, etc. will need to be a community effort. Atheros states the Linux platform will be the reference public code base so folks wanting to add support for other platforms will have to scrape the information from there."
(Log in to post comments)
FCC? Posted Dec 1, 2008 16:49 UTC (Mon) by dmarti (subscriber, #11625) [Link] Did the FCC have anything to say about this? iirc that was one of the early arguments for keeping the HAL proprietary.
FCC? BS! Posted Dec 1, 2008 17:01 UTC (Mon) by pizza (subscriber, #46) [Link]
As someone whose day job involves writing drivers for various wireless network devices, I've always maintained that the "we can't release code/specifications due to FCC regulations" excuse was utter BS.
Atheros's (very welcome) about-face (first ath9k, then ath5k, and now madwifi-hal) leaves me feeling quite vindicated.
FCC? BS! Posted Dec 1, 2008 20:22 UTC (Mon) by pr1268 (subscriber, #24648) [Link] As someone whose day job involves writing drivers for various wireless network devices, I've always maintained that the "we can't release code/specifications due to FCC regulations" excuse was utter BS. Well, that's refreshing to hear. Perhaps this notion (and Atheros's change of heart) will dissuade other Wi-fi manufacturers from hiding behind that (non-)excuse.
Not really 100% true. Posted Dec 1, 2008 21:41 UTC (Mon) by khim (subscriber, #9252) [Link] As someone whose day job involves writing drivers for various wireless network devices, I've always maintained that the "we can't release code/specifications due to FCC regulations" excuse was utter BS. Reality is somwhat more complex, of course. FCC regulations are pretty vague but if FCC has an issue with you it's big deal. Given benefits and drawbacks someone decided that to keep source secret is good defence. Later s/he was overriden - someone from the top said "yes, I know what I'm doing - this is not needed". That's how corporations do such things. I seriously doubt actual members of FCC were involved...
Not really 100% true. Posted Dec 2, 2008 0:11 UTC (Tue) by rahvin (subscriber, #16953) [Link]
He didn't say they were. He said someone at <wifi company> made the decision that it was safer to not release the code for the rare chance it would get them in trouble with the FCC because that trouble can be very dangerous to the survival of the company. Later someone higher up said it was an acceptable risk because the risk was so minuscule resulting in the code being released.
FCC? Posted Dec 1, 2008 18:31 UTC (Mon) by schwaang (guest, #19827) [Link] It may have helped that when the FCC issued a rule on software-defined radios last year, the Software Freedom Law Center made a legal argument for why this doesn't restrict FOSS made for SDRs. However, I distinctly recall an Intel employee speaking at a FOSS conference (probably linux.conf.au last year) blaming the equivalent of the FCC in Japan for their not releasing wifi source code. That was a new one on me, and I expect there will always be some country somewhere with laws that provide that excuse, for companies that want to use it.
tentative summary of argument Posted Dec 2, 2008 11:17 UTC (Tue) by pjm (subscriber, #2080) [Link]
If I may be so bold as to summarise said legal argument even though I've only skimmed through it and IANAL:
They conclude that it is permitted for individuals to use uncertified FOSS in software radios (consistent with the FCCs position on experimental or specialized equipment. Such equipment is exempt from the certification requirements if it is not marketed to the public and is only used under controlled conditions, citing 47 C.F.R. Part 2.803). In §3.3, they go so far as to argue that it is permitted for equipment manufacturers to ship uncertified software intended for individuals to use in this way, so long as they do not ship that software as part of a device; though this stronger proposition (that radio equipment manufacturers can develop and distribute such non-certified software) is conspicuously absent from the Conclusion section.
I suppose this means its fine for Ubuntu etc. to ship such software, but if Dell etc. wants to ship Ubuntu preloaded then they need to modify it to use certified software (which will probably involve proprietary software due to the higher burden to prove FOSS as sufficiently secure), probably instead of, rather than in addition to, the uncertified software.
Obviously youd be foolish to rely on my casual summary; I provide it only for the sake of the curiosity of casual readers like myself.
Comments of confirmation or denial by people who've read more carefully than I are welcome.
tentative summary of argument Posted Dec 4, 2008 12:08 UTC (Thu) by farnz (guest, #17727) [Link] One quibble - although certifying FOSS generally (e.g. all versions of Linux on the market) is harder than certifying proprietary software, certifying a clearly identified release (e.g. Ubuntu 8.04.1) is not. Thus, Canonical could certify their Atheros drivers in Ubuntu to suit the commercial needs of Dell, and could update that certification whenever the drivers change. This would avoid Dell needing to use proprietary software to prove that things are sufficiently secure.
Sam Leffler releases HAL source Posted Dec 4, 2008 12:38 UTC (Thu) by mkflint (guest, #50223) [Link]
I'd welcome the release of the code, but I'm a little saddened by the termination of their agreement. To say "OK, you can release the code, but now you're on your own" is quite an odd thing to do.
Are they supportive of open source code, or not? Or is it just me who's seeing mixed messages here?
Sam Leffler releases HAL source Posted Dec 4, 2008 17:18 UTC (Thu) by JoeBuck (subscriber, #2330) [Link] I suspect that the real issue is money. Times are hard, companies are laying off staff, and they aren't in a position to pay engineers to work on the code.So they are open-sourcing as well as dumping the code. Better than nothing, I guess.
Sam Leffler releases HAL source Posted Dec 6, 2008 18:09 UTC (Sat) by giraffedata (subscriber, #1954) [Link] So they are open-sourcing as well as dumping the code. I didn't see anything about dumping the code. The comment was about Atheros no longer giving Sam information about its products, and how that seems inconsistent with Atheros no longer keeping the source code closed. But I think it's entirely consistent. Sam's special NDA access to the information for use in producing closed-source code, making him the only person who could develop it, is the opposite of open source. If there's an inconsistency, it's that Atheros didn't release the information to the public at the same time as the code.
|
|||||||||||||