Dotan Cohen wrote: > On 28/11/06, Erik Hemdal <ehemdal@xxxxxxxxxxx> wrote: [...] >>As I understand it, Intel has to do it this way so that they can >>have an open-source driver and still maintain compliance with >>radio-interference regulations. I'm not an expert about the rules, >>but I have seen RF interference cause some really difficult problems >>at work, so I don't minimize that requirement. > > Ah, have people been hacking the open source driver to make it more > powerful? Ruining everybody else's good time just to boot their own > signal? You'll have them in every lot. I don't know that there is any evidence of that happening. It's more likely a CYA decision on the part of Intel's lawyers. There's an article on kerneltrap about an OpenBSD developer, Damien Bergamini, creating a driver that didn't need this binary daemon[1]. He suggests that Intel used the binary daemon "probably because they wanted to reduce the complexity of their firmware which were bloated and buggy." The binary daemon is the one big strike against the ipw3945. If I'd have done my homework better I would have avoided this card because of that. For the Dell's, the alternative is the Broadcom-based card which is only recently working with an open-source driver, so that would have meant choosing a different brand of laptop entirely. I have read that the binary daemon may be going away at some point though, as there's no way the driver will make it into the mainline kernel with such a requirement. But I don't know the status of that. -- Todd OpenPGP -> KeyID: 0xBEAF0CE3 | URL: www.pobox.com/~tmz/pgp ====================================================================== It's important that people know what you stand for. It's more important that they know what you won't stand for.
Attachment:
pgp6LNUqjSOOG.pgp
Description: PGP signature
