RE: Keyspan Firmware fun
> If there were no reason to ever want to change it, then they wouldn't have
made it
> uploadable in the first place
True. Although I don't remember the last time I had the urge to tweak the
BIOS of some hardware device. Not only am I likely to break it (certainly in
the case of devices with flash/eeproms rather than RAM) but in doing so I'm
likely to void the warranty. And as a hardware vendor I'd probably rather
people not touch my BIOS based on the increased support costs alone! Why not
ask for the schematics too... I'd say you've got about as much chance as
getting that from them. Thus this information is nothing more than a string
of 1's and 0's to me, and probably the vast majority of Linux/Debian users.
However, there is clearly some legal issues with this approach. Tainting the
kernel code with non-free is obviously bad. Hey, how about I go write my own
driver, get it put in the kernel and then go sue all the linux distros/ftp
sites/magazines/etc. I think we need to find a balance, and storing the
firmware in files which are uploaded by a userland utility or (better yet)
the module itself may be just that.
> This amounts to the statement that we should abandon our principles
> because other people will never change to recognize that our
> principles are good ones.
>
> In fact, other people can and do change!
IMHO every extra device supported by Linux is one step closer to the
desktop. When we have the support of the masses hardware vendors will be
itching to get out of the box support, and will no doubt be lining up to
hand over the code (it seems we have keyspan's attention now anyway).
Insisting on it now is, IMHO counterproductive.
- samj
Reply to: