Slashdot Mirror
More on OpenBSD 3.7 Release
putko writes "As previously reported, OpenBSD 3.7 is released. Here's some interviews with the people behind the release about the new features, including information about which companies are complying with requests for documentation and permission to freely distribute required firmware, and which are not. Ralink Tech and Realtek 'GOOD,'Intel 'BAD.'
The next time I build/buy a wireless product, I'll want Realtek or Ralink Tech inside -- because getting software to work with it will be easier. Ralink Tech and Realtek are Taiwanese, by the way."
Yes, Ralink provides Free (as in GPL) Linux drivers for their rt2400/rt2500 cards here. My roommate has a laptop with an rt2500 wifi card, and it works beautifully in Linux.
0 1 - just my two bits
The madwifi drivers are not entirely Free - there's a large closed section of driver that runs on the host processor (it's not merely firmware for the card). People are working on drivers for the softmac prism54s, the Intel 2200 has an entirely open driver (but awkward restrictions on distributing the firmware. Thanks, Intel), there's an experimental driver for TI's acx111 hardware, and the RT2500 is an 11g part.
the full comment that gave it that legendy reputation is in the following source file (you have to scroll past the license boilerplate first).L ENG5
http://fxr.watson.org/fxr/source/pci/if_rl.c?v=RE
note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
Would you care to share which packages are broken?
I've install 60+ packages with no problem whatsoever.
The only catch is that firmware is still closed-source. It can be downloaded, but I am not certain about redistribution conditions.
redistribution is not permitted by intel's license (hence the 'Intel BAD' in the story leadup)
...
and, straight from iwi(4)
The official person to state your views to about this issue is peter.engelbrecht@intel.com at (858) 391 1857.
vodka, straight up, thank you!
To be 100% precise, the original port by William Jolitz and his wife to the i386 architecture (probably one of the smallest BSD development teams of all time, and probably the least-credited for the work they did) had potentially encumbered code.
Most of the "potential" encumberances were header files and a few relatively minor bits of code that AT&T could easily have just released to the public. However, AT&T wanted to play dirty, and decided that ANYONE who saw AT&T code was "encumbered" and that therefore any code later produced must also be encumbered.
As was noted by commentators at the time, this means that those who use constructs such as "for" loops that use the ++ operator are in violation of AT&T Intellectual Property terms, as this specific practice originated from a book by an AT&T coder.
AT&T did lose their lawsuit against iBSD, but iBSD folded not long after because of the suit, IIRC. (The suit actually came about, not because iBSD were using AT&T stuff - the Jolitz' had been doing so for some time by then - but because they were selling it and not giving AT&T any kickbacks.)
Interestingly, as part of the SCO vs IBM case, SCO want to revisit that decision and get it overturned. Again, it seems to have little to do with Linux actually including any of this supposed code, but because IBM and SGI are making money from Linux and SCO isn't.
Personally, I think anyone who demands kickbacks of this kind should be banned from visiting any website other than MSNBC's weblog. Well, at least they'll find kindred spirits there.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
I bought the same card and it's very easy to tell the difference between revisions. The latest version of the card (revision 4) which contains the RT2500 chipset is a half-height PCI card, unlike the previous revisions which were full-height.
Please don't give up. We must discover what compels people to ignore the very simple rule that the only time you should ever use apostrophes to pluralize is with single letters (ie, I got four A's).
LOAD "SIG",8,1
So, yes, faulty designs do exist, and one of the best-known for it is also one of the best-known for not releasing hardware specs, which does tend to make for some interesting implications.
Of course, even when documents ARE released, there is often a lot that is UNdocumented. The 486 had an interesting "load/save all registers" instruction, which basically allowed you to preserve or restore a complete CPU state. The hardware industry is littered with all sorts of other obscure undocumented syscalls, which is one reason why Open Source drivers for 3D graphics cards generally underperform - not because they are no good, but because the proprietary drivers include undocumented calls which improve performance.
This goes along with why manufacturers are dead-set against reverse-engineering. Not because they fear someone learning some "industrial secret" that really IS something the manufacturer shouldn't divulge, but because they fear people discovering device commands that they currently sell to the highest bidder.
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
Henning Brauer: Nobody ever gave us anything back. A plethora of vendors ship OpenSSH--commercial Unix vendors (basically all of them), all of the Linux distributors, and lots of hardware vendors (like HP in their switches)--but none of them seem to care; none of them ever gave us anything back. All of them should very well know that quality software doesn't "just happen," but needs some funding. Yet, they don't help at all.
That just blows. A while back the OpenBSD team had to raise funds to acquire Dell hardware so that their CVS server could scale up. The CVS server that holds repositories for all Open* projects. You would think that one of these companies would have just donated the hardware. But nope.