Originally posted by bridgman
View Post
Announcement
Collapse
No announcement yet.
Nouveau: NVIDIA's New Hardware Is "VERY Open-Source Unfriendly"
Collapse
X
-
I go to bed and wake up to pages of discussion. Time zones differences suck.
-
Vegan comparison has its limits
Originally posted by bridgman View PostGood analogy, but you missed the part where the vegans are fine with getting meat sauce on their steakhouse salad as long as they don't have to explicitly order it...
"Yum, this place has the best salads. I don't know what they put in the dressing but it's awesome !!"
Now an example of reality: I use a system derived from Ubuntu with AMD open drivers and the firmware. No auto-updates can occur in my systems, and unlike on Windows a malicious "linux-firmware" update requires the attacker to have Ubuntu's signing key or an APT exploit. It's a HELL of a lot tougher to crack than a Windows box, it's worth noting that the recent cross-platform Firefox exploit used by the FBI to work around Tor in the Freedom Roads case used a Window-only payload and a JS loader for it. No amount of unfree firmware,not even using that huge Nvidia driver blob plus a wifi card with an ndiswrappered Windows driver would have made that exploit work on Linux Mint. You would have had to run a Windows Torbrowser install in Windows, maybe in WINE for it to work. Maybe in WINE the exploit would have crashed for that matter.
You can also bet that both closed driver blobs and video firmware are very closely watched for odd behavior (like network access) for exactly the same reason systemd is watched the same way-trust or lack therof by many users.
I do not have and cannot get open software (GPL, BSD or even NDA to me only) to run my camera, however. This is of little security consequence so long as the camera stores only to camera card, does not have wifi capability, and the card is physically moved to the computer to be read and wiped. The separates a trusted and untrusted device. Since smartphone vendors are known for malicious baseband radio firmware (due to CALEA) I choose not to have one at all, I do not trust any OS replacement to secure them unless the cell radio is removed to a separate device. Here it is the Tor Project's recommendations I trust, and a phone meeting them does not do anything my netbook cannot do. The printer was chosen to run with open drivers in the computer, but no doubt has its own CPU and closed stack upon it. Thus, it is not trusted not to keep copies of files, nor not to "tag" them for the pigs. As for the "Internet of things," I keep connected shit that can watch your movements out of the house. Why do you think there is an active movement against electrical "smart meters?"
Leave a comment:
-
I guess the most accurate analogy would be :
- our vegetarian restaurant serves salads with yoghurt dressing on the side, so you can add it to the salad yourself
- a second restaurant serves the same salad with the same yoghurt dressing, but the salad and dressing are pre-tossed
- the steakhouse down the road has the oil and vinegar dressing you prefer, on the side, but you have to bring your own salad greens (or, as you say, this could be the "bland salad")Last edited by bridgman; 02 May 2015, 09:56 PM.
Leave a comment:
-
Originally posted by chithanh View PostIf you open a vegetarian salad bar but insist on selling your delicious salad only with yoghurt dressing, don't expect any vegans to recommend your place. They'd rather go to the steakhouse restaurant and order a bland vegan salad that this place happens to sell. All your complaining about how you work with local organic farmers and what you've accomplished for the salad community and how much better the lacto-ovo vegetarians and flexitarians are served now will not help.
"Yum, this place has the best salads. I don't know what they put in the dressing but it's awesome !!"Last edited by bridgman; 02 May 2015, 09:11 PM.
Leave a comment:
-
Originally posted by nanonyme View PostHm, I don't immediately get why it needs to be either or
At the start of the project we discussed this with developers and distro packagers -- fortunately there was unanimous agreement that freeing as much driver functionality as possible should be the priority, and that binary microcode would be OK as long as the license allowed easy redistribution.
Leave a comment:
-
Catalyst is a bit of a special case. It's really aimed at workstation customers who in turn tend to use enterprise distros, so we have dedicated support and direct liason with a couple of distros (RH, SUSE and more recently Canonical/Ubuntu).
What we weren't covering was the consumer market, so that's where I figured we could do the most good with the open source stack. Even with a common kernel driver, the open stack is going to continue to be the focus for most markets outside of workstation.Last edited by bridgman; 02 May 2015, 07:53 PM.
Leave a comment:
-
Originally posted by chithanh View PostI didn't think that you deny that this effect exists, I just said that you (or your management) did maybe not grasp the full extent of it.
Case in point: As a distro packager for X11 related packages I was astonished on how much NVidia catered to our nvidia driver maintainer, including providing a contact within the company and responses to our problem reports. However when reporting problems via the unofficial tracker for Catalyst there was hardly if ever a reaction from AMD people.
.
Leave a comment:
-
Originally posted by bridgman View PostNo, that was actually one of the key selling points we used when setting up the program.
Case in point: As a distro packager for X11 related packages I was astonished on how much NVidia catered to our nvidia driver maintainer, including providing a contact within the company and responses to our problem reports. However when reporting problems via the unofficial tracker for Catalyst there was hardly if ever a reaction from AMD people.
Originally posted by bridgman View PostFSF is saying open microcode and no vendor support for open drivers is better than significant open support and binary microcode. That seems to be the opposite of what Linux users want and need.
If you open a vegetarian salad bar but insist on selling your delicious salad only with yoghurt dressing, don't expect any vegans to recommend your place. They'd rather go to the steakhouse restaurant and order a bland vegan salad that this place happens to sell. All your complaining about how you work with local organic farmers and what you've accomplished for the salad community and how much better the lacto-ovo vegetarians and flexitarians are served now will not help.
Leave a comment:
-
Originally posted by bridgman View PostNo, that was actually one of the key selling points we used when setting up the program.
The whole point is that we had to choose between providing good open source driver support while keeping the microcode closed or providing minimal support and opening up some of the microcode. We can't do both, and no other vendor does either.
FSF is saying open microcode and no vendor support for open drivers is better than significant open support and binary microcode. That seems to be the opposite of what Linux users want and need.
Leave a comment:
-
Originally posted by chithanh View PostAnd I believe (but don't have proof, so feel free to disagree) that your employer totally underestimates the importance of winning the minds of the Linux nerds, who are often relied on in hardware purchasing questions. Poor Linux drivers (and the Catalyst Linux driver is deficient in many aspects) would have hurt AMD's Windows sales too.
The whole point is that we had to choose between providing good open source driver support while keeping the microcode closed or providing minimal support and opening up some of the microcode. We can't do both, and no other vendor does either.
FSF is saying open microcode and no vendor support for open drivers is better than significant open support and binary microcode. That seems to be the opposite of what Linux users want and need.
Leave a comment:
Leave a comment: