No - chromeos is the whole operating system with specific components turned off and optimized for specific vendor architectures. There are ton more differences but that's what jumps out to me at the moment.
Still ChromeOS and Android are both based on Linux, thus both export the same API to the userspace that Chrome could use on GNU/Linux.
There is a point about the hardware being predictable, that applies much more to ChromeOS than to Android. But the Chrome code could be the same for all three.
If you read the comments someone points out that they are using the exact same drivers for intel stuff on ChromeOS that people like me are using on on Thinkpads (it has an nvidia GPU but I rarely use it, next machine won't have) with no issues.
Not sure what is up with Chromium just recently in light of the potentially breaking adblockers thing but I'm happy that Firefox are back in the game (not that I ever stopped using them, Chrome/Chromium is only for dev and testing on my machines) since the single largest browser vendor also depending on advertising for the vast majority of it's income is a massive conflict of interest frankly.
Firefox for Android has gotten really good and with ublock origin and the lack of constant "sync everything to google forever" nagging makes mobile web much more pleasant.
I found this guide about how to remap the keybindings in firefox without having to build from source, which pushed me over the ledge into giving firefox a serious shot at replacing chrome: https://github.com/nilcons/firefox-hacks#binary-hacking-fire...
I'm biased but for me Chrome never replaced Firefox as a general use browser, it did for development (chrome dev tools has forced FF to up it's game - win/win for me) and I have to use it for testing for obvious reasons.
The thing with Firefox is that I know generally while they've made the odd misstep they do actually care about what they preach, I feel like Mozilla never had to say "Don't be evil" because there wasn't really any way they could be (on the scale of Google).
On that page they say "Extensions that request access to all websites still get installed with that access, so the default behavior has not changed." That seems counter to the whole discussion about Chrome's changes yesterday.
The linked blog doesn’t say anything about adopting this change - on the contrary, it asks Google to let them have more input on the extension spec.
“So I would like to publicly invite Google to collaborate with Mozilla and other browser vendors on manifest v3. It is an incredible opportunity to show that Chrome embodies Google’s philosophy to “focus on the user,”...”
ChromeOS is directly supported by the hardware vendors, so they work directly with manufacturers to provide stable drivers. In the general case with Linux on the desktop you have no idea what GPU is installed, so supporting all of the combinations of proprietary, open source, reverse engineered drivers, etc with various features is a nightmare.
This bug was closed three years ago, with the following
"Today we don't have a Linux GPU decode owner, so we have chosen not to activate something which we have no one to support and may have to break arbitrarily in the future for ChromeOS."
Is there anything to suggest that has changed?
Did someone offer to do it and were turned down?
I've seen patches but no offer to maintain the subsystem, etc that i can find.
But maybe i missed it?
I only did a few searches of the relevant mailing lists.
(Also, in my experience commenting on 3 year old closed bugs is not the way to make anything happen)
>(Also, in my experience commenting on 3 year old closed bugs is not the way to make anything happen)
Perhaps not, but this seems to be the only avenue available to many.
What would you suggest people who encounter a problem such as this do if they personally don't have the ability to personally fix the problem themselves?
This isn't an issue specific to chromium or even google. There are a lot of projects that seem to have an inside and an outside with those on the outside rightly-or-wrongly feeling unheard.
"What would you suggest people who encounter a problem such as this do if they personally don't have the ability to personally fix the problem themselves?"
There's a difference between getting someone's attention, and then them disagreeing with you.
If you are trying to get attention, i'd go with the mailing list in this situation.
If they just disagree with you, sometimes, you gotta accept that someone doesn't agree with you about the product vision and let it go (this assumes your constraints - they don't have ability/time to fix themselves). They will either be right or wrong, and it will succeed or fail.
"This isn't an issue specific to chromium or even google. There are a lot of projects that seem to have an inside and an outside with those on the outside rightly-or-wrongly feeling unheard."
Sure, and i think companies/people should be very clear with governance models, participation models, etc. I also think they should be very clear about things like this.
Actually one of the hugest clashes i often see is open source projects that do have a product vision end up clashing with minorities of their user bases who don't fit in that product vision (either due to changes or never fitting in the first place), etc. Those clashes often seem much more emotional for people.
(I honestly don't know enough about chrome anymore at all to know whether that's the case. I'm only talking about the general situation you are describing).
A lot of the inside vs outside cases that i've seen fall into "product vision decided inside, you are welcome to participate if you want but we aren't changing it" style of project.
It's more rare (though i've seen it) to see "we have no vision of where we are trying to go but we still are going to semi-ignore random contributions/etc. We're gonna make it so you think we might care, but not actually try enough that anything happens "
Worse, a lot of the time the latter prevents other competing projects from existing because people spend time trying with the existing one then are too burned out to do anything else.
The product ones often at least give straight answers, even if people hate the answers.
This seems totally reasonable. Google is primarily focused on building Chrome -- a product, with business oriented product goals. Anyone using Chromium has already agreed to be a tag-along. That was always part of the deal.
In general I find the category of folks who -- won't use Firefox, but want to pretend they're not basically just using Google Chrome -- to be very strange indeed.
That said, I agree that you're basically just using chrome.
I happen to use chromium for work because I do development for extensions and it's nice to keep chrome around as my primary test browser and chromium as my daily driver.
Plus I can dig through the source on chromium and recompile locally (admittedly, not super fun) to see if I might suggest additional extension features/apis.
Ok. And achieving those goals can be affected by people being unhappy about their decisions. Google can hire better people for lower salaries, the more they can convince the world that they are a force for good. And governments and the like (e.g the EU) can make rules and enforce them if they see behaviors that go strongly against the best interest of society as a whole. And so on.
I never understood the motivations of people who publicly defend corporations' short sighted decisions on the basis of "they are a business, they are only supposed to consider their bottom line, and aren't supposed to consider their public image in the process." Yes, corporations are allowed to make short sighted decisions, and people on the internet are allowed to complain when they do. (and yeah, I guess people are allowed to defend the corporations, but it simply encourages the companies to do things like this)
Obviously this isn't the worst thing that Google has done, but I support people calling them out on things like this.
Before anyone chimes in with "just fork chrome".. just to give you an idea of the magnitude of the project. The chromium source code has some 16k cpp source files. Millions of lines of code. All the libraries combined the source advancea at a rate of thousand commits a day. You fork chromium now and tomorrow you're already way behind on merging upstream changes. It takes serious engineering effort just to be able to build that behemoth.
And yes the OpenGL drivers are unstable and no, Khronos CTS passing doesnt mean anything wrt stability. The most trusted gfx platform is currently DX11 backend in libAngle.
Heh, amusing about the memory requirements. The vast majority of chromebooks are 4GB ram. I suspect less than 1% are 8GB ram (they cost 3x as much as the 4GB models).
Which is by far the biggest pain point of using Ubuntu on my XPS 13. YouTube absolutely destroys battery life if I watch it in my browser. Only solution I've found is to watch Youtube videos with mpv, but I feel a bit bad about that because it deprives the creators of ad revenue.
This seems worse in every way than just enabling libvaapi, which is already somewhat supported, except for being specifically disabled on non-ChromeOS Linux.
I wish Firefox would already implement GPU accelerated video decoding / encoding. Right now WebRTC is very CPU heavy since acceleration is missing completely.
42 comments
[ 3.2 ms ] story [ 97.0 ms ] threadThere is a point about the hardware being predictable, that applies much more to ChromeOS than to Android. But the Chrome code could be the same for all three.
Not sure what is up with Chromium just recently in light of the potentially breaking adblockers thing but I'm happy that Firefox are back in the game (not that I ever stopped using them, Chrome/Chromium is only for dev and testing on my machines) since the single largest browser vendor also depending on advertising for the vast majority of it's income is a massive conflict of interest frankly.
Firefox for Android has gotten really good and with ublock origin and the lack of constant "sync everything to google forever" nagging makes mobile web much more pleasant.
The thing with Firefox is that I know generally while they've made the odd misstep they do actually care about what they preach, I feel like Mozilla never had to say "Don't be evil" because there wasn't really any way they could be (on the scale of Google).
https://blog.mozilla.org/addons/2018/10/26/firefox-chrome-an...
“So I would like to publicly invite Google to collaborate with Mozilla and other browser vendors on manifest v3. It is an incredible opportunity to show that Chrome embodies Google’s philosophy to “focus on the user,”...”
Bullshit. They say that they are maintaining multiple chromeos patches to that driver that you don't have.
"Today we don't have a Linux GPU decode owner, so we have chosen not to activate something which we have no one to support and may have to break arbitrarily in the future for ChromeOS."
Is there anything to suggest that has changed?
Did someone offer to do it and were turned down?
I've seen patches but no offer to maintain the subsystem, etc that i can find. But maybe i missed it? I only did a few searches of the relevant mailing lists.
(Also, in my experience commenting on 3 year old closed bugs is not the way to make anything happen)
(For those wondering, the actual patch set appears to be tracked here: https://chromium-review.googlesource.com/c/chromium/src/+/53... )
Commenting on an HN thread about a 3-year-old closed bug is definitely not the way to make anything happen.
https://careers.google.com/jobs/results/5104577030586368-tec...
Graphics/GPU Software Engineer
https://careers.google.com/jobs/results/6469646855372800-gra...
Perhaps not, but this seems to be the only avenue available to many.
What would you suggest people who encounter a problem such as this do if they personally don't have the ability to personally fix the problem themselves?
This isn't an issue specific to chromium or even google. There are a lot of projects that seem to have an inside and an outside with those on the outside rightly-or-wrongly feeling unheard.
There's a difference between getting someone's attention, and then them disagreeing with you.
If you are trying to get attention, i'd go with the mailing list in this situation.
If they just disagree with you, sometimes, you gotta accept that someone doesn't agree with you about the product vision and let it go (this assumes your constraints - they don't have ability/time to fix themselves). They will either be right or wrong, and it will succeed or fail.
"This isn't an issue specific to chromium or even google. There are a lot of projects that seem to have an inside and an outside with those on the outside rightly-or-wrongly feeling unheard."
Sure, and i think companies/people should be very clear with governance models, participation models, etc. I also think they should be very clear about things like this.
Actually one of the hugest clashes i often see is open source projects that do have a product vision end up clashing with minorities of their user bases who don't fit in that product vision (either due to changes or never fitting in the first place), etc. Those clashes often seem much more emotional for people.
(I honestly don't know enough about chrome anymore at all to know whether that's the case. I'm only talking about the general situation you are describing).
A lot of the inside vs outside cases that i've seen fall into "product vision decided inside, you are welcome to participate if you want but we aren't changing it" style of project.
It's more rare (though i've seen it) to see "we have no vision of where we are trying to go but we still are going to semi-ignore random contributions/etc. We're gonna make it so you think we might care, but not actually try enough that anything happens "
Worse, a lot of the time the latter prevents other competing projects from existing because people spend time trying with the existing one then are too burned out to do anything else.
The product ones often at least give straight answers, even if people hate the answers.
Now if you had a patch vs Firefox that actually makes libva usable for everything, sure, go ahead, even add a checkbox in settings.
In general I find the category of folks who -- won't use Firefox, but want to pretend they're not basically just using Google Chrome -- to be very strange indeed.
That said, I agree that you're basically just using chrome.
I happen to use chromium for work because I do development for extensions and it's nice to keep chrome around as my primary test browser and chromium as my daily driver.
Plus I can dig through the source on chromium and recompile locally (admittedly, not super fun) to see if I might suggest additional extension features/apis.
Ok. And achieving those goals can be affected by people being unhappy about their decisions. Google can hire better people for lower salaries, the more they can convince the world that they are a force for good. And governments and the like (e.g the EU) can make rules and enforce them if they see behaviors that go strongly against the best interest of society as a whole. And so on.
I never understood the motivations of people who publicly defend corporations' short sighted decisions on the basis of "they are a business, they are only supposed to consider their bottom line, and aren't supposed to consider their public image in the process." Yes, corporations are allowed to make short sighted decisions, and people on the internet are allowed to complain when they do. (and yeah, I guess people are allowed to defend the corporations, but it simply encourages the companies to do things like this)
Obviously this isn't the worst thing that Google has done, but I support people calling them out on things like this.
And yes the OpenGL drivers are unstable and no, Khronos CTS passing doesnt mean anything wrt stability. The most trusted gfx platform is currently DX11 backend in libAngle.
Not to mention resources. System requirements for building on Linux:
" >>A 64-bit Intel machine with at least 8GB of RAM. More than 16GB is highly recommended.
>>At least 100GB of free disk space."
https://chromium.googlesource.com/chromium/src/+/master/docs...