@BlackBloodRun they don't have a blanket policy, because blanket policy is a luxury only the truly rich can afford.
Blanket policy would require them to hunt down every violation(or lose the trademark protection completely based genericide).
What they've chosen to do instead is to hunt down every violation that causes big enough problems.
The closest example I can name of the top of my head about any open source project to blanket policy trademark enforcement policy is mozilla.

Which is why Debian has Iceweasle instead of Firefox

View this comment - View article - View full comments

That's an interesting take on the FOSS freedoms.

@BlackBloodRum freedom to copy, freedom to use, freedom to modify, freedom to read, but not freedom to imitate the original dev.
If they were to take this to court it would be a trademark conflict, which is something essentially all free software licenses maintain(the gpl is explicit. Others just don't touch the subject.)
This is often used to combat phishing campaigns, build trust and theoretically make money from open source development.

I can't find the "quote this post" button anymore.

View this comment - View article - View full comments

So despite being a slow moving company in a slow moving business, we will move to Linux over time.

@Eike
Being slow is actually what pulls many industries to Linux in the end.
Compared to most other major industry players Linux is good with things from the past.
Game consoles are often only temporarily with their most direct predecessor after which they will start dropping this compatibility as fast as they can.
Apple has a constantly shifting set of syscalls, even changes its architecture quite often and changes its notary policies at the drop of a hat.
Microsoft is slowly abandoning its DOS roots and tries to enforce its notary policies(although they fail hard).
Linux has an active community of compatibility layer developers for all systems known to man, has no notary policies by default(although distros tend to include them) and its leaders get refreshed at the speed that they burn out(mostly common in middle sized projects, because large sized projects become one or more full time jobs and small projects don't have enough pressure).
The only real competitor in this space to Linux is bsd with higher stability, but lower compatibility. Its maintainers burn out slower than those more in the Linux/GPL space, it also has only afterthought notary policies, but its full suite of software including compatibility layers is smaller.

View this comment - View article - View full comments

@tohur according to the github issue [External Link]:

• reboot the game


• copy the save data from the Remake prefix to the Rebirth prefix


• change to windowed, set resolution to 4K and change back to Full screen in order


• pin dynamic resolution to 50-50%

View this comment - View article - View full comments

Actually, developer world does turn in the Linux direction. Ever touched Docker and Kubernetes on Windows? No? Lucky you! :D

gcc and Java are much easier to work with in Linux too and Valgrind doesn't really work on Windows(source: am a student in a class with devoted Windows users).
Basically if you want something to go fast, directly interact with the hardware do something from the past or be compatible with something weird a Posix compatible OS(and mostly Linux) is "strongly suggested" in the dev world.
Some might say, "but WSL and Cygwin exists". Cygwin installing and using is too "Linux like" scaring away the Windows users and although WSL usage is pretty acceptable the setup is a special kind of punishment and all structural changes can force you to set it up again.
Still WSL was and is a breath of fresh air for Windows users.
The primary things keeping my Windows using classmates from using Linux are descending:

• Games


• Visual studio


• Market share


• WSL

View this comment - View article - View full comments

@F.Ultra
Something trying to imitate external(Windows) behavior still runs in the kernel.
Something will try to access an edge case and render the entire thing catatonic.

View this comment - View article - View full comments

I see little hope for it.
The primary problem with kernel involvement remains: crashes.
Wine can't help, but be a really crashy program.
In the kernel a crash means a reboot, while in user space it means a signal: making it recoverable.

View this comment - View article - View full comments

It seems that Google is outsourcing cost to the Linux foundation.

This is no cost outscourcing.
They lost a big antitrust fight and one of the damages that might get awarded is chrome getting broken off.
Once it gets broken off they would prefer to at least keep access to the source code.

View this comment - View article - View full comments

We're literally one CrowdStrike-style incident away from all these rootkits to completely crumble and probably cease to exist altogether in the aftermath. I don't care if this is fucked up to say, but I'm praying it actually happens.

Crowdstrike made a few days uncomfy for virus scanner devs that's all.
Kernel side virus scanners still exist and that's a good thing people fought for by the judge.
These issues will come and they will go, but structural change requires much more than stopping a tenth of the world in its tracks.
The incentives will have to change or vastly superior alternatives will have to be constructed.

View this comment - View article - View full comments

For years I didn't believe anti-cheat on Linux was possible, market share or not.

Today I do believe it, but not just for "tens of millions of users".
It would take years of development and innovation and a giant shift of mindset.
Most of the theory and hardware are there.
The software and market are still nowhere to be found.
MFC is still "advanced cryptographic theory".
Homeomorphic encryption is too slow.
Measured boot is pretty mature.
Retail specific secrets are known, but too sparse and under developed.
Tpms are currently underutilized.

Anti-cheat has always relied on security through obscurity and thus been subject to all the problems as stated in CWE-656.
This was never a strong defense and the more source available you kick in your stack the worse it gets, which is why PC is more sensitive to cheating than console and Linux is more sensitive to cheating than Windows.

If you want to rid yourself of that problem you've to rid yourself of CWE-656.
Some will say this is impossible.
To this I say two things: MFC and homeomorphic encryption could make this possible and we don't have to get fully rid of it just enough to get to Windows current level.

The reason the industry stayed there is in my eyes, because change is expensive and it's in its very nature to attract those to work on it who believe in CWE-656.
To believe in CWE-656 you've to be arrogant enough to believe that all your adversaries together won't afford to hire enough talent to approach yours.
On a professional level this means that you'll be convinced that the choices you make for your customers will be so vastly better than what even they can make(otherwise they could be that talent) that you could be willing to force these choices on those stupid customers who don't understand.
Anti-cheat is forcing your way of playing on others.
I understand that it's also a popular feature, but that doesn't change what it's and who it attracts.

View this comment - View article - View full comments