Adrian Vovk
Adrian Vovk
OK as this is going in, we're hitting some design constraints of the `sysregistry.d/` proposal we agreed upon in our meeting. So here's some reasoning and an alternative design proposal....
we generally try to avoid copying over whole config files when we could get away with using drop-ins. Otherwise they'll drift over time. In other words, if you copy the...
Target logical DPI for laptops is something that's ultimately going to be DE-specific. Anyway, for Mutter and thus GNOME you can find the target values here: https://gitlab.gnome.org/GNOME/mutter/-/blob/422ee451/src/backends/meta-monitor.c#L1845-1867
I was researching this for the purposes of picking defaults in Mutter. TL;DR: Here's the target DPI values. https://gitlab.gnome.org/GNOME/mutter/-/blob/422ee451/src/backends/meta-monitor.c#L1845-1867. 135 is the target for mobile too. On screens over 20...
I don't know if this is a good idea security wise. It's entirely possible for a service to allow access to an interface, but now allow generalized access to all...
Instead of the system install scanning all the user installs, we'd instead have each user install to tell the system install that it's using the runtime (like pinning the runtime,...
> I would like to take on the task of dmverity support. I've been exploring a slightly alternate approach that would also be suitable for use in bootable distributions. Filesystem...
> IMHO, EROFS core on-disk format is simple and secure enough if you even compare with other writable fs, because basically it's just like an advanced archive format (which will...
Just to be super clear what I'm proposing: - We (I?) create a new "blobfs" in the kernel, which acts as a generic and flexible CAS filesystem. - Containers can...
> I don't agree with your point here: EROFS is absolutely designed and implemeneted to be robust against rogue images. Awesome to hear. I wasn't aware of this. > My...