I would like to introduce people to bootc or Bootable Containers, a new project that looks poised to take the Linux world by storm. RedHat sure thinks so, but I do too.
Essentially all it entails is putting the Linux kernel into a Docker container. But the trick is that it is bootable. You can use a dockerfile to create your own custom OS image and deploy it to your personal machines, at work, or share it as your own custom "distro"
I'll go over some history of ChromeOS and atomic updates, A/B boot, ostree (Container Linux and flatcar/coreOS). And how we got where we are and how I believe that container tooling is much more abundant and accessible than similar systems like Nix.
Take a quick look at the future where companies and individuals can will be able to verify all of the code running on their machine via container signing, hermetic builds, composefs, UKI.
I'd like to clear up some misconceptions about "immutable" OSes. Show off how and why using an OS with guard rails encourages good/smarter behavior and doesn't have to be restrictive.
I'll demo how it works and how I manage my own personal infrastructure via Bootable Containers
Linux in the Cloud dominates, but falls way behind on Desktop. This may change that.
Containers are fun, cool and weird
This is the year of the Linux Desktop
There is promise of technical depth, in terms of coverage of various techniques, in this talk. The history bits should be a good background setter. "Containers are fun, cool and weird" is possibly true in the many ways they can be used.... Suggesting that bootable containers can fix the linux desktop seems far fetched but does the trick as a controversial title :)
I personally find this very interesting technically. There are a lot of conversations around using nix or nix + docker for better deployments in various scenarios. This seems like an interesting effort and I think the audience will also be curious about it.
I think this talk will create interest.
This talk covers a relatively new concept that might be of interest to folks. I'd vote to accept it.
Interesting talk. Much needed for people working in systems and FOSS.