qyliss changed the topic of #spectrum to: A compartmentalized operating system | https://spectrum-os.org/ | Logs: https://logs.nix.samueldr.com/spectrum/
xantoz has joined #spectrum
pie_ has quit [Remote host closed the connection]
<xantoz> just saw the part about wayland on the web page. have you looked into the cage wayland compositor? it runs a single application in its own wayland compositor, somewhat similar to xpra. it could probably be used as a building block for isolated wayland applications
<xantoz> or really any wayland compositor used with its wayland backend on top of a "master" compositor could probably be used for isolating
<xantoz> (set-frame-font "-1ASC-Liberation Mono-normal-normal-normal-*-13-*-*-*-m-0-iso10646-1")
<xantoz> speaking of wayland and clipboards, I pasted the wrong clipboard :D
<adisbladis> xantoz: Cage looks super useful :)
pie_ has joined #spectrum
hyperfekt_ has joined #spectrum
<hyperfekt_> Mind that the work around crosvm also includes VirtIO Wayland, meaning the ability for guest clients to use the host compositor via shared buffers.
hyperfekt_ has quit [Remote host closed the connection]
pie_ has quit [Ping timeout: 252 seconds]
<adisbladis> hyperfekt_: Yeah they are probably orthogonal, cool regardless :)
pie_ has joined #spectrum
pie_ has quit [Ping timeout: 248 seconds]
hyperfekt_ has joined #spectrum
<hyperfekt_> adisbladis: Definitely, I just wanted to point out that an implementation using Wayland would not be hindered by the transport if using nested compositors could work out without too much effort. But I think if either Xpra or Wayland turn out much easier to implement than the other it'd be advisable to choose that (it's conceivable at least that we
<hyperfekt_> might be able to just take the Qubes programs for things like Clipboard sharing, which are built for X).
hyperfekt_ has quit [Ping timeout: 260 seconds]
pie_ has joined #spectrum
puck has joined #spectrum
pie_ has quit [Ping timeout: 245 seconds]
pie_ has joined #spectrum
<qyliss> Not been around much on IRC in the past couple of days, but will have a status update on Spectrum tomorrow, along with a summary of what we learned at CCCamp.
<qyliss> Tomorrow is going to be writing all day. That update, along with my project plan for NLnet.
<hyperfekt> qyliss: Awesome, looking forward!
<qyliss> At some point I should update the website too... It doesn't really reflect what Spectrum is going to be anymore.
<hyperfekt> qyliss: What are the main points that have changed?
<qyliss> Focus on crosvm instead of containers.
<qyliss> Possible Wayland instead of Xpra, pending research.
<qyliss> Focus moving from being "almost as good as Qubes, but usable for more people" to actually having net security benefit over Qubes, while still being able to run on way more hardware (because of new information that makes this seem feasible)
<qyliss> Possibly using some other means than the file system for sharing state.
<FireFly> hm, sounds interesting, looking forward to reading in more detail about what you learned
<hyperfekt> Security benefit over Qubes? Sharing state via different means? Sounds interesting, also wanna read more!
<qyliss> tomorrow :)
<hyperfekt> :D
<tazjin> Would a session with some
<tazjin> people from the ChromeOS team be useful?
<tazjin> (thinking specifically people who worked near crosvm)
<hyperfekt> I'm getting way too hyped for this :o
pie_ has quit [Ping timeout: 245 seconds]