💾 Archived View for gemini.ctrl-c.club › ~phoebos › logs › kisslinux-2023-02-19.txt captured on 2023-03-20 at 19:42:49.
-=-=-=-=-=-=-
[2023-02-19T00:19:24Z] <niceguy5000[m]> What would happen if we all use the tor? will it be impossible to track everyone? [2023-02-19T00:20:39Z] <niceguy5000[m]> or is it pseudo anonymity? [2023-02-19T01:13:45Z] <niceguy5000[m]> guys I'm switching to the framebuffer full time like aligrudi. [2023-02-19T01:14:19Z] <shokara> nice [2023-02-19T01:14:34Z] <shokara> the framebuffer is comfy, at least when not needing multimonitor [2023-02-19T01:25:55Z] <niceguy5000[m]> yeah it is. I'm going to use fbvnc with qemu to run firefox and other heavy apps. [2023-02-19T01:29:14Z] <shokara> you might as well just use x instead of x through vnc then... [2023-02-19T01:58:17Z] <niceguy5000[m]> I want something super suckless so framebuffer is the way to go. [2023-02-19T01:58:54Z] <niceguy5000[m]> I can use netcat to pipe to the framebuffer directly to watch videos [2023-02-19T01:59:36Z] <niceguy5000[m]> ffplay plays video on the vm and I use netcat to pipe to host framebuffer. [2023-02-19T02:07:25Z] <shokara> fbvnc sounds to me like the opposite of suckless :/ [2023-02-19T02:21:53Z] <niceguy5000[m]> I can run a full bloated electrons apps and not effect my system in anyway if I had too(but won't) using fbvnc and qemu. Only if their was a suckless vm emulator that's full featured to remove qemu but none exist. [2023-02-19T02:22:16Z] <niceguy5000[m]> and still be on the framebuffer all day. [2023-02-19T02:23:01Z] <shokara> I think you should try not using the bloated programs that don't run on the framebuffer. [2023-02-19T02:23:04Z] <davidgarland> is that really less bloated / easier to set up than a chroot? I haven't gotten the latter working (..for lack of really trying, to be fair) but I imagine that'd be simpler [2023-02-19T02:24:12Z] <shokara> The way I see it is that you'll still be running a full Xorg installation in the VM that you'll just be accessing from the framebuffer. [2023-02-19T02:24:20Z] <shokara> (or wayland) [2023-02-19T02:24:48Z] <davidgarland> yea [2023-02-19T02:25:33Z] <shokara> so you'd be better off not having to rely on this tedious workaround or just running x directly when you need to and exiting once you're done using the gui program :/ [2023-02-19T02:25:57Z] <davidgarland> anyhow im gonna hop off for now, friend is asking me to play vidya and those are on windows pc still so.. seeya tomorrow most likely o/ [2023-02-19T02:26:59Z] <shokara> also for a "suckless" vm program, have you tried using bochs instead of qemu? [2023-02-19T02:27:22Z] <shokara> I don't think it's actually suckless, but it should still be much simpler than qemu [2023-02-19T02:51:47Z] <niceguy5000[m]> <shokara> "also for a "suckless" vm program..." <- no but I build Qemu statically, so it's only one binary file. [2023-02-19T02:52:54Z] <niceguy5000[m]> I have a full static system. [2023-02-19T02:54:47Z] <shokara> "Only if their was a suckless vm emulator that's full featured to remove qemu but none exist" I was referring to this btw [2023-02-19T02:55:14Z] <shokara> for x86, it seems to do most of what qemu can. [2023-02-19T02:55:24Z] <niceguy5000[m]> can it use kvm? [2023-02-19T02:56:24Z] <shokara> I don't think so [2023-02-19T02:58:31Z] <niceguy5000[m]> KVM is need for hardware acceleration. [2023-02-19T02:58:42Z] <shokara> yes [2023-02-19T03:00:29Z] <shokara> for a simple featured alternative to qemu, bochs seems to be one of the best [2023-02-19T03:00:43Z] <sewn> Hi [2023-02-19T03:01:11Z] <shokara> otherwise there's virtualbox, but that's basically proprietary with the extensions needed for usb3 and other stuff [2023-02-19T03:06:44Z] <niceguy5000[m]> <shokara> "for a simple featured alternativ..." <- I also use qemu for security reasons. [2023-02-19T03:07:54Z] <niceguy5000[m]> Redhat is on top of qemu so it's more secure I think. also I use scripts instead of gui or virsh to run vms. I know the whole documentations of qemu arguments. [2023-02-19T03:08:10Z] <niceguy5000[m]> I might make them mircovms using qemu too. it's much smaller to build. [2023-02-19T03:08:30Z] <niceguy5000[m]> bootup less than a second. [2023-02-19T04:33:39Z] <testuser[m]123> Hi [2023-02-19T12:22:52Z] <sewn> should -f or rm/ln force flags be used in kiss build manifests? [2023-02-19T14:33:18Z] <midfavila> niceguy5000[m]: you need to weigh the size and scope of the codebase against the complexity of its tech stack and the number and competency of its maintainers in order to effectively gauge how secure something is [2023-02-19T14:34:18Z] <midfavila> qemu is massive and has piles of features, so despite its large maintenance team and their high competency, it's unlikely to be as secure as it could be. of course, if it's good enough for qubes, it's probably good enough for someone who just wants to spin up some VMs at home to mess with... [2023-02-19T15:15:02Z] <sewn> i shall name my kiss fork piss, for piss it simple stupid [2023-02-19T15:18:14Z] <midfavila> already done [2023-02-19T15:18:16Z] <midfavila> python kiss [2023-02-19T15:18:23Z] * midfavila makes a buzzer noise [2023-02-19T15:18:58Z] <sewn> what about ciss [2023-02-19T16:55:31Z] <phinxy> herro [2023-02-19T16:56:01Z] <sewn> jelheo [2023-02-19T17:04:03Z] <midfavila> really wish C had block structure jfc [2023-02-19T17:04:22Z] <midfavila> i'm too used to programming in scheme and having that as an option for program organization [2023-02-19T17:14:01Z] <testuser[m]123> ioraff: illiliti: what do you think of swapping "out of scope" stuff from busy box for full featured alternatives? Eg modprobe -> kmod, ip -> iproute2 [2023-02-19T17:14:23Z] <midfavila> >he uses ip [2023-02-19T17:41:45Z] <illiliti> i think we should keep busybox as is [2023-02-19T17:43:24Z] <illiliti> actually i think provides system would help here [2023-02-19T17:44:09Z] <illiliti> just make busybox "provide" those components, so user can switch them effortlessly [2023-02-19T17:44:35Z] <illiliti> that would be much better tbh [2023-02-19T17:44:53Z] <midfavila> mfw [2023-02-19T19:09:22Z] <midfavila> so after a few minutes of thought last night before bed and a half hour or so of half-hearted debugging i found out why i was getting redirected to onionsites from riverside when using apport [2023-02-19T19:10:05Z] <midfavila> as i thought it was due to my lazy implementation of a response key-value extraction routine matching falsely against one of the responses. didn't realize that riverside in particular provided Onion-Location: keys [2023-02-19T19:10:40Z] <midfavila> so when apport looked up "ocation", it matched that instead of "Location" or "location", which is the original reason why i didn't bother implementing proper lookups - didn't want to deal with mixed case [2023-02-19T19:45:31Z] <midfavila> https://inv.riverside.rocks/watch?v=uKo_IJDCO9c&listen=false