14:02:26 <alebastr[m]> #startmeeting Fedora Sway SIG (2022-08-24)
14:02:26 <zodbot> Meeting started Wed Aug 24 14:02:26 2022 UTC.
14:02:26 <zodbot> This meeting is logged and archived in a public location.
14:02:26 <zodbot> The chair is alebastr[m]. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
14:02:26 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:02:26 <zodbot> The meeting name has been set to 'fedora_sway_sig_(2022-08-24)'
14:02:33 <alebastr[m]> .hello alebastr
14:02:34 <zodbot> alebastr[m]: alebastr 'Aleksei Bavshin' <alebastr89@gmail.com>
14:02:39 <Fale[m]> .hello fale
14:02:40 <zodbot> Fale[m]: fale 'Fabio Alessandro Locati' <me@fale.io>
14:02:58 <jkonecny[m]> .hi
14:02:59 <zodbot> jkonecny[m]: Sorry, but user 'jkonecny [m]' does not exist
14:03:20 <jkonecny[m]> .hello jkonecny
14:03:21 <zodbot> jkonecny[m]: jkonecny 'Jiří Konečný' <jkonecny@redhat.com>
14:05:06 <alebastr[m]> #topic Name for the ostree spin, Ep. 02
14:06:00 <alebastr[m]> so... Fedora Atomic with Sway, anyone?
14:06:38 <Fale[m]> based on the feedback on https://fale.io/blog/2022/08/12/fedora-sway-ostree-spin-name and the linked reddit conversations, I would say many people are not happy with the minerals based way of naming spins, though I'm not sure this is something we can override or if we have to stick to this nameconvetion
14:06:43 <jkonecny[m]> +1 for the new naming variant. Altought we don't know the exact naminh
14:06:47 <jkonecny[m]> Naming
14:07:25 <Fale[m]> Sodalite is already used by an unofficial phanteon-based Fedora spin, so we should probably avoid it
14:07:49 <alebastr[m]> yep. the message is that people want to see an unambiguous reference to Sway and to the immutable desktop
14:07:58 <jkonecny[m]> I mean the Fedora SilverBlue with Sway
14:08:59 <jkonecny[m]> Or better Fedora Ostree with Sway?
14:09:38 <alebastr[m]> I suspect that less people know what ostree is, compared to silverblue.
14:10:05 <Fale[m]> the part that I don't like about this, is that our ostree version is waiting for a name, and I see this as a way broader conversation (someone is also putting into discussion "SilverBlue" itself) and I have the feeling that we will not have a clear naming convention for a few months, at this point
14:10:59 <jkonecny[m]> Yes but people are take SilverBlue == Gnome
14:11:10 <jkonecny[m]> But that might be less of a problem
14:11:21 <jkonecny[m]> So we can go with the SilverBlue
14:11:23 <alebastr[m]> people are split on that part :(
14:12:51 <alebastr[m]> the more important question, can we use SB in this context? do we need to work with council/design/branding (do we even have a branding team?) to ask to use the name?
14:13:17 <jkonecny[m]> Fale: we don't need to follow the current solution and come with something new
14:13:29 <jkonecny[m]> Can be changed later
14:13:44 <jkonecny[m]> If the name will be self describing
14:14:33 <jkonecny[m]> alebastr[m]: We can ask Matthew about this or just ask on discussion
14:15:44 * alebastr[m] is looking at the struggles of Ubuntu Sway developer with getting Canonical's permission to use the name :)
14:17:29 <jkonecny[m]> I don't expect that to be problem here
14:17:48 <jkonecny[m]> But definitely we should ask first
14:20:04 <pxl_sg[m]> The spin is only going to be immutable? Or also the “normal” version?
14:20:18 <alebastr[m]> both
14:21:02 <pxl_sg[m]> Ok, so the name problem will arise for both version:
14:21:02 <pxl_sg[m]> - fedora sb (or whathever) with sway
14:21:02 <pxl_sg[m]> - fedora workstation with sway
14:21:44 <alebastr[m]> yep. but in this case things look very clear for the classic variant - `Fedora [Linux] with Sway`
14:22:34 <alebastr[m]> I don't think there's a clarity whether the Linux part is necessary in the policy discussion thread... wait, does it mean that things are not clear?
14:24:38 <alebastr[m]> ok. let's not block on that.
14:25:13 <alebastr[m]> are we fine with Fedora Something with Sway, or should we talk about trees and rocks? even if it would be a backup plan
14:26:48 <jkonecny[m]> I would go with Fedora with Sway and Fedora SilverBlue with Sway
14:27:00 <Fale[m]> +1
14:27:23 <enfermerocrypto[> Fedoras’ s way
14:27:25 <jkonecny[m]> It's simplest and we csn change that later
14:28:18 <jkonecny[m]> I can ask Matthew about the SB in the name
14:28:26 <enfermerocrypto[> Your way or the highway. Choose Fedora’ s way.
14:28:57 <alebastr[m]> jkonecny: that would be appreciated, yes
14:29:25 <jkonecny[m]> Sorry but I have only 14% of battery with no charging option😬
14:29:32 <alebastr[m]> someone just asked on discussion about a single brand for immutable os :)
14:29:50 <alebastr[m]> ok. let's not waste your 14% of battery :)
14:31:57 <alebastr[m]> #accepted Use Fedora with Sway/Fedora Silverblue with Sway (with the necessary approval) (+3)
14:32:10 <alebastr[m]> (one voice is from me)
14:33:17 <alebastr[m]> #topic initial environment and VM compatibility variables
14:34:23 <alebastr[m]> jkonecny: you mentioned that you like the approach with the wrapper script better, right?
14:34:42 <alebastr[m]> as opposed to the /etc/profile.d
14:36:04 <alebastr[m]> Fale: any opinion on that?
14:36:43 <Fale[m]> wrapper script to sway itself?
14:37:34 <alebastr[m]> so we're talking about the 2 options here: https://gitlab.com/fedora/sigs/sway/fedora-sway-configs/-/merge_requests
14:38:15 <alebastr[m]> we have to pick one because Sway is very much broken on VMs and we need to inject environment variables as a workaround
14:38:19 <jkonecny[m]> Honestly I'm more inclined to the profile.d seems to me as easier to be discovered
14:38:38 <Fale[m]> I'm currently using the script wrap, but I see it as a workaround
14:39:28 <alebastr[m]> I don't think anyone disagrees that this is a workaround :)
14:40:00 <alebastr[m]> but it's also an only interface provided for some tunables (WLR_RENDERER/WLR_NO_HARDWARE_CURSORS)
14:40:11 <alebastr[m]> s/an/the/, s/WLR_RENDERER/WLR\_RENDERER/, s/WLR_NO_HARDWARE_CURSORS/WLR\_NO\_HARDWARE\_CURSORS/
14:41:15 <Fale[m]> I think the wrapper scipt can be found fairly easily, but surely is not the most standard way of doing things
14:42:11 <pxl_sg[m]> How does the greeter start sway? Is it thru a script?
14:44:49 <alebastr[m]> we can use a script there. in fact, that would solve some of my sddm environment woes, but eh, another script...
14:45:26 <alebastr[m]> https://gitlab.com/fedora/sigs/sway/fedora-sway-configs/-/blob/fedora/sddm/wayland-sway.conf#L10
14:46:00 <Fale[m]> I'm +1 on the script
14:46:00 <jkonecny[m]> I don't have so strong opinion we can go with the wrapper script
14:46:01 <Fale[m]> at lesat in the short term
14:46:12 <alebastr[m]> (and https://gitlab.com/alebastr/qmlgreet/-/blob/main/data/sway.config#L4 for the greetd option, which also may use a script)
14:46:55 <alebastr[m]> okay
14:47:40 <alebastr[m]> #action review and merge https://gitlab.com/fedora/sigs/sway/fedora-sway-configs/-/merge_requests/1, use wrapper scripts for DM greeters
14:49:10 <alebastr[m]> and I'll make a preview of a greetd-based login option once we have ostree publish for merge requests set up
14:50:05 <alebastr[m]> #topic Software
14:51:33 <alebastr[m]> I want to get rid of dmenu in the default config. That is long overdue and some decisions should be made :)
14:52:26 <alebastr[m]> I do hope that people went through the list [here](https://gitlab.com/fedora/sigs/sway/SIG/-/issues/2#note_1026003944) and made some opinions
14:53:14 <Fale[m]> I'm +1 on the drop of dmenu
14:56:19 <Fale[m]> I'm not a fan of rofi/wofi UX, but it probably makes sense for majority of users
14:57:36 <Fale[m]> at the end of the day, imho, rofi is the best option at the moment, since it suppots all workflows
14:58:14 <alebastr[m]> Fale: does `rofi -theme dmenu` look too different from the real thing for you? :)
14:58:53 <Fale[m]> I'm on bemenu, but I get your point ;-) rofi is very flexible
14:59:50 <alebastr[m]> ok. so I'll have to move the rofi-wayland fork packages from my copr
15:00:12 <Fale[m]> @alebastr#2053 if you want a package review, ping me
15:00:13 <jkonecny[m]> +1 for Rofi
15:00:28 <alebastr[m]> which is going to be fun, esp. with all these `Conflicts:`
15:00:32 <jkonecny[m]> I'm using it with no issues
15:00:37 <pxl_sg[m]> how about wofi?
15:00:46 <pxl_sg[m]> wayland native
15:01:09 <alebastr[m]> Alpine maintainer just did a package with binary that uses themes and other noarch content from the upstream rofi package
15:02:18 <alebastr[m]> pxl_sg: other than being not quite alive, wofi also lacks a few important features
15:02:49 <alebastr[m]> (and it's slow, but that's my subjective opinion not backed by benchmarks)
15:02:57 <Fale[m]> @pxl_sg rofi has been around for more time and is way more active upstream (my2cc)
15:04:02 <pxl_sg[m]> got it
15:05:51 <alebastr[m]> #action alebastr to file a package review bug for rofi-wayland and switch the default config once that's done
15:07:03 <alebastr[m]> ...but f37 is frozen and we will have to get the package from updates-testing until the beta freeze ends
15:08:02 <Fale[m]> @alebastr#2053 oren't we aiming for rawhide (ie: F38)?
15:09:32 <alebastr[m]> Fale: yes, but we do test builds for other branches
15:10:09 <alebastr[m]> because while rawhide is stable and definitely not broken, using it as a daily driver may be too much
15:12:30 <alebastr[m]> #topic Open floor and general progress discussion
15:12:35 <StefanoFiguraret> <alebastr[m]> "(and it's slow, but that's my..." <- I agree, I like it but it is slow AF
15:13:07 <jkonecny[m]> 😃
15:13:19 <alebastr[m]> we're over time, so foot vs. alacritty gets deferred again :)
15:13:44 * alebastr[m] still needs to check if alacritty works on RPi4 with the latest kernel/mesa
15:14:36 <StefanoFiguraret> Am I wrong or foot is now part of the base SB tree?
15:15:19 <alebastr[m]> it is a part of our tree right now, as it was in the sway defaults
15:15:31 <StefanoFiguraret> Ah I see
15:16:05 <StefanoFiguraret> For some reason I thought I had seen it on my vanilla SB
15:16:14 <StefanoFiguraret> I need to double check
15:16:28 <pxl_sg[m]> I start to like foot actually. I don't need the complexity of alacritty at all and foot is lightweight enough for me
15:16:48 <StefanoFiguraret> Does it have vi edit mode like alacritty?
15:17:59 <pxl_sg[m]> I doubt
15:18:01 <StefanoFiguraret> s/edit/navigation/
15:21:21 <alebastr[m]> seems like we don't have other important things to discuss :)
15:22:29 <alebastr[m]> just wanted to thank jkonecny and anthr76 for getting classic live/installer iso builds working. now we have something to play with that does not involve rpm-ostree rebase
15:23:26 <alebastr[m]> #endmeeting