kixik

well, for me there's no need for eye candy. I'm happy with sway and its tabbed mode.

BTW, labwc is sort of the openbox for wayland, in case interested

2

See tabbed mode on sway. Not all tiling compositors are about just tiling, :)

If you're not into tiling, but still want several of the advantages of sway, it offers a couple of additional modes, stacked and tabbed. I really loved tabbed setting some things to be floating. It's like it sounds, it offers a horizontal tab with all windows within per workspace, maximized below the tabs... Stacked is similar but it stacks the tabs vertically. If you'd tell me before a tiling compositor has such functionality I wouldn't have believed it. I like it better than stacking compositors, :)

1

kixik @lemmy.ml 1y ago

Both Tubular and Libretube support SponsorBlock, in that regard both are cool. I would use Libretube first, and if no piped site works then use Tubular, but Libretube itself allow for using local connection or it could be set to only work that way as NewPipe/Tubular do, so perhaps there wouldn't be a need for two apps.

The advantage to start with Libretube is a bit extra privacy by using a 3rd party piped site, so what youtube sees are requests from that site. But piped sites are more identifiable and therefore vulnerable to be rate limited or totally blocked. Though lately youtube has managed to block everything pretty close in time, Freetube, Pipeline, HarmonyMusic (a bit more resilient but also gets blocked), NewTube, Tubular, LibreTube, whatever.

Google is shortening the periods on which any of these frontends can be used to access youtube. Sadly, as Google offers some profit depending on the visits (more adds) then people seems not to be interested on leaving youtube, :(

uSo it doesn't seem to be a frontend which is less prompt to functionality disruptions. Some times it takes less time and some times it takes long, but usually most if not all frontends get affected. In that regard I don't see much difference, so that's why I recommended Libretube first and then Tubular as an alternative. Given there's not frontend supported by Google, in the end it's almost a matter of taste, with any Piped solution offering a bit extra privacy if used with 3rd party piped sites, and if that site is blocked or rate limited, another one can be chosen, or even it can try on the run to connect locally, or you can actually stop using a 3rd party site at all and only connect locally. BTW, freetube can be set to use an invidio site as well, but there are way less invidio sites up and running that there are piped sites. One can self host, but then the requests to youtube will be yours any ways, though you get local subscription and nice stuff on all frontends.

I like to have an alternative though in case at least for a little while one mechanism works but the other doesn't.

BTW, if interested on music, HarmonyMusic is a bit different frontend that can run on several platforms, gnu+linux, android and so on, and sometimes it keeps working while the other frontends don't.

kixik @lemmy.ml 1y ago

just now? :)

kixik @lemmy.ml 1y ago

I guess there was an attempt to move away from the email flow, to allow more people to contribute (I read that was part of the motivation), perhaps that made sourcehut (although it's in their plan, it hadn't become their highest priority) not an option, however both can be self hosted (that's what I would have expected from an organization as the Guix one, so that there's no dependency on a cloud service, as good as it might be), and both have really good TOS and are non profit. But cloud services are still something its users/clients do not really own. Perhaps as I understood, savannah will still be used as a mirror, but not just temporally, rather for good, so that if something happens on the cloud, there's plan B available... That's why for such big and important project I would have preferred a self hosted service. But oh well, I'm not part of the decision, and not an user yet, hopefully to become one later on when getting some minimal understanding of both guile and guix configuration (still guile but I believe simpler), because no matter the distro I always have to write and maintain a few packages myself. Hopefully at some point doesn't become never having the time to do so, hehe.

So all in all yes, the two best cloud options by far, but I'm surprised a Guix instance was not chosen, not sure if even considered.

3

kixik @lemmy.ml 1y ago

It would have been better to self host forgejo, rather than trusting a cloud git service using forgejo. But to be honest, its TOS, as well as the sourcehut's TOS which I even like it better, sound way better than GH's...

kixik @lemmy.ml 1y ago

Also a devclass post Mozilla quietly makes Microsoft’s GitHub the authoritative home for Firefox code suggests FF is making the GH repo the place to go as the source of truth for FF, :( This move to me is really sad, instead of moving to FLOSS alternatives it's preferring a proprietary with a terrible hosting licensing (gitlab one is much better for example, not sure about codeberg's one, but for sure is better as well), and what's worse, one that uses anything hosted in there for its own purposes, including feeding openAI stuff with FLOSS code violating any licenses and so forth. Which actually makes me strengthen the idea that mozilla is trending to go in the wrong direction making things worse on every step they follow.

I use a derivative, Librewolf, but in the end it depends on the FF code... Sadly, using GH is still like the norm, and I can change that. servo browser engine and verso (browser based on servo) are also hosted on GH. But at least they started there and migrating is always a hard decision, FF is just moving there having other options, so it means they don't care about GH mistreating users code...

kixik @lemmy.ml 1y ago

The battle is still there, and the GrapheneOS guy always bark at microG, like he really hates the whole concept of microG. What I have gotten from the discussion is that GrapheneOS is more secure, but although it sandboxes GPS denying some permissions, and some of those might be needed to be given away for some services any ways, it doesn't try to fake anything, which microG does. In that sense my preference has been microG, and I don't regret it.

That said, what you mentioned is true, both still access google app store, and still have to give some minimal information to google.

There's a 3rd option the OP didn't mentioned. If they are mainly interested in app store, and not the google services in general, there are a couple of somehow recognized 3rd party app store mirrors, which keep the same original signatures of the packages hosted by google app store, and they offer packages from other sources not provided by the google app store, in case interested on those packages: apkmirror and apkpure. From the two apkpure still allows to install and upgrade packages through FLOSS 3rd party apps like apkupdater, so that might be an option. For some months apkpure packages weren't able to be installed through apkupdater, but it seems that got corrected already.

But in general, the OP would benefit from always looking for FLOSS packages on the F-Droid repo, then other non official F-Droid repos which can be used through the F-Droid app, then see if they can be installed from their web site and updated without intevention of any installer, and then if there's no option but using proprietary software maybe looking for them on the apkpure/apkmirror sites or on apkpure through apkupdater or similar, and then aurora store, or if using grapheneOS finally google play if anything else fails, :)

I do understand the need for proprietary software, like bank OTP apps. It's sad banks, governments, medical services and so on never look for FLOSS software, they always require users to get proprietary software. I don't live in the EU, but I hope current hate/banning tendency ends up doing user a favor by starting to require banks, and the like to start using FLOSS apks, though doesn't really helps me, I hope in the end it helps people in the EU.

2

kixik @lemmy.ml 1y ago

This is a security feature. Other communication mechanisms having the keys somewhere else not owned by you is rather something I wouldn't stand. And to me it's unsafe that messages would be kept way long on the servers.

On xmpp the sync happens from server to all syncing clients, and the proper XEPs need to supported in both the client and the server.

1

kixik @lemmy.ml 1y ago

I'm curious about which programs if you can share. I write few bash scripts which used to call sudo, and I replace sudo with doas in those. And in case of muscular memory I also added a bash alias so that if by mistake calling sudo in reality I'd be calling doas. So far no issues. O course I don't use fancy args, and what I really needed from sudo I used to include it in /etc/sudoers and now on /etc/doas.conf, and I believe I couldn't include a couple of options but they were not critical since I've lived without them so far. And it's weird to find actual software that requires sudo, perhaps proprietary software. One can actually live without sudo and without doas, as long as there's still su.

Not judging, rather curious, actually I've met several guys who write scripts which would benefit from using sudo/doas, but they claim better call the scripts through sudo/doas rather than adding them as dependencies.

3

kixik @lemmy.ml 1y ago

A way smaller alternative therefore less prompt to vulnerabilities is OpenDoas found on Arch/Artix/... and other distros. From the GH project:

doas is a minimal replacement for the venerable sudo. It was initially written by Ted Unangst of the OpenBSD project to provide 95% of the features of sudo with a fraction of the codebase.

kixik @lemmy.ml 1y ago

Jami on desktop, not on phones yet:

https://jami.net/eleutheria

Available only for desktop users for now, the new Push-to-talk feature offers a new effortless way to communicate: simply press a button for hands-free, instant, and convenient audio messaging. It’s like in the olden days of gaming when gamers would key bind the Push to talk feature to be able to talk when necessary.

So jami all the way, 🙂

1

kixik @lemmy.ml 1y ago

You can simply decline by stating to an agent that you do not want your photo taken and want to opt out of a face scan. From there, a TSA agent should follow standard procedure of looking at your ID and your face to verify your identity.

kixik @lemmy.ml 1y ago

Are you sure?

https://guix.gnu.org/manual/en/html_node/Bootloader-Configuration.html