Elementary OS 6.1
blog.elementary.io
blog.elementary.io
You heard me right: The official explanation on how to upgrade to Elementary OS 6 is to back up your stuff and do a full reinstall.
Considering that Elementary OS is somewhat marketed as being ideal for, say, elderly folks or people less experienced with their PC, as something that a Linux user might install for non-techie friends, this is beyond unacceptable and also shows a certain lack of technical competence for the folks running it.
Edit: Also, I've made the tragic mistake of installing previous Elementary OS versions for non-techie friends I rarely see. Not anymore! What am I supposed to do Elementary? Inform my friends I've got to basically rebuild their computer the next time I see them at Thanksgiving to keep them safe? I'm trying to avoid using profanity...
Options from $200-$1000+, with good options around $400 on sale.
Issues? Power wash it.
Really broken? Go buy another one.
Lost or stolen? Buy another one and log in.
Need office? Google’s products work well. Need more? O365 subscription.
Want more security? Two factor auth with a Yubikey.
I recognize that this relies on Google and Microsoft and is not FOSS and has privacy issues and everything else. But I do not have enough time to spend managing my extended family’s IT beyond giving them chromebooks that achieve 99% of what they need to do with very little input on my part. If they want to invest time in learning about Linux and get away from Google and MSFT, that’s fine, but they can do it on their own time with a test machine and keep Chrome OS as their daily driver until they are confident on their own.
https://support.google.com/chromebook/answer/9367166?hl=en
https://support.google.com/chrome/a/answer/6220366?#zippy=%2...
Most Intel-based ChromeOS devices from before 2017 are still easily recyclable into capable machines if you can flash the firmware and install Linux. I use several for for DNS/DHCP/media/file servers and test machines. Some of them have batteries that outlast my newest laptops, so a UPS is unnecessary.
It stinks that recycling unsupported devices is getting more difficult, but it's happening across the board regardless of platform. I'd still recommend a ChromeOS device to nontechnical users for a worry-free experience.
Naturally they didn't want to hurt my feelings and I only discovered their undo action several months later.
For them Windows just works, and there are plenty of shops around that can go to in case it doesn't, which wasn't the case with anything Linux.
He had bought a second hand machine from a school or something and the sysadmin there refused pirate licenses and pre-installed Ubuntu for them.
When the disk broke (or something, it wasn't software related IIRC) he asked me to get back the brown system because he liked it.
And that my intro to Ubuntu. I had used Linux for a couple of years already but Gnome before Ubuntu wasn't something I used voluntarily so it was a reql surprise to me that someone had made Gnome great.
Anyways, since then I have said that Linux is for people who don't know about computers - and for people who know what they are doing.
The people it doesn't seem to work for is people who have used Windows or Mac before and have picked up a few tricks and refuse to learn anything new.
But that was a home build Frankenstein desktop that eventually had a part fail. By then, chromebooks were available for $200. I got one and they used it until the wheels fell off (screen issues, battery life <10 minutes, charger would only work in certain location) two years ago when I got a free chromebook through a promotion to replace it.
I think we got one in 2012, 2015, 2019 (to replace the 2013), and 2021 (to replace the 2015). They are laptops and used daily.
Worth checking.
Other chromebooks with comparable hardware are still getting updates. There’s nothing wrong with the hardware. It’s still snappy to use. Battery life is still north of 4 hours. Just no more updates suddenly from Google.
https://reclaimthenet.org/google-docs-policy-is-updated-hate...
I submitted a bug report and essentially got a DM from someeone that it was still on their todo list, and had been that way for a while. That was the last time I used them.
For easy to use, I really miss CrunchBang Linux.
BunsenLabs (very conservative, basically just updated CB). CrunchBang++ (modernised a little, with some new features and shiny).
AIUI:
The BunsenLabs community are aggrieved because they respected Corenomial's rights over it and so didn't use the original name.
The CB++ people are bolder: they asked him, got permission, and were a bit more radical.
TBH I wish they would either settle their differences and re-unify, or the BunsenLabs folks be a little more radical and (for instance) move to Devuan instead.
There's a name i haven't heard in a while :) Look up Archcraft, the OpenBox variant might scratch the same itch as CBL.
This is the issue with Linux Desktop - every time you want to try something new you need to effectively reinvent the wheel for yourself. Most things end up being the same though, my browser is my browser, my data is my data. I can compartmentalize some of that to make it portable between distributions but I feel like that is solving the problem from the wrong direction.
At some point I would hope this becomes a DE like Gnome or KDE. Inventing your own programming language, app marketplace etc... is admirable but seems like it is being done in vein.
They've been sponsoring GNOME, which has been pushing the simplified, mobile-friendly look for almost a decade now. They've also (I believe) been behind the scenes in GNOME's push to kill theming on the behalf of developers who hate their apps getting themed, which of course distributions hate because that hurts the ability to stand out. If GNOME looks the same on every distribution, your distribution sticks out less.
Red Hat has also been behind sponsoring SystemD, which for better or worse brought every distribution kicking and screaming into a more consistent "low level" layout. Again though, this makes management across Linux systems somewhat easier, while making distribution-specific differences smaller.
Red Hat has been behind Flatpak, which compartmentalizes applications and means that they can run on any distribution when built against a common runtime. For app developers and new people, this is great as a "Linux app" just runs on any "Linux" distribution, easy. For old-timers and technical folk (even on here)... again it's with kicking and screaming. And if you are a distribution author, every distribution now runs the same apps, so what's so special about your distro again?
Red Hat has also been behind PulseAudio and PipeWire, which changed how audio worked on Linux... again with kicking and screaming.
The point is, Red Hat keeps working on technology that, I would argue, makes Linux more consistent and approachable to newcomers, but massively changes how it works which angers old-timers and also happens to reduce diversity within distributions as to how things are done. This makes things easier for programmers, but erodes away the proliferation of distributions because the differences between them get smaller and smaller.
Because, if Red Hat gets their way, consider: Every desktop that uses GNOME looks the same because there isn't as much theming. Every desktop runs the same apps using Flatpak. Every desktop has similar low-level internals with SystemD. If you are a distribution maker, not much left to stand out now, is there? Again, app developers love this, but if you are a distribution maker, Red Hat is trying to kill your edge.
I never would have noticed the change except that detecting and setting audio profiles/codecs on my bluetooth devices from my sound control GUIs works much better than before!
It seems to be damn good.
It is. PipeWire is good. PulseAudio was bad. End of the story for me.
PulseAudio's hate phase was caused by it breaking everything. If PipeWire had been introduced and worked, out of the box, correctly, then I doubt that it would have been poorly received.
I have no way to evaluate the claim, and don't really care what the real story is.
PA mostly works, now, and PW mostly works, now. But PW does have bugs, and is, inexplicably, coded in C, for which there can be no excuse at this late date. It could be switched over to build in C++ in probably under a day, and then begin to be modernized, anytime, as was done with Gcc and Gdb with unalloyed success.
PW is written by the main guy from Gstreamer, which used to crash all the damn time back when PA went silent a lot. But he is older now and has lots of help.
Most of the biggest communities are around distributions that have releases every 6 months - 1 year and proudly proclaim all the things that they have improved and changed.
It's an entirely bad faith argument.
Audio prior to pulseaudio on the Linux desktop was hot garbage primarily for lack of hardware support for various built into the motherboard sound chips. The only viable solution was to buy supported sound cards which of course was only easy on desktops. With the launch of pulse even those machines experienced curious problems with a crashy, confusing, poorly engineered face to the already bad audio situation on desktop Linux. It ended up with both the credit for its own bad engineering and for the underlying shoddiness of the substrate it operated on by being the thing that visibly failed to produce sound when a user clicked on a video. The cases wherein audio was magically fixed by disabling it using alsa alone created loud and persistent critics whose criticism persisted even when both pulse and the underlying audio problems decreased.
This is indeed how things work. If I put syrup and lots of salt in a glass and float a radish in it and proudly plaster my brand across the cup I shall never sell you another cup no matter how much I improve the formula for my "soda".
Labeling people who came by their criticism honestly thoughtless Luddites doesn't help. You might notice that most products don't go through a "hate phase" and new software often does. It's not just because people don't like change its because as a species we are remarkably bad at making software and new software is often garbage unfit for purpose. If much software were a toaster you should have girded your foot with a steel toed boot and kicked a field goal out your back door with the disgraceful junk you were tricked into buying.
I don't hate change; in the case of PipeWire, I absolutely welcomed it. The alternative was obviously broken and couldn't be iterated on in any meaningful way, so replacing it with a higher-level interface made all the sense in the world. I loved a lot of the iterations GTK3 made over GTK2 so the interfaces would be equally at home on a tablet as they are on a desktop. I appreciated the work that went into writing GUI package managers even though I'd never really use them, and while I don't agree with them, I'm happy that people are experimenting with containerized packaging.
Linux users don't hate change, we just hate regression. A lot of people's workflows rely on exploiting extensibility and niche features, much as others do on Windows and MacOS. When people pull the plug on old systems, or limit the interoperability of their application, they shouldn't wonder why their reception is negative; people want to take your project to the next level, all you have to do is let them. PipeWire was beloved because it did let people extend pre-existing features. It was a powerful tool for managing audio cart-blanche, and it released in a fairly feature-complete state. It's an example of the perfect Linux software overhaul in my opinion.
There's a lot of generalizations in the grandparent comment about how distributions 'hate' the lack of theming and whatnot, and all I can say is that those kinds of opinions are almost entirely founded on a premise of not understanding the majority of Linux users. Extensibility is king, and it makes a lot more sense for them to build a flexible toolkit that allows for both theming and accessibility options instead of locking it into a one-track design philosophy that just-so-happens to benefit a small handful of users. That's what people are mad about.
The 'stop theming my app' crowd is pretty small. The number of apps on Flatpak is pretty small compared to say Snap and tiny compared to 'traditional' package managers on distributions.
You don't have to worry how each distro handles booting or services - it's SystemD.
You won't have to worry how each distro handles audio with Jack, PulseAudio, or ALSA - it's PipeWire.
You don't have to worry about the libraries the distro provides for your app - just use Flatpak.
You don't have to worry where to host your app for every distro - just use Flathub.
You don't have to worry what toolkit to use - just use GTK. You can use Qt if you really care, but GTK is clearly the "one true road" (if that analogy makes sense).
And so on.
You can with minimal effort theme both similarly by installing a theme and icon theme that has versions for both gtk qt but your gnome desktop wont provide a built in gui to configure such, again their choice, their limitation not a limitation of QT. A given app QT app isn't going to pull in a particular theme because that would be the tail wagging the dog.
kthx baiiiii
Citation needed, and number of users is more important than number of apps. Snaps are pushed strongly by Canonical and they actually pay people to port apps to their store, which is pretty much used by Ubuntu users only, while every other distro uses Flatpak.
Snaps are not gonna last long, like many other Canonical-only efforts. Nobody is interested in adopting them, for good reasons. Flatpak, with their shortcomings, are more flexible, actively developed by multiple entities and adopted by more people.
While the raw number of apps on Flatpak may be less, it's more often the apps people care about.
I say this as someone who prefers native packages, except for in some specific exceptions.
I currently used KDE btw, because I like customization, but I would install a boring but predictable Gnome for casual users that want to try Linux.
Why does every Linux desktop have to accomodate all the whims of every user? Why does GNOME have to be compatible with your homegrown DE? Do you also expect emacs to ship features that are compatible with vim?
I want a consistent experience, that is open source. I want a libre Apple experience. There should be nothing shameful in having a group of people pushing towards that goal.
Red Hat isn't a walled garden and its not clear that there would be anything virtuous about it if it were. It certainly wouldn't provide you with an Apple experience on Linux. Apple makes i think about 9 billion per quarter on Macs. Red Hat on the other hand makes 1.25B and virtually zero from the desktop. Caring about desktop experience and having a pile of money to pay people to work on it gets you an Apple experience not limiting user choice nor abandoning the idea of building an ecosystem of interchangeable parts. Let me turn your question on its head.
> Why does GNOME have to be compatible with your homegrown DE?
Why instead would one desktop environment be in any fashion incompatible with any other desktop environment? If standards it depends on need to evolve logically they ought to evolve in such a fashion as other applications can non trivially evolve with new and improved standards.
> Do you also expect emacs to ship features that are compatible with vim?
Actually yes. The most prominent example would be evil mode but there is also viper and several newer implementations of modal editing and features first dreamed up for vim.
Similar to != compatible. Evil-mode is inspired by vim, but you can't use it on vim. Elisp isn't written to be as compatible as vimscript as possible. They are two distinct projects.
GNOME wants to become its own distinct project, not because of superiority or dislike about other DEs, but because they want to go for strict conventions and exotic ideas which are often at odds with other DEs philosophies. I am still waiting to hear why this is so bad. I don't hear anyone hating ElementaryOS for trying to do the same.
It's hard enough to get people to support Linux.
> Why does every Linux desktop have to accomodate all the whims of every user?
is exactly right.
I think having strongly opinionated desktops to choose from is a good thing, and that that kind of pluralism is more likely to lead to the development of new ideas and execute them well than pluralism within desktops.
I wish that XDG standards ran deeper (for example, that there was just a single, well-supported VFS implementation for all DEs, and that drag-and-drop could be handled in a universally compatible way). But in terms of experiences, I think differences between DEs is a good thing, and that some DEs should have a narrower focus than others so that they can execute their visions better.
you can install Pantheon, the desktop from Elementary OS, on some other distros
You're not supposed to do anything at all!
https://blog.elementary.io/updates-for-june-2020/
> elementary OS 5 will continue to receive security and stability updates to its underlying libraries until April of 2028.
You're acting as if this is some kind of massive and unprecedented inconvenience, but this is Linux, where users routinely recommend to one another that they try a different distro to solve application problems. By comparison, what you brought up is pretty tame.
They are also very slow in supporting the latest LTS builds. I think people should move on to something like Ubuntu Budgie when they want a simple and clean experience for themselves or others.
Ubuntu Budgie keeps their LTS and interim builds in sync very well with upstream Ubuntu and upstream Budgie & I have never seen fossfreedom shoot down a PR that makes sense. He's often encouraged others to contribute to the distro in very healthy and normal ways.
Of course, sometimes failures do occur, but that happens whenever you do a major OS upgrade. I've had Windows fail on upgrade more than I've had Ubuntu fail on upgrade, but that's just my experience.
This in comparison to, well, Elementary OS saying scrap it all and reinstall. That wouldn't be so bad if they were, like, a technically-minded distribution - but for their primary market who hardly knows anything about updates, completely unacceptable.
ps: I agree with your point about EOS trying to make things invisible for the user and not addressing this well though.
https://ubuntu.com/blog/how-to-upgrade-from-ubuntu-18-04-lts...
Distros like Ubuntu make upgrade failures harder to work through because they go to some lengths to wrap the dist upgrade process in some software that reverts everything if anything goes wrong.
But if you just move all your apt sources forward to the next distro and read any specific error messages you get, you can generally work through them.
Distros other than Debian itself don't typically recommend that you do things this way, and instead have some extra software for going through the process. But the Debian way still works on them.
And it wasn't my personal machine so it would've been unreasonable for me to do so.
1. One that looks nice and comes with bells and whistles and is closer in UX and functionality to Mac than Windows
The same is true for Debian, Fedora, and... well, every Linux distro I've ever known people using on their desktops/laptops. For good reason, because it is a pretty essential feature in that context.
What annoys me is that AppCenter takes up so much space in their blog posts and sponsoring campaigns, because I feel it's the worst of their apps. The navigation model never made sense to me (why is search disabled when I'm on "Installed"?). It also doesn't look good, because there are way more tiny utility apps than there are good icon designers, and often the icons are just shown on a giant, colored rectangle because apps have no banner images.
I wish they'd do it more "magazine-like" as Apple does, just with more unsplash images instead of Apple's terrible Corporate Memphis illustrations.
I'm stuck on an older kernel because the nvidia drivers are borked, but worst of all the AppCenter is truly horrid.
For a distro that does such a good job at making sure that there is a GUI alternative in addition to all the normal things you'd do in a CLI, the AppCenter is really crummy.
Ubuntu actually does a better job of making sure you can sort out your drivers through the GUI.
[1]: https://www.amd.com/en/support/professional-graphics/radeon-...
For example, the Music app. Comes built-in, looks great, only works with your local music library if you copy it over and reorganize it the way the Music app wants. No online music streaming of any kind.
The Videos app? Again, only your local library, and that's it except (IIRC) it has videos from The Guardian newspaper of all places as one of its four online sources? What's up with that?
The Maps app? Looks beautiful, again, but the last time I tried it, it could hardly handle directions. Maybe Directions work well now, but it's just... not good.
Every app is like this - barely functional.
None of the issues you mentioned are really fixable. Online streaming is a bit of a pipe dream for linux, legal(ish?) solutions are going to be hacky and break constantly since nobody has the negotiating power to work out deals with labels. Same with videos, alphabet is not going to just allow random apps to bypass the entire point of youtube (serving ads to eyeballs and tracking said eyeballs to serve more ads). Maps does suck, and that's mostly and openstreetmap issue. You could integrate google maps but those API keys are not cheap.
The Videos app likewise plays your videos. No idea where you're getting that bit about The Guardian, but that's just flat out false.
There is no first-party Maps app, so again, that's not something that the elementary team provides...
Constructive criticism is helpful, but only when it's informed.
I feel like the more internal team relies on them, the better they are. And, well, I guess they don't use multimedia apps a lot.
- Calendar support for Exchange so I don't need to log into DavMail all the time
- A Mail app that lets me view emails in plain text first, or some kind of option to avoid downloading images and trackers in HTML mode.
- The Music app has a hard time with my library which is on a network drive. It keeps trying to reload all the songs.
- A better UI for choosing my graphics drivers (the Ubuntu one works fine)
A list of qualms should probably be followed by the things I love about elementary:
- It looks fantastic.
- The terminal, code, and files apps are where I live and they are great. Code isn't a replacement for VS Code, but it's quick and light and fantastic when I don't need e.g.: a linter.
- Back in the earlier versions, I had a lot of small graphical glitches, but things are extremely solid and reliable.
- The OS truly feels unified; not some hodgepodge of poorly conceived UI elements and metaphors.
I don't think this is as impressive as they think it is. On the contrary, this seems to be a very very small number that I wonder why it's mentioned in the first place. I don't use these GUI frontends, but if we ever want any Linux distro to be more user-friendly. Almost everything should be available.
Also I noticed that Elementary is taking a 30% cut (with a 50c minimum) which I find obnoxiously high. We have been complaining about Apple and Google, yet "we"(?) cannot seem to be doing better ourselves. It's a bit disappointing.
> Also I noticed that Elementary is taking a 30% cut (with a 50c minimum) which I find obnoxiously high
It does seem really high considering they are just taking from one Stripe account to another.
The Calendar app doesn't work with Google Calendar as a backend; or, at least not with a ton of hacks, which is what one chooses Elementary to avoid.
Overall the Pantheon WM feels rather sluggish, even on modern hardware. Managing windows just feels so clunky compared to KDE or even Windows. To be fair, macOS is their direct competitor.
I paid a bug bounty ages ago to get streaming audio to work in their Music app but I don't think that went anywhere.
I think they might be able to get something solid in the next few years.
So, good luck Elementary folks. I really hope you don't succumb to the ever-progressing "flatpak the world" mentality.
"AppCenter continues to fill out with apps from developers—and since the move to Flatpak, all apps that have been released for OS 6 will continue to be available on OS 6.1 and beyond!"
I may give it a shot again.
I especially liked the ability to customize the key combos so that it acted reasonably close to a tiling Wm (I had moved to it from awesomeWM) which was very handy on an Thinkpad x200s with only a track point and no touchpad, but with the added perks of a fully baked DE. And that was in addition to the very nice aesthetics, rather than in spite of.
If I'm going to be staring at something for many hours every day, those details really do matter to me. I tend to be able to fix everything else on my own.
Most people would stare at an IDE and browser many hours and not at the DE control's panel or File Manager, and IMO this core apps should have had the UI fixed years ago and not have stuff still changed at each updated.
I don't think elementaryOS's visuals have changed all that much? Certainly nothing like macOS, which has gone from Mavericks → Yosemite/Catalina → Big Sur in the same time frame.
But perhaps it's okay for this one distro to have an obsession with visual design? Maybe your anger should be at the distros that do all the things you want but are too ugly to use?
>Maybe your anger should
What wording in my comment can be interpreted as anger ? I did not intended that so I would like to understand and express myself better next time.
The linux desktop is a house on fire, and elementaryos is standing in the living room holding up paint to a smoldering wall.
This is regularly claimed but I can only presume it's by people who have never ever used a Mac, just looked at screenshots.
It does not work in any way like a Mac.
Sure, it has a dock at the bottom and a panel at the top, but so do dozens of desktops now. It's the default layout of XFCE. That tells us nothing useful.
It doesn't have a menu bar. It doesn't have a repositionable dock. It doesn't have Miller columns. It doesn't have a folder structure anything even vaguely like macOS'. It doesn't have app folder bundles, or drag-and-drop installation. There are no desktop icons, not even as an option.
Apps don't have menus at all except a hamburger in some. Windows don't have title bars, they have something like CSD.
It's a Linux. It's a slightly weird Linux but it's way less Mac-like than, say, GoboLinux is under the hood.
Just because the desktop is quite clean and minimal, and there's a GNOME-like panel at the top, it has a passing cosmetic resemblance, no more.
You could make Budgie, LXDE or LXQt, or MATE look much more like the macOS desktop than Pantheon is or could be.
If anything it's more like iOS with overlapping windows than anything Mac-like.
- Dock
- Files aka Finder
- Code aka Xcode
- Global status bar icons
- PiP
- DND
- App Store
- Screen Time
- Calendar
- Camera
- System Settings
- The folders inside your Home folder, which absolutely resembles macOS
Is everything a pixel for pixel rip? No. Does it use the same technology? No. But the UI/UX is more than just “inspired by” Apple. “Cosmetic resemblance” seems to suggest these things are coincidence, but when it’s clearly not.
> Dock
Also used by default in XFCE, Unity, GNOME 3, Budgie; optional in LXDE, LXQT, KDE, MATE. Also used in Windows 10 and 11. Rejected.
> Files aka Finder
Not significantly more Mac-like than Nautilus, Nemo, etc. Rejected.
- Code aka Xcode
XCode is an IDE. Elementary Code is a text editor. Every desktop GUI has a text editor. Rejected.
- Global status bar icons
As also used in XFCE, GNOME, MATE, Unity, LXDE, LXQt, & every version of Windows since 1995. Can't be customised; ordinary Linux apps can't add new ones. Rejected.
> PiP > DND > Screen Time
Since I am not sure what you mean by them, no comment.
If "DND" means Do Not Disturb, GNOME 3 has this too.
> App Store
As used in every modern distro/OS. GNOME has one, Ubuntu has one, Windows has had one since 2012. Rejected.
Again, every GUI OS has this. Rejected.
> Calendar
Again, every modern OS has one. Click the clock in XFCE, for instance. Rejected.
- Camera
Ever seen "Cheese"? Rejected.
- System Settings
Every OS has this, and most call them Settings because Microsoft uses "Control Panel". macOS calls them "System Preferences". Rejected.
> The folders inside your Home folder, which absolutely resembles macOS
100% standard Linux layout, and share the same names as the defaults in Windows as well. I remove the Linux ones and symlink the ones on my Windows partition so they stay in sync, and even the icons automatically reappear because the names are 100% identical.
It is not only unreasonable to claim that a list of features that basically _every modern OS_ has mean that one OS in particular is a copy of OS X, it is laughably ridiculous.
On this basis, I could also list a set of features that OS X copied from Windows which were missing in Classic MacOS and NeXTstep (Alt-tab app switching, keyboard dialog box navigation, Safe Boot Mode, Fast User Switching, etc. etc.) to claim that OS X is a copy of Windows.
No. This is completely weak and bogus.
Ok I’m not going to go point by point as it seems you’ve completely missed mine. I’m not suggesting a Mail client coming with the OS is unique to macOS. I’m suggesting the UI/UX is, or in this one case, actually comes from NeXT which became macOS. If you can’t look at these and understand where it comes from, regardless if it’s in other Linux DEs (look at its history evolution), or tell the difference between a taskbar and a Dock (and know their history), then I don’t know what else to say.
Lastly, telling me that my argument is weak and bogus makes me want to engage negative percent. It’s not a nice way to interact, and is frustrating when you don’t even get my point when I clearly said “ But the UI/UX is more than just “inspired by” Apple”
I will have to try Elementary's new email client – it replaced Geary with its own home-grown one in recent releases. I did not give it more than a cursory look. I have used Mail.app in several OS X releases, but these days, I use Thunderbird instead. I am know the GNUstep Mail client a little bit.
For me, the menu tree is a big part of an app's UI. I favour older versions of MS Office that still have menus rather than the horrible Ribbon UI of Office since the 2007 release.
Since Elementary and its apps have no menu bar and no menus, I don't really feel that their UI can be like macOS apps' since a core element of the UI is missing.
Regarding the difference between a taskbar and a dock – well, ISTM a lot of Linux distros don't, frankly. I think there's a big difference. But saying that, the Windows taskbar (that is, the original of the species) has been getting increasingly Dock-like in recent releases. Windows 8 even dropped the Start button and Start menu. 8.1 put it back starting a full-screen app launcher not unlike Launchpad in recent macOS releases; Win10 and 11 have made it more and more Mac-like, so that in Win11 it's centred by default and can no longer even be arranged vertically at the side of the screen.
(Personally, I hate the new version.)
So I'd say that the company that invented the taskbar is trying very hard to make it more like a dock. And of course imitation is the sincerest form of flattery.
I had a quick play with an older version of Elementary 3-4 releases ago. I didn't like it much. Last week I spent a few days getting to know it better. It's a very "opinionated" distro. I like macOS a lot – I'm typing on it now – and if Elementary was more Mac-like, I would like it more. As it was, I was impressed with the clean, integrated design, but actually using it, no, I don't find it Mac-like at all. Not really in any way.
You can, if you wish, read my thoughts on it here: https://www.theregister.com/2021/12/21/elementary_os_61/
It is clean, it works well, its programmers have *views* on how to do stuff and they've found a model that pays and they're working on it. Good for them. I do not think I will be using it myself.
Would be good to get it out of the PPA and into official ubuntu repos.
- having windows automatically start at "empty" coordinates if available,
- having sufficient shortcuts to position them on specific square coordinates.
Since I have a wide screen, I have configured win-[, win-] and win-\ to go to the left third, middle third, right third.
On a thinkpad, I complement that by having win-left to the left half (50%), win-right go to the right half (50%), win-pgup toggle half size top and win-pgdown toggle half-size bottom.
Then I can assemble the windows very precisely with just the keyboard (full screen, half screen, quarter screen, one-third of a screen, one-sixth of a screen) both on the laptop display and the external wide screen, while keeping the full convenience of normal titlebars and mouse operations to reposition windows by hand if needed.
Add to that some custom scripting to make win-left and win-right move between physical screen, and you get a tiling solution that do not require any specific window manager.
Of course, space taken by the title bars might be a problem, but with Mica and the like (more generally, titlebar moving from "full of wasted empty spaces to "where tabs are shown", like Edge now does), I think tiling WM are quickly becoming a thing of the past.
All distros are basically the same but with different default packages and package managers.
Fortunately, regolith is open source and has repos for their default configs. Get the right packages installed on Arch and copy the regolith configs for those packages. and you have yourself an arch regolith clone.
But it would just be nice to have the full DE backing behind i3 as a distribution. Yes I know you can do everything here with list of tools, but I just want a configuration that I know has been thought out, with best practices, and easy to modify in a reproducible way. The list of dotfiles and little adjustments alone you need to marry i3 with a power manager, menu bar, network manager, is a lot! I just want the work done for me (a DE).
It seems to me you are introducing complexity for complexity sake, with nothing to be gained - except maybe nerd credential points for using i3?
On my laptop and even w/ proprietary drivers installed, I cannot get eOS to hardware accelerate videos with either my embedded Intel graphics card or my discrete nVidia card in Firefox or Chromium.
The UI is full of really weird inconsistencies and is extremely opinionated without any way of changing some basic things (previously adjustable via `gsettings` but it seems that has been intentionally deprecated). For example, the elementary File Manager forces single-click-to-open ("web-style") behavior for any folders, but then switches to double-click-to-open for any files within those folders, so you can select a file but not open it but you cannot do the same with a folder. If it weren't for elementary tweaks (neutered as it has become for eOS 6), elementary would be completely unusable.
The decision to not even have an official upgrade path besides "nuke everything and reinstall" makes me question the clarity of vision of elementary OS developers/decision makers. You want to be the macOS of the Linux world but you have a worse upgrade experience than any other Linux distro bar none? You purposely advertise to "less techy" users that are exactly the kind of users that won't be able to handle a "back up your files, wipe everything, and clean install" upgrade path?
They added multitouch gesture support... but the only gestures available are for interacting with the desktop window manager (show all windows, move to prev/next virtual desktop)? Fortunately my `syngestures` that I wrote for elementary OS 5 works just fine alongside eOS 6's so-called "multitouch support" but for something that was going to be their killer feature, you'd think they would integrate some of the TouchEgg configuration into their touchpad gestures control panel UI, wouldn't you?
Why did they bother trying to redesign a window switcher of all things?
See also this clip by Chad, I mean Nick.
Coming from MacOS, it does so many things right and Danielle and the other contributors have such a great mindset focusing on accessibility and platform building.
With their dev conference, they’re doing great work encouraging people towards native app development with an eye to performance and HIG.
Everything looks so good.
And yet, as a daily driver for development, it’s just never quite there. The edge cases stand out enough that I always want to reach for Manjaro or Zorin or the like instead.
But I love their approach and think it’s ultimately “the right one”, so even when I’m not using it, I can only support them and cheer them on.