Elementary OS 6 Odin
blog.elementary.io
blog.elementary.io
Can't help but notice that the Sound indicator's dropdown doesn't line up with the icon above it. But the fact that a single UI glitch is a notable item now says a lot about the overall quality. Very tidy, still customizable, and a suite of built-in software that ought to cover most users. Good stuff.
Looking at screenshots of Elementary, it's so much more obvious how things work and what widgets do than on modern Mac OS. I've been running it in various VMs for a few years but didn't seriously consider switching to it full time until Big Sur was announced. I have too many paid-for Mac apps like Alfred, Keyboard Maestro, Hazel, Photo Mechanic, and others that prevent me from moving to Linux full time.
With all OSX Finder's faults and all the annoyances of the OS, I still think it's miles ahead of any Linux distro I've tried to get on with.
Two words "Quicklook" --OK. that's one word made of two. I find it incredible that no Linux desktop environment has something like this. Pressing spacebar to preview the content of a file without needing to open it first is something I do literally every single day and often multiple times. Why, in 2021, does every single Linux desktop require me to open a file first to see what's inside it?
That's my biggest bugbear with Linux desktops. Other annoyances are more vague. Mostly to do with the inconsistencies in the interface. Coming from something like OSX which, for all its faults, presents a completely consistent user interface, every Linux desktop I've tried [even the polished ones like Ubuntu & Mint] always have those annoying glitches & inconsistencies which just scream 'Designed by Committee' [which I suppose they are, to an extent].
As regards apps; a big part of what I do is graphic design and that pretty much rules Linux out for anything work-related. Whatever you think of Adobe, Photoshop and Illustrator just blow The Gimp [dire!] and Inkscape [tolerable but clunky] out of the water.
I did have high hopes that new kids on the block Affinity might bring their Photo and Designer [which actually do give Photoshop and Illustrator a run for their money] to Linux. But it seems they've decided to follow Adobe's lead and make thei offerings OSX & Windows only.
About the only quality graphic design app I've seen available for Linux is Krita. But it's more of a digital painting app than an image manipulation one. And, of course, no vector graphics.
I used Mac OS X from 2009 to 2012.
It was sold as consistent and keyboard friendly back then.
However: Describing a desktop interface where selecting a single word using the keyboard needs a different keystroke in different apps, that is - rich.
Maybe this is fixed now but back then even official applications had different ways to select a word using the keyboard: in one it was cmd+shift+arrowkey in the next it was ctrl-shift-arrowkey or even fn-shift-arrowkey.
Especially infuriating because one of the most used ones in other applications meant "navigate back" in Safari.
Meanwhile in Windows and Linux ctrl+arrowkey always moves one word and ctrl+shift+arrowkey almost always (hi Excel!) selects a full word.
Maybe it is fixed now. But I'll never let a Mac user lecture me about consistency again.
This was a problem in Snow Leopard and probably the two next releases as well: I always upgraded as soon as possible.
The combination of this feature (including sort-by this directory size when listing dir contents) and quicklook makes the Mac the only place I ever want to actually manage files, instead of dumping them in Downloads and declaring bankruptcy. Two basic features that would have me on Linux in a heartbeat. I'm begging you!
(Ironically, Windows Search is so useless compared to Spotlight that the give-up-and-dump-and-search approach works best on Mac too...)
Maybe not very Mac. ;-p
KDE's file manager has a previewpane that you can enable if you wish.
(Though the whole thing, while useful in practice, feels like essentially a workaround for unnecessarily heavy and app-centric UIs. Having both an “Open” command and an “Open, but faster and does less” command feels like humans succumbing to the needs of computers, not computers providing an additional convenience for humans. But that is a much more complicated problem that I don’t think has ever been solved in a GUI context.)
Right, I'm sometimes puzzled by how people approach their personal "deal-breakers" that prevent Mac->Lin (or Win->Lin) switch. Many of the issues are presented as though they are bugs or clear-cut missing features, but are in fact subjective preference statements.
In this case the "spacebar to preview" feature might be a must-have for some, but to others it might not be important, and maybe even to some be perceived as an anti-feature. I personally see this particular feature as redundant (given there is already thumbnailing and a means to open files) and unintuitive, and thus just more clutter. But I also don't care too much, since I could always disable it if it came pre-enabled.
The worst cases of this is when some insist that not only is their particular preference the The One Way, but that stubborn Linux DE developers are somehow at fault for not implementing The One Way, and that's why Linux will never be popular. There are plenty of things that might keep "normies" from switching, but random obscure DE features are not them. People tolerate FAR worse UX disasters than any major Linux DE.
When I read how Linux DEs are "obviously" garbage, I just generally throw up a thought bubble with few question marks above my head in confusion and go on with my day. I find Mac OS a frustrating experience personally, but assume it's mostly a result of not being used to it.
I used it at my previous job, 2014-18 -- got the choice between a Mac and IIRC a ThinkPad when I started, and never having used a Mac before I wanted to finally try it -- and got a MacBook Pro, can't recall the vintage. Broke the screen a couple years later, got a 2016. But, anyway: Took a while to get used to the OS, and certainly didn't feel all that "super-intuitive" as the hype has had it for all these years.
So, observation #1: If I could get used to it (and I got quite comfortable after a while), I can't see why people can't go the other way just as easily.
Can't recall the OS version at the beginning (one of the last of the big cats, or some US body of water?), but later it upgraded via various Californian (or Oregon, Washington?) rivers or lakes to Maverick and finally, I think (for the last few months), to El Capitan. Can't recall if it set in right away, but after a couple of years at the latest I noticed the same thing as many here have mentioned, namely
Observation #2: With every new version, the OS gets a little more limited and locked down.
Exactly! It really feels like there's this massive disconnect in these conversations that I can't explain.
Whenever I have to use macOS, I'm a complete mess, and to me like it's missing basic features and doesn't support my workflow without lots of 3rd party add-ons, etc. That said, I probably am using it wrong, so I don't go around saying that macOS desktop is obviously garbage. I know that other people prefer this style, so that's A-OK.
Many of them were asking for email preview popups directly from the inbox. Instead of implementing it, he realized the need wasn't for a popup but for a wait to see your email quickly. The real issue was that it was slow when you clicked on an email, so he made it fast instead.
Since preference is pretty important, does it really matter? If they have a workflow they like and no significant pressure to change, why would they? You speak as though it is everyone's most pressing goal to transition to Linux Desktop for some reason.
> The worst cases of this is when some insist that not only is their particular preference the The One Way, but that stubborn Linux DE developers are somehow at fault for not implementing The One Way, and that's why Linux will never be popular.
There's a lot of reasons Linux will never be popular, developers ignoring the workflows and use cases of potential users is just one of them. The condescending attitude of many of its supposed proponents is another. Why would I want to switch to a Desktop that's as clunky and haphazard as Windows 10, just in different ways that I have to relearn?
>Having both an “Open” command and an “Open, but faster and does less” command
>I personally see this particular feature as redundant (given there is already thumbnailing and a means to open files)
When I'm browsing a folder of my image files and can't remember which of dozens of files is the one I actually want, QuickLook lets me pop open a screen sized modal preview window, which is as good as having that file open in the app. If it's not the file I'm looking for, I hit the up or down arrows and the next or previous file's content is displayed in the already open preview popup. If some of those other image files were created by Illustrator, I've just saved myself an additional ~30 seconds, by not having to open that app too. And, once I find the image I want, I can open it in its associated app directly from a button in the top of the preview pane.
But, even if you're not a graphic designer, QuickLook has its uses for other types of document too. It can preview the content of source code files [with syntax highlighting] preview the rendered output of things like HTML, PDF, Markdown, AsciiDoc etc. You can even scroll through the pages of a multipage document.
This is another timesaver I use all the time. Being able to check the date or bottom line on a received PDF invoice, without opening it... being able to check whether I mentioned something in a letter I sent, without firing up a text editor or word processor... being able to see a rendered representation of what an HTML file contains... etc. etc.
I just think that it the apps that make it necessary shouldn’t do so: if Photoshop (or GIMP, or LibreOffice, or whatever) takes 30 seconds to open a file, the solution shouldn’t be to layer on yet another option and make me think whether I want to suffer that delay or reduced functionality, it should be to eliminate the delay.
Yet (as I’ve also said) I don’t think anybody came up with a plausible way to do so (even a purely technical one, disregarding economic incentives for app developers).
[and, if the file in question is a Photoshop or Illustrator document, you'll really appreciate how much time QuickLook saves you, by not having to wait for those behemoths to creak slowly into action]
There are also certain plugins which can enhance this functionality by providing extra info, such as pixel dimensions on image files, allowing browsing inside archives, syntax highlighting of source code files etc.
Maybe peoples' descriptions on here aren't doing Quicklook justice, for those who aren't familiar with it. Once you've used it for a while, you can't imagine why it's not a standard feature on any modern desktop OS.
If remembers the state globally so if you leave it open it will always be there when you open a file manager window if you toggle it off it will be hidden next time too.
It's not open by default perhaps because not everyone regards it as maximally useful. Say I open a file manager and have it sized to 1/4 of a 24" 1080p screen. On the left hand side of the screen is a panel which lists common and places and other items. This leaves a space for 15 icons on the screen at 48px size big enough to comfortably read and note some details in a preview.
Opening the information pane reduces that to 10 icons greatly reducing the space available for the main purpose of the application.
What if I open it to the full size of the screen? I can now view 56 icons at 80px where now they actually are a pretty decent size to see whats in them in their own icon.
I think most of the time having the info panel shown would decrease the utility of the window.
Indeed, I don't have any use for information or preview panes, I disable them all when encountering. IMHO they are a distraction from the content itself and make the view arrangement clumsy.
Initially I missed QuickLook a lot after switching away from macOS, but now I haven't thought about it in years.
Aside from that, QuickLook remains one of the few examples of a generic extensible document reader in modern operating systems. The way it gains the ability to read new types of files just by virtue of the owning app’s bundle being present on the system (no installation necessary, and it goes away when you delete the app) makes so much sense, and it’s a shame that there’s no equivalent on Windows and Linux.
...were there other ones in classic operating systems? QuickTime <= 10.2 did let you install third-party components (I wrote one recently), but what else was there?
I wrote an AppleScript to convert DOCX or PPTX to TXT using qlmanage -p | textutil
Using QuickLook to turn anything to HTML:
http://hints.macworld.com/article.php?story=2010071209045151...
Using textutil to convert HTML to TXT:
https://stackoverflow.com/a/32621215
AppleScript source code: https://pastebin.com/JRbXTv8d
I'm on GNOME 3.38.4 on Debian sid.
Well there's the web based Figma, somebody even created an unofficial desktop app for Linux. Figma has plugins, tons of them for many graphic design tasks. It blows the Affinity stuff out of the water imo.
Additionally, it’s very much geared specifically toward UI design and prototyping, whereas something like Affinity Designer or Sketch also work well for generic screen-targeted vector work.
True for project files, but you can export images and SVGs.
> it’s very much geared specifically toward UI design and prototyping, whereas something like Affinity Designer or Sketch also work well for generic screen-targeted vector work.
That's where the plugin ecosystem comes in. It's quite usable for simple/modern vector work that's more component oriented graphic design than hands on illustration.
tl;dr Web Sharing, Notes sync, iCal sync, HFS+, apps sync.
https://news.ycombinator.com/item?id=26245164
I gained NVME drivers for using larger SSDs, so tolerated the changes, but I'm not sure if I'll be able to make the leap to ARM M1 if Big Sur is holding me back so much.
APFS isn't writeable on Linux or Windows without additional drivers. Nor is HFS+. It's quite inconvenient. If ZFS was usable in that way, I'd consider it! But for now I still have to keep my boot volume as HFS+ so I can at least read files, and use FAT32 to shuttle data around.
Purely from a visual standpoint I like the Big Sur redesign way more than I expected. It looks like it was made for dark mode, unlike previous versions where dark mode was kind of an add on. The whole material/translucency thing is really coming together. Not so happy about the non-visual aspects though, like all the unlabeled, barely distinct line icons.
Yeah, I actually like Big Sur more than I expected to, and it's the first version of macOS with a dark mode that I actually stuck with. Most of my quibbles stem from choosing minimalism over discoverability, e.g., hiding the document proxy icon, making all the keyboard shortcuts in menus grey because somebody clearly decided that looked prettier and more subtle rather, etc. But in practice, all the shirt-rending over how the excessive transparency would make everything unreadable and illegible just hasn't matched my experience.
I think iTerm is probably not just the best terminal on any platform, but also one of the best software applications ever made, period. I would looove to have something like it on Linux, but I've spent several evening beers googling it, after switching to Linux, and there just aren't any.
Random things I like about iTerm, off the top of my head:
- not only is it extremely customizable, it can sync those settings across multiple machine easily via any shared-folder mechanism (I use Syncthing, but you could use iCloud Disk or Dropbox or anything)
- infinite history/scrollback retention, even across restarts
- extremely good performance, regardless of having dozens of windows open (maybe only on high-spec Macs? that's the only kind of Mac I use, since macOS is so crazy slow generally)
- timestamps on every line, hidden by default but can be shown after the fact whenever you want (e.g. hmm... when did this script output this line of text? oh, last night at 1AM, it must be hung then, kill it)
- detaches your sessions when it updates itself, then restarts and re-attaches them
- great shell integration (https://iterm2.com/documentation-shell-integration.html) enables useful extra features
- Instant Replay
- scriptable, so each project can have a little script that opens iTerm and opens a tab that runs a local dev server, another tab with REPL, another tab with unit tests in watch mode, another tab with whatever... and so on
It's not "mine" but I adopted it years ago and cannot live without it.
I don't know about detecting soft-splits, but tilix has a really nice split pane setup.
See uLauncher, AutoHotKey and Darkroom, respectively. They're not 1:1 in terms of feature parity, but you can get pretty close for free.
Also since you're on Linux I'm looking for a linux alternative for this tool. https://github.com/chipsenkbeil/choose It's a gui chooser I don't know if such a thing exists for linux. I suspect it might be quite easy to implement one with electron. By the way what I'm not looking for is something like alfred for linux. That would be an overkill.
(I think there was a nunber of issues with extensions but I guess that is fixed now.)
I think it was meant to be perhaps not necessarily noticed but at least deliberately positioned like that, and not a UI glitch. Look how it's situated directly between the microphone and speaker indicators on the top bar, and the screenshot appears in a section that says, "the indicator now shows both input and output devices right in the popover".
> I think I actually like elementary's evolution of the UI styling (from when it started as a pretty direct clone) better than I like Apple's
I'm a lifelong freedesktop user who doesn't even own a Mac but will nonetheless defend OS X's UI, from around the time of Snow Leopard or Mountain Lion or so, as the best visual design ever with respect to graphics for a conventional desktop operating system and utterly timeless—with some notable exceptions. One of those exceptions was Apple's longterm insistence on its terrible tab look—with tabs shown detached from the content its associated with and attached to the window chrome above.
It's really irking that Elementary copied that tab look, because absent any admiration, it's basically indefensible. Elementary's copying it strikes me as an example of the kind of irrationality that originates from conservatism-for-the-sake-of-it mixed with how people come to take pleasure in idiosyncrasies and other incidentals. (Similar to how when Pluto was designated not to be a planet, it led to people sort of staking their identity on choosing to insist that it be treated as one and publicly aligning themselves with "Team Pluto" for the quirkiness.) It's one of the things that they should have deliberately opted to break with the influence in order to be "better than Apple's".
And none of the other screenshots show the microphone icon, it's only in that one. Only shows up with a mic connected I assume.
Or could the mic be there to give a "something is currently using your mic" indicator?
- Keyboard shortcuts.
- Drag-n-drop behavior.
- Menus and menu item organization.
- Application interoperability.
If I ever had the money to invest, I'd finance the making of a legitimate replacement for macOS out of KDE or something where the dulling the thousand paper cuts was the focus, not the landing page screenshots.
I knew that MacOS’ menu organization was a bit better on average and application interop more common (especially with things like Automater) but the keyboard shortcuts thing really surprised me.
Consistently at the system level and application levels, shortcuts seem so much more natural and easier to learn on MacOS. Use of the super key, a letter, and maybe a modifier is the norm, whereas Linux seems to have heavy use of control plus multiple modifiers and even function keys for just about everything.
Moreover, the super key feels so much better placed than control for common shortcuts.
The problem that persists is related to what you mention. Nearly every application seems like it was making it a point of pride to be a maverick about application menus and shortcuts. I think a lot of it stems from decades of trying to be like Windows in some attempt to lure Windows users over to Linux. The result instead was still a persistent lack of commercially popular/essential software that gives Windows its value and a user experience in terms of consistency and execution that's no better than Windows. It ends up being the worst of both worlds, but... it is free, and basically a miracle that any of it exists and works at all considering the arduous and meandering way it tends to all come together.
So, on the one hand I'm in bewildered awe of it and on the other I'm constantly frustrated by it thinking, "Ugh. What? Why? It doesn't have to be like this." If I were ever in a position to do so, I'd gladly drop $10+ Million on creating a Linux Distro that was little more than KDE Neon or Kubuntu, plus popular applications, all patched and built with menu organization and keyboard shortcuts that follow Apple's HIG from ~2005.
I put Elementary onto an external SSD with about 20 other Linux versions, yet it's the one I choose to use most often. It's some combination of familiarity, discoverability, and compatibility that made me like it most.
In practice I still use macOS 10.13 for myself (I really like the AppleScript API into every process) and Windows 10 at work (because my boss tells me to), but Elementary would be my choice of Linux distro if/when the day comes when I need to move.
Elementary doesn't have that issue and so can iterate with small quality of life improvements, without having to throw the baby out with the bathwater
That said, I wish them all the best, because I use 'Files' (but not the distro) since it's the best I've found for the odd occasion where I think using a GUI file manager will be easier than the command line. I still wouldn't say it's good, just the best I've found. (Not opening everything on a single-click is a vast improvement!)
Linux distros used to love to do this (maybe some still do?). They'd put whatever the "native" browser was for a desktop in as the default, and do the same for a bunch of other stuff.
Early Ubuntu's success was partly due to ignoring that crap and just installing whatever the user'd almost certainly actually want. Its defaults were way better—actually somewhat helpful, rather than harmful—as a result. Browser, on Linux in 2006? 95+% chance you want Firefox, so here you go. And so on.
It's still something I'm leary about. I switched from a mac to ubuntu (gnome) recently, and wonder how much lack of facilitation/communication leads to so many separate or incomplete apps. For example, I may use the gnome calendar app but also still have thunderbird setup to handle calendar links. Or evince (akin to preview.app) doesn't support drag/drop editing of pages so I have to also install 'pdf arranger' to do this.
I think there's a chaos tolerance one has to have in approaching linux, which is fine! On the plus side I've really appreciated hearing quickly from developers when filing bugs.
For anyone curious, I keep track of cross-platform workflows here: https://docs.google.com/spreadsheets/d/148zTJUwfVv9xfDcpSoH3...
Elementary is aimed in part at non-power-users, moreso than most other distros besides Ubuntu. Most of them don't care or even know what browser they're using.
So yeah, there's definitely a value IMO. In fact, I know for a fact that my mother in law has been using linux for years because they just set it up for her and installed whatever she needed to get stuff done (libreoffice, zoom, whatever).
To that end, why not just go with Firefox, which would appease the more technical users, would be closer to what most other distros out there are doing and would also have a higher chance of it being familiar software to all users?
I think that for the most part custom browsers are only good when you want to include something functional, yet minimal in your distro and want to save space or something like that.
Edit: admittedly, an argument could also be made about having software look and feel consistent with the rest of the OS, where such a solution could be better than off the shelf browsers, at least without heavy modifications.
If I installed this, even for an elderly relative or whatever rather than myself, probably my next step would be to install Firefox. (And for others it might be Chrome of course.)
But mostly it's the first, I couldn't possibly have picked GNOME Web before this conversation, I didn't know it existed!
A summary of some of the cooler features:
Performance: General performance improvements on all hardware resulting from optimizing for Pinebook Pro and Raspberry Pi - namely, reducing and asynchronizing inter-process communication between desktop components, removing unused code, and reducing disk I/O.
Firmware: Linux Vendor Firmware Service now built-in, enabling firmware updates from within the OS.
Flatpak: all-in on flatpak, all AppCenter apps are flatpaks, as well as some Elementary apps like Web.
Portals: apps must explicitly request permission to get access to files or interact with other apps. Can tweak these permissions in System Settings.
Mail: The Mail app now sandboxes html emails.
Multi-Touch: Extended from supporting just desktop to various apps now too.
Multi-Tasking: Better hot corners + new window and workspace controls.
CalDav: Tasks and Calendar now designed around the CalDav format, making importing and sharing of tasks and calendar items with other CalDav apps easier.
Dark Theme: system-wide, applies to GTK apps too.
Terminal: smart-paste protection extended from sudo pastes to multi-line pastes.
More OEM/Vendor friendly:
- Installer is simplified and streamlined - network connectivity, user account creation, and updates moved out of the installer and into the installed OS. Better for vendors & OEMs.
- Startup is intentionally non-Elementary-branded, better enabling OEM/Vendor branded startup splash screen — "we don’t need to constantly advertise your operating system to you".
There's a separate blog post on hardware-specific improvements here: https://blog.elementary.io/hardware-improvements-coming-to-e...
Dang, I really like their logo and it sorta feels nice to see it on bootup. I'm sure there is some way to bring it back though, and it's a tiny thing that, like they said, will probably be far outweighed by the benefit it brings to OEMs.
Maybe the awkward stretched out time zone map is gone too?
Having recently begun using a Mac for the first time, and after having had to suffer with Win10 for a few years, I have to say that I didn't appreciate that several out-of-box linux desktop experiences are now superior to commercial options.
Ubuntu for example is quite stable, has attractive defaults that aren't garish or trying to be too unfamiliar, and otherwise never does anything I don't expect. Especially when it comes to finder/explorer/nautilus.
All major desktops now have a dock, some tray icons, and a notification area.
To me the differentiator is just staying out of my way.
I installed it on one machine, and the missing "minimize" button drove me mad, otherwise it's a very nice environment (I generally use Ubuntu LTS).
You expect typing letters while having the file explorer window focused to launch a full recursive file search?
My workflow is based on two main hotkeys plus some extras:
1. Alt+Tab to change between applications. Set it up so it ONLY switches between applications (never between windows of the same application). If you keep alt pressed the icons for the open applications stay on screen, and you can click on them to go straight to that app.
2. Alt+` (the key below "esc"). This rotates between open windows of the same application. It is basically alt+tab but for windows of the same app (that you have focused).
Extras:
- Window-resizing keywords (make the current window occupy the left half of the screen, maximize it, etc.)
- Tabs on some applications (like the browser or vscode). I still use several windows though.
- Disable all animations. Just make things appear/disappear as fast as they can. It sounds silly but it really enables faster "ops, not that window, switch again" when you don't nail it the first time.
This workflow is ingrained in my brain already. I somehow seem to mentally (without noticing) keep track of the window and app switching stacks, so most of the times I know how many times I have to hit tab or ` to reach the window I want.
In cases of transient windows (reference materials, docs I'm reviewing, internal wiki pages, screenshots) I just let them pile up willy-nilly on my desktop. When I need to find a specific window that is no longer visible I either tab through the application switcher or select it from the multitasking view.
It may sound like a "messy" approach to window management, but finding an open window using an app switcher or multitasking view seems neither more nor less efficient than finding that same window minimized in a taskbar/dock.
I've never understood Gnome.
Or just install Elementary Tweaks.
It's one of those little functionality gaps that keeps people on closed/non-free systems. Is it at least on their roadmap? I'd gladly donate/pay just to get a Mac-like minimize/hide experience.
What's the connection between the dock and minimizing windows? I use a WM with a dock and that supports minimized windows just fine.
Personally I put it back to "Windows" layout (min/max/close on the right), even though I mostly use Super+H.
- Their 70/30 split is higher than what Apple currently offers (85/15) for revenue less than 1m (which is going to be everyone on eOS)
- While I like macOS, I think it's a shame they chose to (more or less) copy it, rather than try a new direction. But perhaps that would be too risky.
It's perfectly possible to get money _and_ write open-source software.
This is the problem with your argument. You're looking at it from the developer's perspective, rather than the user's. As the developer you have this choice, and you know which choice you've made. As a user, all I have is your word (which is worth nothing if I don't already have a good reason to trust you).
As a user, source code access is important, for a number of reasons:
1. It gives both users and third parties a way to verify your claims
2. As it turns out, transparency tends to discourage bad behaviour in the first place
3. Users can fork if you no longer maintain the project according to their standards (abandonment, price hikes, bundling malware, and so on)
None of those reasons are direct improvements for you as a developer, but they're a big part of why I am reluctant to adopt proprietary software (free or paid).
> How are movies a counterpoint? Movies and streaming services are paid products.
Yes? I am less reluctant to pay for a movie because they don't demand the same privileges.
> In fact if you are able to make a living through selling software you have less incentive to collect user data and sell it as a revenue stream. Most products that are free are those where the your data is being sold for is exactly that reason.
On the other hand, reality shows that many (most) commercial developers are perfectly happy to pick both.
The developers need to make the apps, otherwise there's no marketplace. Gotta satisfy both, albeit imperfectly.
> As a user, all I have is your word (which is worth nothing if I don't already have a good reason to trust you).
That's not true at all. Experts can audit closed-source software to see if it's phoning home, etc. An App Store (like Apple's) can have strict rules where developers will be banned if they do malicious things. You can install tools to monitor apps. Apps are sandboxed to prevent access. You have much more to go on than just the developer's word.
> As a user, source code access is important, for a number of reasons
Sure, it would be nice as a user to have the code. Most app developers simply won't give it to you.
Something matching the quality of Excel/Final Cut/Photoshop/90% of games/(lots of my favorite misc. iPad/Mac apps) does not exist in the FOSS world because they have been built on millions of hours of work (and not only by developers, but people spanning many industries and skillsets), which needs payment. In the only FOSS model, huge classes of software (e.g a AAA game) simply won't exist because they will be economically impossible. There's TONS of software requiring cross-domain teams to build, that is unglamorous and unfun to build/maintain, and those kinds of things don't get built to free. This whole purist FOSS model dates back to a very different time that honestly makes no sense today.
In fact by your argument, we can extend to many other things that should be completely open source and free. There should be no need for any intellectual property - the proprietary ingredients and manufacturing processes of Coke (and 80% of everything else in a grocery store) should be free (how can I trust whats in it otherwise?), the complete schematics of every car and airplane should be open (how will I know its safe otherwise?), the innards of every piece of electronic that could conceivably contain a mic/speaker/GPS should be free (what if they track me?). You think this would lead to a utopia of free and awesome stuff. I think it would lead to 90% of that stuff never being built in the first place.
I guess you are one of those purists who somehow uses ONLY FOSS software and is satisfied with it so I won't argue further. To me and many others the quality and breadth of commercial software is orders of magnitude better than free software, and we recognize that's because we pay people to work on it, even if that work is not fun enough to do for free.
you can have for-pay open source apps, no problem
> but if it's a consumer product...then it isn't going anywhere as a business
if you live/sell in a country that respects software copyright/license, then you can sue them or get a cease and desist for using it against your license (say prohibiting commercial derivative or violating your copyright without authorization)
and your software, if popular or lucrative, its going to get ripped off anyways like in the real app store (just look at the many clones of many apps), or just plain pirated like the myriad apps (photoshop, word et all) have since basically forever
am i way off-base?
(I know that the dev environment is quite different)
If anything, the Linux community seems to cite it as a strength.
I honestly don't think these are the best of Apple's stuff. When I think about these features I can't really see why they decided they were important enough to implement. Perhaps they're trying to generate some excitement about the desktop environment I don't know.
It's a bit insulting, really.
My favourite feature is that it's all flatpak, allowing smartphone-like permissions for each desktop app — both first-party and any third-party from Flathub.
I like it way more than snap and am glad there's an Ubuntu fork that works so wonderfully with it.
There's also Flatseal (https://flathub.org/apps/details/com.github.tchx84.Flatseal) as a user-friendly way of managing permissions for flatpak apps. This version of elementary OS comes with its own permission manager, but with my limited experience with both, it seems like it allows tweaking less permissions than Flatseal.
Everything in that reference basically says "yep the criticism is valid" and in a few cases the author expands on that with "but it's ok because..." and then lists various reasons why it's fine that it's still a problem but that's it's being worked on. They also had to correct a good chunk of just outright incorrect information they were supplying mid-post about system updates.
One item was addressed, which was that flatpak now notifies the user that sandbox escapes are possible based on the app's configuration.
As a response to the criticisms, it's not a great one.
Consider also, there are better ways that some of these issues could have been tackled. Why not have flatpak prompt for permissions as they're used, e.g.: "This app wants to open your home directory" rather than at install-time. That would make it abundantly clearer to the end users this is aimed at.
I'm going out on a limb and guessing that it is because that's what is done in mobile, and it's usage in mobile has re-taught everyone the lesson that constant dialog prompts just train users to click through dialog prompts without reading them. That's bad, and annoying.
I do still think there are better solutions to that problem, but it would require more effort from users to get applications working if they weren't designed with sandboxing in mind, which is the vast majority of applications, which in turn means that Flatpak probably wouldn't have grown as quickly.
As it stands though, flatpak out of the box has all the security issues of running old unpatched systems in order to mostly have compatible runtime environments, which, in my experience, don't actually buy me that much. The few times my distro hasn't already shipped a copy of an application, AppImage, Flatpak, or Snap haven't had the solution either.
This entire experiment we're doing with "ship the developer's box" as the new standard of software delivery and the different warring philosophies employed to turn that into a reality are interesting. My money is on the least secure, least safe, least functional, but best marketed thing winning out.
They essentially deny every such permission request and for the few they actually care about (getting notifications from 2 or 3 of the 50 apps that want notification access) they come and asked for assistance.
One of the nice things about the iOS ecosystem is that apps aren’t allowed to be nonfunctional if you deny them access to something.
The end result being that you can't get everything in any one of them, and often can't get a particular application in any of them.
Plus I find the general sandbox approach to be a LOT more stable - distro upgrades rarely causes issues, and dependencies are much easier to wrangle.
Definitely still some hiccups, but overall it feels like it's moving in a direction that I find easier to work with - both as a user and a developer.
Plus the markets are generally fairly large - sure I still dip into aur every now and then, but I mostly find the "Software I need at work" stuff available
I think that speaks to a particularly huge failure of Linux Desktop more than anything else.
When it comes to platforms[0], you need to be able to depend on certain things being there and working a certain way. Fragmentation of a platform is bad, and Linux Desktop is so fragmented the various distros style themselves as entirely different OSs!
[0] As opposed to applications. Part of the problem is that historically there has never been a clear delineation in Linux Desktop.
From my experience, most of the popular proprietary ones (think: Slack, Discord, Spotify, whatever) are distributed both as a flatpak and as a snap, while more niche desktop apps are usually flatpak-exclusive (out of those options).
Perhaps you just don't use that many things that aren't really popular or well known? I can't find PuTTY and KeePass (though obviously there are alternatives in those cases), applications that both have Linux versions and that I use every day (and aren't even that unknown).
Also, unfortunately the limitations of Flatpak mean even if what you want is there, say Wireshark, it often won't have full functionality because it is impossible to give it the permissions it needs. In Wireshark's case, it can't actually capture any packets. If you want to actually use it you'll still need to resort to a different installation method.
That being said, unless the flatpak has some features your distro version doesn't, for trusted software package is always preferred.
Flatpak is really for untrusted software/proprietary and projects that update with new features often.
For me this was a much bigger issue when I was on Windows, but also in Ubuntu which groups programs together similarly to the Windows taskbar. I'm on i3 now mostly and just use a normal terminal for ssh, but for quite some time I was using putty on linux specifically for window management.
It's been some time since I last read anything like that from you. Would you mind to comment if ebpf usage and the possible new futex2 syscall helps to close the gap between linux and NT with regards to async primitives?
Ubuntu LTS base is also great for developers, as it seems most cloud images these days are just that. I've tried installing/building tools on other distros that I like more than Ubuntu but they always seem to use wonky options, libraries that are too old/new or in strange places, and I end up back at Ubuntu.
Their own apps similarly look ok at first, and then you go to use them and you realize it’s all very bare bones.
My system also froze/locked up a lot at random, and I don’t know why. Installing Ubuntu fixed that issue. Not sure what happened there.
Again I don’t blame them; not sure how many people are working for them, but creating a modern desktop and all the apps you’d expect from scratch is a huge undertaking. I hope they do well and provide Linux a real alternative experience. I’d love to see some way that they can extend into apps to make their look and feel more consistent somehow (gnome/kde skins? Something much more? Their own forks of popular apps?), and that the community ends up focusing around them so we get a distributed effort.
That's a weird criticism. How many competing UI frameworks with completely different look'n'feel are there in a default win10 install ? How many control panels and file dialogs ?
There is more value in design than consistency on production oriented systems. It seems the contrary is preferred for mobile.
Mobile devices are by and large meant for and used as consumption tools. So consistency is desired there.
Or perhaps better, use a windows theme for a linux desktop - and stick to gtk apps. https://github.com/grassmunk/Chicago95
The dashboard feature is totally underrated. What would be even better would be if you could actually put any window in there.
Or if any window could be made semi transparent and float on top. And have that entire layer be toggleable with transitions....
It's total shame that they've removed it entirely.
Interestingly, Microsoft Apps are the largest offenders, historically. They like to ship them with new versions of their common controls library, so it may take a while until you see changes in the OS itself (the ribbon being a large offender).
However, things work similarly. All the keyboard shortcuts work (unless they messed up on purpose), screen readers work, most settings work across the board (even if it makes the app look even worse). Change the color scheme, everything changes. In most cases, you can use the same API functions to interact with the controls, and WM messages tend to work consistently.
Not so with linux GUI toolkits. There has been a lot of effort in trying to make them look the same, but it's pretty obvious when an app is written in GTK vs QT or TK. Even the way they respond to sizing events change. Widgets work totally differently, keyboard shortcuts and conventions change.
It's less of a problem than it used to be, but it is still there, and it is sometimes very jarring. I don't know how to fix it, but it has to be fixed at some point.
I’m really comparing to macOS, which clearly is what eOS is trying to be.
If there are parallel realities and alternate timelines, then somewhere out there is the one where everyone settled on GNUStep/WindowMaker in the 90s and Linux took over the Desktop by 2005.
Sadly, I can run neither Excel nor PowerPoint under Gnome. So I'm stuck with windows.
They do their own apps. It's actually one of their most frequent criticism - they recreate the wheel in building their own apps instead of focusing on the OS. And they do it for consistency, which is also goes against your other point.
On linux, there is no paradigm to stray from and there's no consistent design language or UX because of the fractured nature of its ecosystem. There are incredibly well made applications for KDE, GTK, and Electron, all with their own ideas of what UX should be and what UI it should look like.
The only difference with macOS is the number of apps that make at least a minimal effort to follow the OS paradigm.
(We might be saying the same thing!)
> all AppCenter apps are now packaged and distributed as Flatpaks, a modern container format that keeps apps siloed away from each other—and your sensitive data
All apps are siloed from your sensitive data? That is a little um simplified or misleading or something. They may have done a great job with security now that they have Portals, but it is surely a much more complicated story than the above quote suggests.
Has that changed recently?
1: https://linuxize.com/post/how-to-install-xrdp-on-ubuntu-20-0...
On Linux, interaction is overwhelmingly through SSH. Or a client/server app.
Of course, the canonical answer to that is remote Xorg. It needs a high performance connection and X running on both sides, but it works. And the most Windows-like answer is VNC.
Also, I am not sure what you mean “jumpbox”, but it sounds like you don’t use the remote machine locally.
Performance seems okay for me as long as you don't use fancy backgrounds. Not as good as native windows RDP, but definitely usable.
These are secured jumpboxes in our lab/production environments we RDP (or SSH) into via the VPN, but we used to connect to them from the office LAN back in the days when we actually worked in an office.
I've been using elementary OS since. I've tried some other distros, but those experiences usually ended up with me trying to recreate elementary feeling, failing, and then returning back to elementary OS a couple of weeks later.
Before then only the obscure Wayfire had a similar gesture for desktop switching (actually good) and there were a few projects hacking it up with simulated keyboard shortcuts (obviously not the same at all).
I always wondered how an open-source project develops from scratch so many different apps (e.g. Web, Mail, Calendar, etc). Why not focusing more on the OS rather than wasting resources on apps that there are mature open-source alternatives (e.g. Thunderbird, Firefox, etc)?
One aspect of it I feel often goes underappreciated is how lightweight it is. Provided you have a reasonable amount of RAM for the tasks you're trying to do, e.g. 4GB+ for normal desktop usage and an SSD, you can run it on fairly old CPUs and it will be blazing fast still.
Congrats to the team!
VM's partly solves the problem, but You need to have a decent computer and small technical chops.
I think each OS release must include data about the performance (Speed and resources at steady state).
We want performant and lean OS's.
You and I survived the trials amd now wamt performance. I personally will lilely never use elementary, but you and I are not the target audience.
https://en.wikipedia.org/wiki/Drangarnir
https://visitvagar.fo/welcome-to-vagar/activities/hiking/dra...
[1]: https://github.com/elementary/houston/wiki/Submission-Proces...
1) My personal desktop has Windows for games
2) My laptops (personal and work) are Macs because Apple simply makes the best laptops
When I've dual-booted in the past it's been a huge pain, and I didn't end up bothering to switch to Linux very often in practice anyway. Linux laptops exist, but they tend to be spotty in terms of build-quality and power. Gaming on Linux can technically be done, but I don't want my expensive machine to be unable to play some things because of the OS.
There are lots of little things, most of which could be overcome with effort, but these days I'd rather use my devices than tinker with them most of the time.
I guess I just wish I had an excuse to use something like Elementary or PopOS. I would if I had an a) desktop that b) was mainly for projects. But alas, I don't.
Valve has been doing amazing things in this space, but for now my understanding is that it's more of a "most games will work pretty well" scenario, which is not really what I want for my primary games machine.
As another comment points out, its called VFIO. From that subreddit: "VFIO stands for Virtual Function I/O. VFIO is a device driver that is used to assign devices to virtual machines. One of the most common uses of vfio is setting up a virtual machine with full access to a dedicated GPU. This enables near-bare-metal gaming performance in a Windows VM, offering a great alternative to dual-booting"
I haven't worked in windows space in awhile though, and state changes rapidly, so there may be other options now that utilize a single GPU across host and vm without losing performance.
I already tried to migrate to Pop_OS! but it failed to install.
So I had to keep with Ubuntu.
> The primary motivation for this change is the eventual move to Wayland (a protocol that will make your device faster and safer), but this also gives us more creative control over these features.
https://blog.elementary.io/platform-changes-in-elementary-os...
maybe just me but i felt a lower cognitive load compared to big sur, and especially monterey where everything is super low contrast, large margins between items, and all the icons are hard to differentiate same-color outlines...
looks nice!
There's far fewer flat icon in its design, the shortcut hints are styled in a completely different manner, there's no global menu bar (not that they have a choice with GTK+), and the top bar widget actions are more complex than the ones you see in macOS
Those sort of things would make it a "UX alike" instead of only "look and feel"
I just tried this in the previous Elementary OS version and, yes, it works.
(Tested using "Files" & "Terminal" applications.)
It also shows a little green "+" circle hint (i.e. copy) when the file icon is dragged over the Terminal window.
Running Linux natively on Apple Silicon is not ready for daily driving.
It's the Isis drama they invented all over again.
however, i'm still gonna stick to XFCE4