> While speed may be the single most important element of a UI, there is plenty about GNOME that is going to disappoint Unity users. The biggest gripe I have is keyboard shortcuts. Unity had them in spades. For as graphical as Unity is/was, it was also very easy to drive without taking your hands off the keyboard. GNOME lacks that level of shortcuts.
Can we please fix this before the convergence? Muscle memory and Linux go together like rosemary and garlic.
+1 although an Ubuntu user since it's very first release, i3wm has been my go-to productivity DE for the last 5 years or so. Plus it's multiple-monitor support is superb, much more useful than Unity's
Indeed, I have a dual monitor setup (horizontal + vertical) and it's been a joy to finally have independent control of each one of them. Hated that Unity treated them as a single desktop.
xfce (Xubuntu) hasn't unexpectedly changed on me in about 10 years. YMMV but I haven't been surprised once and haven't seen any features in other DEs I wish it had.
Relatedly: making "safe" keybinding choices as an application developer has been a problem for a long time. Especially when compiz was the new hotness, many application shortcuts needed to be rebound by the end user, because the authors had chosen ones that were now in conflict, especially ones involving Ctrl and Alt together.
Before that, Mozilla had gone through something with the Mozilla Suite, where seemingly every contributor wanted to ship their new feature with a keybinding for it, which meant there was an internal friction. But product leads and code reviewers were also burdened with respecting platform conventions, too. Eventually, there came a separate approval process for any proposed shortcut, with specific people tasked with managing allocations (and reclaiming past ones—this project management approach is what eventually led to Firefox).
The headache persists, because there tends to be no pattern for the assignment of shortcuts. When Unity was introduced, I'd hoped that they'd seize the opportunity of a clean break to reconcile problem, but that didn't happen.
Here is the scheme I've adopted:
1. If we pull back our scope a bit from the application, then operations on the window itself are performed with Super. (It helps that this key usually has Windows iconography on it).
2. Finally, if we pull all the way back, operations to be handled by the environment/desktop/graphical shell/what-have-you are performed with Alt+Super.
The implication of 1 is that to move a window around without having to drag its title bar, you'd use Super along with a mousedown anywhere in the window, rather than Alt+mousedown. It also means that the shortcut to close the current window is not Alt+F4, but instead it can be remapped to Super+F4. Same goes with Alt+Tab, which becomes Super+Tab. This is jarring at first and perhaps controversial. But even when I wanted to abandon my own scheme because of how bothersome this stuff felt at first, I found that after one day of use, I had become accustomed to it. So switching is really not as bothersome as kneejerk might have you believe.
Flipping between virtual desktops is Alt+Super+arrow rather than Ctrl+Alt+arrow (or other combinations that might be in use) and revealing the desktop by hiding all windows is Alt+Super+D. I have global launchers set up like this, too. E.g., Alt+Super+T brings up a new GNOME Terminal window.
What this means is that any combination of Ctrl or Alt or Ctrl+Alt (along with Ctrl+Shift, Alt+Shift, and Ctrl+Alt+Shift) in the keybindspace are reserved for use by the application. Go wild.
I put this scheme into place once whenever I buy a new machine or otherwise do a new system install, and then never have to worry about it again. The important thing to note that it entirely comes down to remapping the default "global" shortcuts so that the desktop/WM is not stomping on the application keybindspace. This is very easy—it doesn't involve trawling through dozens of apps' settings panels or config files so they comply; you remap the "global" desktop/WM shortcuts and you're done.
The result of adopting a scheme like this is zero conflicts, ever, and you end up feeling like you have more space. It might sound so cumbersome as to be unworkable, but don't underestimate the advantages of what a clean break, once-and-for-all decision like this brings around.
3pt14159 put it well: There are more than a dozen really useful keyboard shortcuts for UI management, and tiling your windows. I mean, it's powerful - maybe one of the most standout features of Unity. You know what is really the icing on it? Hold down the "Super" key, and you get a menu showing all the UI keyboard shortcuts. It's a quick modal - no Alt-F1 help dialogs to navigate and close; no link to a website; a handy cheatsheet instantly accessible and dismissed.
I'll give GNOME a try, but everything I hear is that it's a directionless blob that makes productivity difficult.
In 2013, when I have first chosen gnome, it was to replace the PC of my old parents. I have contributed to a gnome extension (translated in french) https://bitbucket.org/LukasKnuth/backslide to make it looks like their previous computer on windows XP.
AFAIK, ubuntu gnome is very stripped down by default, but it is easy to tailor to fit many needs. The extensions are often written in javascript.
The trouble is, that different people have different ideas what 'make OS work' means. An essential feature of yours is other's bloat. Hence extensions.
The one big one for me is Super+number to jump to a window (super+3 to switch to the third window on the toolbar). Both Windows and Unity support this, but there's no way to get it on most other linux DEs (with Cinnamon there are plugins but they are really buggy). Not sure about gnome though.
Alt-` cycles through the windows within one application, such as terminal or Chrome. Similar to what can be done on the Mac. I use it & hope Gnome can do the same.
It's exactly the same in Gnome. In fact, Gnome is better in that you can activate an extension which will change the behaviour of alt-tab so that it cycles through all windows.
I feel like I'm being asked to list all the shortcuts I use, which seems impossible and I've never used Gnome, so I don't know what it's missing. But off the top of my head the must haves are:
1. Open a new terminal and max screen it with focus in the prompt, not on the "window".
2. Move windows between workspaces.
3. Screenshot with copy to clipboard.
4. Disable "quit all" for applications on an application-by-application basis (I never want to nuke chrome, but I often want to close all finder windows).
5. When plugged into a monitor, move active application to the monitor and have it show up correctly. If it was maximized or full screen, make it maximized or full screen in the new context and make sure to handle different pixel resolutions correctly.
6. Fullscreen toggle.
7. Maxscreen toggle.
8. Paste from clipboard without formatting.
9. Make capslock an additional control.
10. Remap individual control keys and have them work in all contexts and in all ways resemble those keys (something that Unity doesn't do which drives me insane in Slack in the browser where I can't always copy and paste because Slack somehow knows capslock isn't a "real" control).
I never understood why Canonical's attempts to be "brave" (as Apple would call it) result in Canonical pulling out, while attempts by other companies like Apple don't.
For example, Canonical made a very brave decision to switch to Unity, and now they're having to move away from Unity, which hints that there were enough users against it, and they pushed Canonical into this position (along with Canonical trying to allocate resources better).
But Apple "killed skeuomorphism" and there were tons of people complaining "MY NOTES APP WHITE INSTEAD OF YELLOW EWWW" (especially my parents), yet today they're still getting more users and they have so many on recent versions of their OSes. I supported the removal of skeuomorphism, but seeing so many around me and on the internet complain, I thought Apple would be forced to put it back everywhere. Instead, people just moved on.
When Apple makes a "pro" laptop without any USB A ports, people laugh, but then buy it. When Ubuntu makes a phone, Ubuntu regrets it. Why?
Apple's user base is one that wants Apple to do all their thinking for them as far as computer decisions are concerned. Often this may lead to sub optimal solutions for them, but they are ok with this because the overall lack of worry with Apple products, and the fact that they have a good track record of making good decisions, still leaves them ahead in the net (this may be changing with pro Macs though).
Linux users, OTOH, have an opinion on everything. Worse, a vocal minority is simply anti anything that is popular, which means Ubuntu's goals were entirely in conflict with this segment of their base.
Let me rephrase it in less condescending terms (or, more accurately, /different/ condescending terms):
Apple users want to get shit done, instead of compiling custom kernels to fix font rendering issues when using Wifi while headphones are plugged in.
"Pro user" means, for Apple, s/o using a computer to create something. It is not someone who's a professional at using computers, i. e. constantly optimising their 20+ USB gadgets and swapping hardware components.
Agreed here. I have compiled probably a half dozen linux kernels in the last 15 years, and especially in college, tried many flavors of linux. So I've historically not been intimidated by tinkering.
But now that I have a career that relies on getting stuff done, and don't have time to play with kernel modules and window managers just to get my computer running in a stable way. I have zero patience with the clusterf*ck that seems to still be the Linux desktop.
I mean, look at this thread itself. When I see I hear all kinds of things like Wayland, Xwfm4, Compton, Unity, Gnome (the only one I recognize), multi monitor issues, tearing? Who has the time these days to deal with all that, unless it's part of their jobs, they're curious college kids, or they're kernel hackers?
I mean, geez, it's 2017. Apple (and MS to an extent) figured these issues out for the most part more than a decade ago. Yea yea I get it, drivers, blah blah. But I simply don't have time anymore to futz with this stuff. I have to get things done and need a reliable machine. And my clients don't pay me to fiddle with my kernel to get nice animations.
It's the same reason I don't build a Hackintosh, although I'm really tempted. SO this isn't an "I love apple" type thing – it's an "I want to get work done and not stare at a boot loader" thing.
That may be your experience, but it's not mine. I've been using Ubuntu since 2006 or so and I have never had to do the things you mentioned or had any issues. I don't even know how to mess with kernel modules or window managers.
My experience with Apple has been far from "Getting things done". Just to name a few issues:
* I had a screen saver (one of the stock ones) turned on and it would cause my fan to go crazy
* It still doesn't always sleep, I sometimes find it running low on battery in the morning because an application woke it up
* I have had to use the Apple store which requires a phone number, credit card, etc to just get XCode
* Maximize doesn't seem to always work
* I can't keep a window always on the top
* Changing the password for wifi sucks
* If I change my system password, for some reason I always get these annoying and constant popups saying that so-and-so wants access to something, which only works on the old password
* Copying and pasting does weird stuff, adds in a bunch of crap that I don't want often
* Updating my OS takes forever and does whatever it wants without giving me much insight into what is going on
For the record, since macOS 10.12.4, I get to enjoy tearing on a rMBP13, on both built-in and external display. So no, this stuff was not being figured out 10 years ago.
By the same token, the last time I compiled a Linux kernel, it was 2001 in calendar, or something like that. Since then, the distribution provided ones are fine. Even if it means running 3.10 in 2017. It works, it's fine.
I think both groups want to get shit done. It's more a question of what each group can tolerate. Apple users can tolerate a lack of choice. Linux users can tolerate a constant struggle with flaky device drivers.
There was a lovely TV show from the UK back in the 1980's called Yes Prime Minister - one of the frequent exchanges between the senior public servant, and the PM, when the former wanted the latter to back down on something, was for the phrase 'That's very brave of you ...' to be uttered. Brave does not necessarily imply profound, good, worthwhile, etc. Canonical's 'brave decision' to go out and create their own DE (despite two very competent options already existing) does not imply it was a good decision.
> When Apple makes a "pro" laptop without any USB A ports, people laugh, but then buy it.
This is the classic mixup of 'people'. The people that scoffed at a laptop with a single USB port (or a phone without a headphone jack, etc etc) are NOT the same people that bought those items.
> When Ubuntu makes a phone, Ubuntu regrets it.
The proverbial 800lb gorilla made a phone, and also regretted it. There are plenty of propositions explaining why two mobile phone platforms were all that the market could sustain.
> which hints that there were enough users against it
I'm pretty sure those users are a vocal minority.
Personally I loved Unity. The only thing I disliked about Unity and Ubuntu's UI in general is the lack of configuration options, having to install extra buggy configuration managers to do really basic stuff, like turning Caps Lock into Ctrl. But that's a problem with the whole Gnome ecosystem and not Unity.
And Unity I loved, because it has pretty good keyboard shortcuts, is space efficient and generally worked for me just fine out of the box.
Going back to why Canonical dropped it, I'm pretty sure that's because they are giving up on the Linux desktop in general.
As for the comparison with Apple, dude, you're comparing David and Goliath, except this is the real world.
Going back to why Canonical dropped it, I'm pretty sure that's because they are giving up on the Linux desktop in general.
Indeed, Canonical had the same realization as Red Hat ~2000-2003: there is no money in the Linux desktop, there is a lot of money in enterprise services and support.
Since Canonical has to make a profit, it is smart of them to drop unprofitable projects such as the phone.
And now Ubuntu has passed Red Hat in the # of servers metric because devs like me choose their Linux server OS based on familiarity with the Ubuntu desktop OS.
On the other hand, Red Hat's revenue is steadily growing[1] and there are very few RHEL desktop users. Fedora is probably also less popular than Ubuntu and its derivatives.
Interesting question. Some data points that I could find.
Canonical: 2009: ~30M, 2013: 65.7M, 2016: 103.3M
Red Hat: 2013: ~1.4B, 2015: ~1.8B, 2016: ~2.1B, 2017: 2.4B
Consequently: Canonical 2013-2016: ~1.57x, Red Hat 2013-2016: 1.71x.
So, their revenue seems to be increasing at a slower rate and Red Hat's revenue is almost twentyfold that of Canonical. Moreover, Red Hat is operating on a quite large profit, while Canonical is on a loss.
Apple/ios users don't really have a choice of hardware vendor. They're locked in, so they're stuck with what apple serves up or can feel the pain of migration (I'm in that boat).
Linux users are more fickle and can and will move to different distributions if they don't like one particular one.
Canonical needs to listen its users in a way that apple doesn't.
To be fair, we haven't seen what these changes Apple has made are going to do long term. Its clear there is some worry over there (the mac pro 2018 preview is very unlike apple)
As soon as Canonical blows something off in Ubuntu, users can switch to whatever other Linux distribution they like (or remove/replace the misbehaving component, like it happened with the Amazon thing or early-ish versions of Unity). Apple users can't very easily switch to a better version of macOS (especially when they need to migrate their data out of the walled garden it's hosted in).
(Yes, you CAN run Linux on a Mac... but the experience isn't great. As far as I know, Linux still doesn't have support for suspend/hibernate when you close your MacBook Pro's cover. And although the Macs have the nicest touchpads, it seems weird to pay a 50% "Apple tax" if you're not even using Apple's OS. You lose half the benefits that Mac people buy into. So if you're a Linux person, then you might as well be using a PC.)
Since most Apple customer's are using Apple's OS, when they don't like a change, their options are:
(1) to switch to PC's, or
(2) to hold off on upgrades, but eventually get on board with the changes begrudgingly.
I've been seeing more #1's lately, but most fall into the #2 category.
In the Linux world, however, there are hundreds of distro alternatives. And half of them are Ubuntu-based, anyway. Don't like the UI/UX choices that stock Ubuntu is shoving down your throat? In an hour you can switch to Xubuntu, Kubuntu, Lubuntu, or even Mint... and still retain compatibility with the same APT sources. Don't like systemd, or something else deeper under the covers? Then it's a little more effort to switch to a completely different distro family, but not MUCH effort.
Open-source Linux distros just don't have the lock-in leverage to muscle their users around the way that propriety OS's do. By design.
Admittedly, when Apple "killed skeumorphism" it had already been killed by most of their rivals some significant time before. Microsoft already had Metro; Google had flat, skeumorphic-free designs for a long time but never had made a big fuss over it (until that Duarte android thing got big). Ubuntu's Unity situation is rather different.
I would think it would be due to the sheer power differences of each organization. While I'm sure Canonical isn't strapped for cash, it's clear Apple is far larger of a company. The fact that Apple has built its user following, along with the huge number of nearly religious followers, against Ubuntu's (or Linux's, really) much smaller audience gives concessions to Apple to make moves such as this. If I had hundreds of billions in cash as a company, I'd feel much more comfortable making "deal with it" decisions affecting my users than if I were running a much smaller operation.
The fact that Apple has built its user following, along with the huge number of nearly religious followers, against Ubuntu's (or Linux's, really) much smaller audience gives concessions to Apple to make moves such as this.
That's too easy. Linux also has a contingent of nearly religious free software fanatics. Moreover, Apple is now strongly represented in niches that Linux was traditionally strong in, such as academia and developers.
Let's not forget that Apple was in dire straits in 1997 and was only a tiny fraction of its current worth. It has only been the behemoth it is now since 2010 or so. The difference with Apple and Canonical is that when Apple introduced products they were often miles ahead of the rest, while Canonical is mostly catching up.
OS X provided a UNIX with a modern display server and good usability in 2001. The first version of Ubuntu was released three years later and had a lot of shortcomings that were only solved later or just now (such as a modern display server).
Apple defined the modern smartphone in 2007. The first preview of Ubuntu Phone was released six years later.
Apple is now extremely large, but 2001 Apple and even 2007 Apple was not.
I'd feel much more comfortable making "deal with it" decisions affecting my users than if I were running a much smaller operation.
Except that Apple canceled unsuccessful products as well. Ping, iTools, .Mac, MobileMe, Aperture, Xserve.
I think Canonical is attempting to identify other investors, and the 'due diligence' on that resulted in a desire to focus on server/container type activity and to pull out of projects that have no obvious commercial value in the medium term.
When you're making controversial decisions, it tends to help to be right.
But Canonical (and every other distro) are in a somewhat difficult position. UI isn't just an incredible hard problem. There's also a dearth of talent because the relevant people aren't a natural part of the OSS community.
And, most importantly, UI suffers tremendously from the bazaar-model of development. It needs someone with near-dictatorial powers, and excellent taste, who can control the complete user experience from init to shutdown.
Developing Unity is a cost center. Presumably there's not much revenue from developing it. Presumably the money being spent on it is better spent on server side where people do pay Canonical money for support.
Microsoft is not even in a position to think that way. They are constantly doing ass-backwards things that their enterprise and government customers want. So many of the messed up decision with Active Directory are because some upper manager that didn't know what was going on demanded a feature or threatened to switch away from microsoft.
I'm not sure it's that simple. For example, the people I knew who hated the removal of skeuomorphism didn't get over it when they watched WWDC, or some TV commercial. They just kept using their devices, and waited a long time before upgrading. But they then upgraded and shut up, and not because they saw a commercial.
It not that they were convinced the new look was better. It's just that no sane individual would stick to a iPhone from two generations ago just because he doesn't like his notes' background color.
As someone who uses Ubuntu on both desktop and server here are my concerns about Ubuntu and Canonical at this stage. I think Ubuntu is awesome and has done huge things for Linux, particularly on the desktop.
A) Canonical's R&D investment is shifting from desktop components to server components. On the one hand I'm sure the server version will get even better which is great. But it's hard to imagine how this could translate to a win for Ubuntu desktop users.
B) Bug #1 (https://bugs.launchpad.net/ubuntu/+bug/1) was closed a while back with a rationale that I thought was redefining the truth a little bit. Canonical's new direction probably makes it even less likely that bug #1 will ever be reopened. Again hard to see how this translates to a win for Ubuntu desktop.
C) I know that at some point Ubuntu is moving to Wayland as well. I am concerned that this change will break things that are important to me -- for instance I use xmonad and it doesn't work under Wayland.
I get 100% that Canonical needs to make money, and it seems no one's ever figured out how to do that with desktop Linux unfortunately. I just can't help but feel we are losing the biggest champion for the Linux desktop here and am concerned about its future.
If things get bad enough (for instance if they move to Wayland and there's still no xmonad port) I would probably switch to Bunsen Labs or some other form of Debian but I would miss losing the confidence Ubuntu gave me that things would just work. Even if I do have to sign off and leave Ubuntu I will do it with a huge /salute in their direction for having created the OS that was good enough to enable me to make the switch to a free desktop.
I hope Canonical succeeds with their new focus. But I have a feeling an opportunity is going to open up for someone else to become the #1 desktop distro.
I believe that the reason that Ubuntu is winning the server market is because it won the desktop market. Ubuntu is on most production servers because devs are more familiar with Ubuntu than alternatives
since they use Ubuntu on their desktop. That's certainly the reason why I use Ubuntu or Debian on all the production servers I've spun up.
I'm sure Ubuntu is well aware of this issue, and of course they're also dog-fooding Ubuntu Desktop, so while progress may slow, I highly doubt it will stop.
In fact it might just speed up, since for the last few years they've been focusing their efforts on Unity 8. Now even though they'll be spending less resources on the desktop, the resources they do spend won't be disappearing into an unreleased project like Mir & Unity 8.
You're lucky. Until very recently, I had to work on CentOS 5 which was very frustrating. To make it more bearable, I compiled from source and installed modern versions of non-server software for personal use, e.g., Vim, GNU screen, Bash, etc. At home, I'm happy running Lubuntu on my laptop.
Indeed. Look at Microsoft. While the home desktop is not their primary focus, they still defend their position there as it is a pitch point when selling their products to corporations.
This with the argument that as potential employees are already familiar with Microsoft products from home use, they do not require as much up front training.
For giggles, I put Fedora on a Macbook Pro. Wayland + Gnome works surprisingly well, and I can see the appeal for some users. But I'm right there with you on XMonad.
I'm not wed to XMonad or any of the extensions, really (for example mostly vanilla dwm, which XMonad was based on, works fine). But I am wed to the keyboard shortcuts and basic functionality of XMonad + dmenu. I'd probably be very happy if there were an XMonad style tiling option in Wayland Gnome.
But, I can't help but feel a little weird about some of the free *nixen moving away from X11. I don't see Wayland running on OpenBSD any time soon, for example. I worry that this will be yet another thing that leaves the old world of diverse, but standards-adhering Unixes behind. I'm genuinely not even sure whether that's a bad thing, though my gut reaction is that a Linux monoculture isn't what I'd want.
Canonical/Ubuntu has practically stopped innovating for the desktop. There are a few ideas which need a big player, but nobody steps up.
1. Copy update-in-the-background from Chrome OS/Android. If you want to actually improve it, try to provide an undo button (which will never work 100% though). Either use two partitions or disk snapshots for the background system. Do the wake-up-at-night-for-update trick of Android.
2. Turn the snap package system into a real app store. The https://uappexplorer.com/apps is not user friendly. There might even be money in here.
Linux users will not pay for appstore. There is no money to be made here. Just another timesink. You need to first attract significant non-developer users to Ubuntu (which is tied to making better laptops).
Nobody said it would be easy. On the other hand, there is Steam, which is an app store on Ubuntu.
The Ubuntu One store had technical issues and Snap solves at least a few of those if not all. For example, "works only on Ubuntu!" or "what about my dependencies?". It would be hilarious if Redhat users would buy their proprietary software from Canonical.
Still, the main challenge of an app store is not the technical side, but the business stuff: Get a critical mass of apps. It is also the reason why only Canonical or Red Hat could pull this off. It would be quite the hustle. Even Steve Jobs had some fights for the Apple App Store.
Some ideas to put there: Sublime Text, Gitlab EE, CLion, PyCharm, Zend Studio, Gurobi, Spotify, Netflix, TeamViewer, Matlab, Mathematica, SPSS, Stata, Maya, VMWare, Crossover Wine, Guitar Pro, Bricscad, Houdini, etc.
Another problem is the barrier of entry. How to get the customers to open an account? Once the account is open, buying stuff is one click and the 1$ apps will be bought on impulse. Cooperate with Humble Bundle and others for exclusive sales. The usual PR drill.
There are opportunities like App Store for Business. Where the company buys the IDE for all its employees via the app store and handles all the licencing there.
Just a minor point but uappexplorer is a community maintained site which uses the snap store API to present apps in the store. It's open source, hosted on github and I'm sure they'd welcome contributors to help improve the look and feel. https://github.com/bhdouglass/uappexplorer
There's a few other ways to view the snaps in the store though. On the Ubuntu desktop there is snap support in Ubuntu Software, and if you "snap install snapweb" you'll get a pretty web portal on port 4200 on the local machine. This can also be installed on remote machines so you can manage the snaps installed there too.
The article somewhat erroneously equates snaps and flatpaks. While flatpak focuses on GNOME, snaps are more broadly geared to any Linux software needing transactional updates (databases, Electron apps, docker/k8s, etc).
These days it is damn hard to tell where Gnome ends and Freedesktop, never mind Fedora, begins.
All too often it feels like Gnome and Freedesktop is out to fix "problems" that only exist within Fedora, in the process turning every other distro they touch into a Fedora clone.
I used KDE until they released that godawful Plasma desktop.
Don't get me wrong, it looks absolutely BEAUTIFUL, but it crashes constantly, and it is a crapshoot on whether or not it will actually even BOOT within a virtual machine.
I will attest to the same. My whole Engineering team is going to be migrating to KDE Neon as the primary development environment. Several of us have been trialing it for months, and as far as Linux goes it's been fantastic.
Yes - it's very annoying. There seems to some strange interaction with Oracle Java that it just can't handle.
The KDE bug reporting process is completely broken.
You need an account on their system just to report a problem. I suspect that a large number of problems are never reported because the process of documenting them is just too difficult for most users. There is no automated crash reporting at all.
As far as I can tell the bugs that are reported go into a black hole and not addressed for years, if ever.
In February 2003, a bunch of the outstanding bugs I'd reported against various GNOME programs over the previous couple of years were all closed as follows:
"Because of the release of GNOME 2.0 and 2.2, and the lack of interest in maintainership of GNOME 1.4, the gnome-core product is being closed. If you feel your bug is still of relevance to GNOME 2, please reopen it and refile it against a more appropriate component. Thanks...
This is, I think, the most common way for my bug reports to open source software projects to ever become closed. I report bugs; they go unread for a year, sometimes two; and then (surprise!) that module is rewritten from scratch -- and the new maintainer can't be bothered to check whether his new version has actually solved any of the known problems that existed in the previous version."
I'm so totally impressed at this Way New Development Paradigm. Let's call it the "Cascade of Attention-Deficit Teenagers" model, or "CADT" for short.
It hardly seems worth even having a bug system if the frequency of from-scratch rewrites always outstrips the pace of bug fixing. Why not be honest and resign yourself to the fact that version 0.8 is followed by version 0.8, which is then followed by version 0.8?
But that's what happens when there is no incentive for people to do the parts of programming that aren't fun. Fixing bugs isn't fun; going through the bug list isn't fun; but rewriting everything from scratch is fun (because "this time it will be done right", ha ha) and so that's what happens, over and over again.
Never heard of CADT before, but I have to admit that I've had experiences like this for sure (as the bug-reporter, not as the developer thank goodness).
I think we see that in all software projects.
It is usually more fun and maybe even easier to redesign than to fix.
I prefer to think of it as "aggressive refactoring" - which is not necessarily bad.
I think the process of software maintenance (in the context of KDE development) could be much improved if their reporting process had less friction and collected more relevant information so that the turnaround would be faster.
It is usually more fun and maybe even easier to redesign than to fix.
I think that there are interesting exceptions. For instance, the *BSDs are sometimes maddeningly conservative. But if you have last used FreeBSD 5.0 or OpenBSD 3.0, you are up and running on the latest versions in no time.
I hadn't looked at FreeBSD for years. Tried the latest version 6 months ago or so and configuration was pretty much the same as 10-15 years ago. There are new cool features (ZFS, bhyve), but it is all logical UNIX.
I installed 17.04 over Win10 on my desktop a couple weeks ago—my first time making a serious go of Linux on the desktop since whenever Ubuntu wasted a ton of my time with their half-baked PulseAudio "upgrade", which was a long time ago at this point—and started with KDE (I know from some dabbling that I hate Unity), but was hitting application crashes and long busy-cursor issues at a rate of several an hour, across many K-series applications—including settings, and the file manager. I constantly had little "something crashed" icons in the tray (which, by the way, failed to be interactive when clicked, so they just sat there). After a few days of that I switched to Gnome (which on an SSD and a four-core desktop class processor with a pile of RAM is now tolerably responsive) and that all stopped.
Still no cure for awful tearing every time I scroll in FireFox :-/
>Still no cure for awful tearing every time I scroll in FireFox :-/
What GPU are you using? There are various ways to fix this, depending on GPU/driver. I'd suggest looking in the Arch Wiki, even if you are using Ubuntu.
Nvidia gtx 1050, official drivers. Steam games are fine, and everything else seems to be fine, but Firefox scrolling looks like something from the Win95 era.
I revisit Kde every few years to see if they have finally managed to get hold of a designer. Since I use Unity and prefer it to OSX on my retina macbook and Windows 10 desktop, the recent announcement sent me on one of my visits.
I tried Kde Neon and was gobsmacked by the scale of change. With a few settings and a within a few minutes I was able to make it look and behave quite a bit like Unity. This was not the same KDE I remember.
It has improved dramatically and while the default is good you can feel the sheer flexibility of the underlying desktop. Kudos to the KDE team.
Unfortunately the Internet has a long memory and many will continue to repeat outdated information.
Thank you. I tried it too, and it installed great, but in the process of tweaking one of the two, I changed some setting for appearance and now Unity has a weird theme.
I think I liked LXDE more than XFCE, though. I will play with them some more and see, thank you.
In my experience, installing gnome-shell (or ubuntu-gnome-desktop) over a working Ubuntu installation can be dicey. For example, If you install both side-by-side the Unity theme can get a bit screwy, and the desktop-background just renders as a black rectangle.
Too bad he talks about unity for half the article, which I never liked much and rarely used. It had a lot of promise but snatched defeat from the jaws of victory. Didn't support the windows key or hot corner for opening the menu. Oh and hid menus by default.
More generally I'm not a fan of the war on menus across desktops, which are useful as a table of contents. Let them and title bars take less room on widescreen monitors, but don't cripple them.
Using MATE on 4k screens and happy. Still can't give up the clock/calendar with day/night map and timezones from gnome2. Could someone port that forward?, it's awesome. Advanced menu supports win key and type to filter.
He mentions the kernel with one sentence about kaby lake support, which was helpful to know, just got a new laptop. How are systemd and wayland coming along? Any other improvements under the hood?
Was very happy Python 3.6 was made easily available, when will it be the default 3.x? Cheers.
What is happening with Ubuntu is what happens when the top executives realize that it may be more profitable to mothball their companies' developments and sellout to the main competition rather than compete with them.
With Canonical and Redhat virtually in bed with Microsoft does any one seriously think that Canonical was going to proceed on a path that would mean getting into head on competition with a major company they had decided to collaborate with?
When Ubuntu collaborates with Microsoft on Bash/WSL what do you think is going to happen? When hardware vendors don't want to open up their drivers for Linux to compete effectively with Windows and Mac how much progress can Ubuntu make with their phones?
The Linux desktop is supposed to be a power users desktop but their poor vision of its developers failed it. It is supposed to be the equivalent of an advanced Smalltalk or Lisp workstation where its administration is concerned, to managed by messaging passing scripting languages like Smalltalk, Lisp or similar. The fact that it is still managed by tedious error prone scripts and configuration files only shows how poor the whole design has been. Coming to think of Linux is just the kernel, and none of the desktop developers had the smarts to develop a sound administration language and build the desktop proper on top of it. All that has happened is that it has given Microsoft and the other big guys the breathing space to retrench and now they have carved up the market by basically co-opting Redhat and Canonical into their camp.
Frankly Canonical has no future, and you can expect it to be absorbed by Redhat at some point in the future given that Redhat controls the Debian project which is the foundation on which Ubuntu itself is built. The only reason why Microsoft would not absorb Canonical itself is to give the impression of competition.
So in effect Ubuntu has turned into the goto desktop for Linux administrators, while Microsoft deepens and extends its reach by making Windows the goto tool for developing cross platform applications.
I am curious to see how gnome handles hidpi. In the past gnome would only accept integer values for dpi, while Unity allows floats (ex: 1.5), which is a deal breaker for me.
I might start using Ubuntu pure again. I honestly never understood the switch to this interface on desktop. Microsoft tried it on Windows 8 and failed.
Not the OP but I think hacksonx might be referring to the big squares UI. They designed an interface that one could use with their feet... elbows.... etc... And it had that Pinterest/Masonry appeal to it. I think they realized quickly that that's not how desktop users interacted with their computer. Was it a failure? I can't say I've met anyone that is trying to re-enable that interface on any other device. I do know they went back to using a more Window95-esque interface for the default Windows 10 desktop.
Sure. It was an interesting experiment (and certainly not one that I enjoyed -- I'm on Win7 Pro on the few VM's that I have to run MS on).
My point was that MS hasn't suffered massive failure despite this ludicrous attempt at promulgating a feet/elbows interface on devices with actual keyboards and mouses.
Yes, lots of people (especially relatively technically aware people that I know) have resisted the upgrade to MS Windows 8 / 10 -- but they're still running Microsoft Windows.
Compare and contrast the Canonical Unity experiment.
I suspect there would have been less of an outcry if it had been more like 8.1 before it launched.
That said, i think perhaps Windows 8 is the first (or second if one consider the Phone 6.5 to 7 transition) time Microsoft had ignored the muscle memory aspect that had kept their products up top for so long.
Can we please fix this before the convergence? Muscle memory and Linux go together like rosemary and garlic.