Slashdot Mirror
Ask Slashdot: Are Linux Desktop Users More Pragmatic Now Or Is It Inertia?
David W. White writes "Years ago ago those of us who used any *nix desktop ('every morning when you wake up, the house is a little different') were seen as willing to embrace change and spend hours tinkering and configuring until we got new desktop versions to work the way we wanted, while there was an opposite perception of desktop users over in the Mac world ('it just works') and the Windows world ('it's a familiar interface'). However, a recent article in Datamation concludes that 'for better or worse, [Linux desktop users] know what they want — a classic desktop — and the figures consistently show that is what they are choosing in far greater numbers than GNOME, KDE, or any other single graphical interface.' Has the profile of the Linux desktop user changed to a more pragmatic one? Or is it just the psychology of user inertia at work, when one considers the revolt against changes in the KDE, GNOME, UNITY and Windows 8 interfaces in recent times?"
What is a "Classic Desktop" and in what way are the other GUIs being discussed not "Classic Desktops"?
Would be funny to have a "Score: --1" for your post.
Linux users just haven't fell victim to the mass hysteria of solving a problem, which never existed. Apple designed an appealing desktop, and as their market share increased, Microsoft began throwing UI designs against the wall. Then people started buying phones and tablets, so designers decided no one wanted a functional desktop anymore. Gnome 3 decided to screw everything up, then Ubuntu decided they wanted everything screwed up in a different way. KDE made the same traditional desktop demand more resources, making it unusable.
Everything has to do with productivity. Sure we all like a bit of novelty and it's fun to tinker with new features of a desktop or user interface, but the majority of these innovations are never used (if the user has the choice), but the recent Linux desktops (Gnome mostly) have forced a new set of heuristics on a user base that increasingly uses Linux for productivity and not just tinkering.
It's a waste of time to have to learn a new way of doing everything when the existing ways work already. That is why 'classic desktop' is favored. It works, and although new things might work, they have not proven to work better.
Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
I don't see it as a "revolt against change" but a revolt to changes for the sake of change (enter gnome 3 and windows 8 as exhibit A and B).
I'd like to have something like the Win 7 Start Menu, but XFCE with the Panel on the bottom is (a) Good Enough, and (b) easy on the brain, since I frequently switch between my Linux box and the company's Windows 7 Enterprise laptop that sits right next to it.
"I don't know, therefore Aliens" Wafflebox1
If you can't have a consistent experience across even one day, why get too reliant on customizations and shortcuts?
Back in the day, I had to switch between Data General (terminals), MacOS, and Amiga keyboards and UIs on a daily basis between work and home. These days, of course, everything has changed - now I bounce from Linux to Android to OSX, and more than occasionally Windows too. It's just never paid off to build a super-custom setup when you can't stick with it.
I use Linux for my main desktop at home partly because it is so quick and easy to reinstall - just keep your data on a backed-up server and you can virtually forget about maintenance or troubleshooting. Get used to the default setup and just reinstall whenever you run into something you can't work around - 15 minutes to get back to a familiar desktop is quicker than any full restore-from-backup I'm aware of. (I actually like Linux internals but every time I learn something, I end up forgetting it before I need it a second time; it gets frustrating...)
I'm aware I'm giving up a fair amount of potential productivity and convenience. I don't care any more. I'm just happy when I remember not to try and touch the monitor on my wife's iMac.
I got friends and colleagues who, for example, use Dvorak. More power to 'em. They're younger and more stubborn than I, and most of the time they have one laptop they use both at home and at work. As a wise man once remarked, I'm older now, I got to move my car on street-sweeping day, I can't be doing just anything I want any more...
Perfectly Normal Industries
I'd say it has less to do with any change in user tastes and more to do with the apparent move from a situation where the present state of interfaces is bad; but improving (which, fairly obviously, creates enthusiasm for new stuff) to a situation where most of the improvements have been mined out; but there are still UI designers around, so they've just been changing random things in some horrible mockery of genetic drift.
When version N+1 was probably an improvement, getting motivated to go poke it until it works was easier. Now version N+1 may have some cool new feature; but it'll probably have 8 regressions, the pointless removal of something you liked, and probably tentacles. Why bother?
"Pragmatism" versus "Inertia"? What a strange choice that doesn't align with pro/con argumentation.
FWIW, let's look at a continuum of Linux/Unix desktop users instead. We know that a core group will tend to prefer a minimalist X-Windows desktop such as IceWM for the least impact on hardware performance. Many users prefer desktops like XFCE, Razor-QT, LXDM, and others that offer lightweight but fuller and more integrated experiences than the truly minimalist ones, acknowledging that the load on a system tends to increase as more features are included and deciding strategically to suit their usefulness-efficiency preferences. At the other end of the spectrum are those users who want an entire desktop environment in which all the bells and whistles are integrated into a particular look and feel, as characterized by KDE and Gnome, but understandably with a heavier load on the underlying hardware. So, I suppose pragmatism enters into such choices. To each their own, and having such choices is wonderful. Inertia? There are those who will say "I use KDE because I learned on it and I'm used to it", but this also is a pragmatic choice and not one of "inertia".
I deny that I have not avoided attaining the opposite of that which I do not want.
GNOME: the desktop that COULD be awesome, if only the dev team actually cared about performance, polish and a reasonable feature-set. Overall this desktop has the best feel and most potential, but sadly it is never quite realised.
KDE: at first this desktop seems powerful and feature-rich, but after a week of using it you realise how little its devs care about usability and sane defaults. Not everybody wants to make a career out of tweaking their desktop.
Unity: has SOME nice usability aspects, but it is only properly supported on Ubuntu, and Ubuntu is an extremely buggy OS.
Xfce: fine for very basic use, but lack of proper OS integration (like GNOME) and some annoying bugs make this desktop unusable.
LXDE: almost total lack of OS integration. It's more like a collection of recommended packages for a minimal X desktop than an actual DE.
Cinnamon: GNOME done badly. Sure, you get your somewhat classic launcher and panel, but it just feels clunky compared to Windows, GNOME or Unity.
MATE: what can one say about MATE? It does the job and GNOME 2 was great in its time, but the desktop is starting to show its age. Probably the sanest choice for getting real work done, but not as satisfying as more modern desktops.
Summary: if GNOME would stop reshuffling the deck chairs and spend a few releases on performance, polish and features real-world people care about, they could easily become the most popular desktop. They've done 99% of the work, but for some reason are blind to that crucial last 1%. Given that this is probably never going to change, the Linux desktop is pretty much an exercise in futility and inefficiency.
Whenever I leave my [former-]chomebook in the bathroom when I take a shower & everything's still running right except for the mouse, I can get by until the next reboot without using it. IMO, that's a clean interface. That said, out of the box you already pretty much require unity-tweak to fix all the shit they got wrong. The opacity settings, the workspace layout, etc....
When I look at all of the major variants mentioned - Gnome, KDE, Windows, Apple - I honestly don't see any great difference.
All of them offer:
- A desktop
- some kind of task bar (top, bottom, left, right - doesn't really matter)
- some form of menus for getting to stuff
- some kind of file manager application
There may be some things that are very different from one to the other (Lord knows that when I switched to a Mac I found some of their choices thoroughly obscure) but in the big picture most desktop systems are similar enough that Joe User can go to one or the other and figure out how to check his Yahoo mail account without problems.
As for why the GNOME variations seem to be prevalent? It's because some form of GNOME desktop was included as the default for the first widely popular "works out of the box" distros - Ubuntu, and Mint. the Son of Ubuntu.
People didn't install Ubuntu/Mint because of GNOME; they installed GNOME because it came along with Ubuntu/Mint. And 95% of those Linux users won't muck about and try different desktop systems because what they have just works.
Three Squirrels
Common Graphical Computer user interfaces haven't really changed that much in the last 15 years, in general.
In general, there's:
- a single place for starting common applications
- a place for starting more unusual ones.
- windows full of files which can be manipulated with menus, dragging, and dropping.
- some method of switching between those applications.
And so people will naturally gravitiate to the variant of this scheme which is most familiar to them.
There's a lot more 'mainstream' users of open source/free desktops these days, and most of them don't actually want the fun hacker 'computers, and interfaces in general are interesting problems, lets hack and play with the concepts and see what we can invent'. They just want something they can use straight away, and customise as much, or as little, as they want.
I'm using awesomewm, and find it almost perfect for me - but largely because it didn't require too much hacking to get it very familiar. (tiling, virtual desktops, 'command space' textual launcher...)
However, if there was a new, interesting, and different model all together, I'd be fine testing and playing with it for a few weeks - but for a very long time now I've not really seen any new paradigms which offer anything interesting (from a conceptial point of view). Things like unity, or gnome3, or whatever, only offer the same boring old models, shinier in places, but more limited in others.
Partly, that will be because the type of applications that we're all running — no matter what desktop — work the same way. If there was an entire suite of programs that worked in another manner altogether, perhaps with circular pop-up menus, any element dragable and dockable into any other, objects having interactions, not applications... then we'd have something more interesting. But that's an awful lot of work - creating an entire new desktop paradigm.
What are the unique selling features of each desktop system? Why would I *want* to change what works?
Linux users have the option of choosing a different desktop environment or window manager. So, if a concept is unwanted or immature, users can and will migrate elsewhere. There's usually not a great risk involved, maybe the programs you use will be less integrated in your new DE/WM. Part of this is a resistance to change. It's something that happens to basically all humans. Another part of it, though, is that end users have a fairly reasonable choice in the matter, unlike on Windows or OS X, where there is only one path forward, at best having some kludge solution that may or may not be reliable.
This is my signature. There are many like it, but this one is mine.
I just (five days ago) spent two days huddled with a half dozen other developers in the corner of a large conference room filled with IT people in Chicago. We were testing our various implementations of a new protocol that we expect to see in wide use during the next two years.
I had brought a brand new laptop, for various unfortunate reasons, on which I had just installed the complete stack of software I needed night before in the hotel room. I put Ubuntu 13.1 on it because I happened to have that particular distro on a flash drive that was at hand just then and I was in a hurry.
Things worked out. The laptop worked well and I got my part done. Thing is, I spent that rather intense period of time using Unity. For development and testing of software. Really.
I get it. Unity is fast and effective, particularly on the limited real-estate of a laptop screen where you end up switching rapidly among full screen applications.
I've avoided Unity like the plague on desktop hardware were I have multiple, large displays, and I think I'll continue doing that. However on a laptop that is not running external displays Unity works pretty well. You can navigate quickly with mouse or keyboard and avoid fussing with things. The fixed position of the large icons (although too large by default) on the sidebar is particularly useful.
So, bust out the fangs and hate me down with your mod points; I found a use for Unity and said so on Slashdot.
Maw! Fire up the karma burner!
I always end up going back to a customized XFCE, but about every 6 months, I decide to try something else, and usually end up wiping my system and reinstalling before I'm done.
My wife has a mildly customized XFCE setup, and she loves it. It almost never gets changed or tweaked.
using too many words. He means that users of personal computers (as opposed to mobile devices) want simply a "desktop."
As in, the metaphor—the one that has driven PC UI/UX for decades now.
The metaphor behind the desktop UI/UX was that a "real desktop" had:
- A single surface of limited space
- Onto which one could place, or remove files
- And folders
- And rearrange them at will in ways that served as memory and reasoning aides
- With the option to discard them (throw them in the trash) once they were no longer needed on the single, bounded surface
Both of the "traditional breaking" releases from KDE and GNOME did violence to this metaphor; a screen no longer behaved—at least in symbolic ways—like the surface of a desk. The mental shortcuts that could draw conclusions about properties, affordances, and behavior based on a juxtaposition with real-world objects broke down.
Instead of "this is meant to be a desktop, so it's a limited, rectangular space on which I can put, stack, and arrange my stuff and where much of my workday will 'happen'" gave way to "this is obviously a work area of some kind, but it doesn't behave in ways that metaphorically echo a desk—but I don't have any basis on which to make suppositions about how it *does* behave, or what affordances/capabilities or constraints it offers, what sorts of 'objects' populate it, what their properties are,' and so on.
I think that's the biggest problem—the desktop metaphor was done away with, but no alternative metaphor took its place—no obvious mental shortcuts were on offer to imply how things worked enough to allow users to infer the rest. People have argued that the problem was that the new releases were too "phone like," but that's actually not true. The original iPhone, radical though it was, operated on a clear metaphor aided by its physical size and shape: that of a phone—buttons laid out in a grid, a single-task/single-thread use model, and very abbreviated, single-option tasks/threads (i.e. 'apps' that performed a single function, rather than 'software' with many menus and options for UX flow).
Though the iPhone on its surface was a radical anti-phone, in practice, the use experience was very much like a phone: power on, address grid of buttons, perform single task with relatively low flow-open-endedness, power off and set down when complete. KDE4/GNOME3 did not behave this way. They retained the open-endedness, large screen area, feature-heavy, and "dwelling" properties of desktops (it is a space where you spend time, not an object used to perform a single task and then 'end' that task) so the phone metaphor does not apply. But they also removed most of the considered representations, enablements, and constraints that could easily be metaphorically associated with a desktop.
The result was that you constantly had to look stuff up—even if you were an experienced computer user. They reintroduced *precisely* the problem that the desktop metaphor had solved decades earlier—the reason, in fact, that it was created in the first place. It was dumb.
That's what he means by "classic desktop." "Linux users want a desktop, not something else that remains largely unspecified or that must instead be enumerated for users on a feature-by-feature basis with no particular organizing cultural model."
STOP . AMERICA . NOW
Come on... KDE... It's not the old KDE 4.0.
All those other guys can keep their all-encompassing UI vision. I don't want their kool-aid. I'm glad I get a choice in Linux. I may have to occasionally beat my head on the computer for days at a time when something stops working, but at least I can avoid having some corporate assholes or desktop environment programmers who like the smell of their own farts ramming their bullshit down my throat.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
I think it has a lot to do with when you came up. When I came up with computers in the 1980s and 1990s we had hard problems and solved them. It was a world of rapidly growing IT spending, with IT taking on more and more tasks. After Y2K the technology sector began to get very conservative, the focus was on cost cutting and reliability. Far more like the world of the late 70s and early 80s in Mainframe and Minis that the PCs had replaced. What's exciting now is that mobile devices have brought back that enthusiasm for change and excitement again. They haven't caught up with desktops but at least they are creating a generation of developers who are used to a market that grows and expands rather than stays put at minimal cost.
I watch the threads on any kinds of change whether it be ubiquitous computing (Windows 8), IPv6 (networking), Wayland, the new hardware designs... and there is a pervasive pessimism among younger IT, a terrible can't do attitude.
Back in the 1990s when Linux was coming up we had sorta GUIs die: FVWM, AfterStep, SawFish, AMI-wm, Openlook (olwm), blackbox... Systems grow change and die leaving behind better ones. What's terrible is that the new generation wants stagnation. Either Gnome 3 succeeds or it doesn't. But regardless of what happens the work on Gnome advances the ecosystem.
Does the summary make sense to anyone?
Dear Slashdot: next time you want to mess with the site, add a rich-text editor for comments.
good for you. To answer the question though I think it's psychology of efficiency. If the tools aren't efficient for the brain to categorize/understand it's not practical as an interface (desktop or otherwise). The problem with Metro isn't that it's different, it's that it's too much visual clutter for the brain to process quickly. This is reflected in GNOME/KDE in that, while neatly organized, it relies on memory association of images to functions. Icons are everywhere these days so those associations aren't as strong or that part of the memory is overloaded to access efficiently. Non-graphical interfaces suffer from something similar in the ability to remember all the commands and their associated flags.
The classic desktop organizes things in groupings, lists, etc and while there's icons associated the overriding organization of alphabetical text gives shortcuts for the brain to compartmentalize information where it can or to simply analyze because all the information is there (where KDE/etc you must hover to get all the info one icon at a time)
The more I look at the whole changes in OS-UIs lately, the more I get the impression that the whole cross-platform thing got lost it's grip to reality.
Sure I like my tablet, my smartphone, my laptop,... and I live with the smudged display I have on my tablet and my smartphone, since the do not really bother me. Probably because I can easily overlook these smudges, but since I can't overlook them on my normal monitor (or laptop display; or my glasses for that matter), I'm no friend of UIs which seem to be designed for tablets&smartphones but get presented to my as 'new' and 'easy' interfaces for my normal displays.
I like the idea of having one back-end, but I also like different frontends for different tools.
-> I get that some designers like their tablets and think that one UI should rule them all, but I don't agree with it.
They are just badly designed or have irritating flaws.
GNOME went to reinvent what turned to be a half-assed tablet UI lacking even basic features that desktop users expect (where is my terminal?? why you can't even log out unless there is more than one user in the system - wtf?).
Classic version also doesn't have half features of the old GNOME2 (which also sometimes strived for too much simplicity, but was still a nice, if a bit minimalistic, desktop).
Unity needs work to even get to GNOME2 levels of. I haven't used it much, but you notice quickly a lack of polish. It also somewhat follows similar design idea as GNOME3.
KDE4 - most similar to the "classic" desktop, but mixed up with stuff that was hyped some time ago (like widgets). Add to that unnecessary controls (rotating icons), and overall bad theme layout (this might also be a fault of Qt), and you don't get a pleasant visual experience. but maybe most usable of the three mentioned.
Windows UI was a bit too simplistic, so Unix GUIs added a few tweaks, like virtual desktops, that worked fine.
The new Metro is crap. I'm writing this from 8.1, though I only use Metro as a start menu and so far it doesn't bother me.
Maybe OS X has the best and most polished desktop around, but notice that they dialed back compositing effects that some time ago they used to produce a "wow" effect (compare this to compiz which had that moment too).
Also the system has some silly limitations - why can't I open more than one calculator? Oh because OSX only runs one instance of each application.
Desktops can these days be considered a solved problem and a mature area. The time of reinventing the wheel here and do radical redesigns (like GNOME3) have long passed, and these attempts often end up in something too experimental or too inferior. Desktops in almost any commercial OS are these days designed by big teams of _designers_, which we are often lacking in the free software world (even those in corporations also manage to screw up, like with Metro). Results of this disparity are desktops that we have now. Copying other UIs worked, but rethinking one not so well.
My car has four wheels. Works best at the moment.
ftfy
what about if a car came along that didn't have wheels? would you not buy it simply because it didn't have wheels?
wheels on cars only works best because you haven't experienced anything better... but that doesn't mean that wheels will always work best.
change for the sake of change sucks, but innovation stems from change and innovation can also lead to change for the better.
What exactly is a "classic" desktop anyway? Are we talking classic Windows? Classic Mac OS? There's a constellation of UI paradigms which work. Some of them are mutually incompatible, you can't use them simultaneously. If you want to come up with something new, it has to actually work better than what we had before. If it merely works "as good" as what it's replacing then users won't be happy. You're changing things for the sake of change. So from those choices you pick the ones you think work best together and create a DE out of them. So we get Gnome Shell, KDE, XFCE, et al. Then there are the numerous eccentrics, throwbacks, and masochists running things like Awesome, DWM, Trinty, or any of the others which don't even add up to 1% all together.
I don't think Linux users are getting more pragmatic. The different camps have mostly just solidified around their own "classic" vision. There's 3-4 different main camps now depending how you choose to slice it, and numerous sub groups and forks if you drill down deeper. It'll always be more fragmented, contentious, and fluid than Windows or OS X. That's a good thing, as long as you have the wherewithal to navigate your way between all the various spin-offs and cousin projects spawned when the devs make a boneheaded change for change's sake. Gnome 2 users need to know enough that MATE is their upgrade path, etc.
I've actually been using Unity these days. It's level of polish and completeness is better than anything else I've found and it replicates the features I most enjoy from OS X. I had to install a less offensive theme and icon pack, change the system font to Lucida Grande, but after that it's a very nice desktop. I only have a few criticisms: you can't move the dock to the bottom; the search features aren't as simple and elegant as Spotlight, lenses are over-engineered and pointlessly complicated for what the achieve even if it's a more powerful tool overall; and there are a couple minor GUI glitches which I've come to find unacceptable after spending so much time in the pixel-perfect world Apple has created.
If you build it, nerds will come. Soylentnews.org
How great for you. You admitted it does not work except in a limited netbook like sense for 1 task.
I do more and do not like where linux is going. So in 2011 I switched to Windows 7 and never looked back ... until Windows 8 :-(
http://saveie6.com/
What i'd like is a terminal with an integrated visual file browser.
A Good Troll is better than a Bad Human.
It was a lousy metaphor when first proposed and it remains a lousy metaphor today.
And ironically while the article defines a "classic desktop" with icons on it (how gauche!) it goes on to mention WindowMaker, which offers a root window metaphor instead. That's my personal favorite.
"Desktops" in the sense of KDE or GNOME are just too creepy, too cluttered, too always trying to make you do things their way. They include WAY too much garbage I will not use and do not want. KDE *is* more tolerable than GNOME for me (since Gnome 2 at least, bleah) but it's not really what I want. It's still too much in the way and it's still doing too much.
Even E tries too hard, though it's a lot of fun. FVWM and the like are functional but too ugly to really make me happy. WindowMaker is the best, minimal, functional, and still gorgeous. Everything a Window Manager should be.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
In the past(late 90s early 2000s) the various machines that I had barely worked. So I noodled and fiddled until the machine was just the way I liked it. But then at some point, I largely stopped. Basically the machines were powerful enough that tweaking didn't buy me any critical functionality or performance to make it worth my time. Also the defaults for almost any OS are close enough that my total "tweaking" might take 5 minutes or less from a default configuration.
In many ways I think that it less that we don't tweak as the machines are coming pre-tweaked.
Obviously this is not for everyone as we all know those people who must spend a full day getting a new machine just the way they like it.
But if I had a new machine built from scratch tomorrow I would say that 50 percent of the few minutes of tweaking would be spent changing the IDE defaults for a few keys and whatnot. The bulk of the rest would be eliminating stupid default icons and putting up a few that I frequently use (Terminal, etc)
I just spun up a raspberry pi and with the arduino IDE sitting right on the desktop I'm not sure that I'll make a single change at this point. Any changes going forward will be 100% in support of critical functionality.
Norton Commander...
Oh, they have the internet on computers now
After 20 years of experimentation, the conclusion is that the desktop metaphor is probably too complex for the average user. Power users appreciate floating windows, file hierarchies, multiple screens, notification bars, hierarchal menus etc. Meanwhile the more typical user maximizes one window at a time, clicks icons, and saves everything in the same place. The "phone/tablet" model is much closer to the average person's mental map of how a computer should work.
The problem is that Linux users are 'power users' almost by definition so KDE/Gnome were terrible places to experiment with replacing the desktop metaphor.
GNU Midnight Commander.
I've been a Gnome user since around 2001, to say things were pretty rough back then is an understatement... In 2012 I switched to KDE. I finally had a machine with 16GB ram to run it on (FWIW KDE seems slightly better at running on limited hardware now, but stil..) Its defaults made me angry, though (especially Konsole - seriously, no keyboard shortcuts to hit a specific tab? Tabs at the bottom [oposite edge to the menus and titlebar]?) but I can actually repair it a lot quicker than fixing Unity/Gnome.
It's been this long and they still can't make KDE remember the orientation/resolution/relative position of any monitor that isn't the primary one - if I'm going to suffer through that sort of thing I might as well give i3-wm a proper go. I was able to use it productively for a whole day recently, which is more than awesome and xmonad lasted for me.
I wish I could disagree, but I help so many users that run one program full screen. I just sit back and shake my head as they constantly switch from one program to another instead of arranging the program windows to see everything they need at one time.
It really start to piss me off when they have two monitors and switch between two programs, both on the main screen, both full screen. Then they wonder why it takes so long to get things done.
Except that the desktop cannot work using the phone/tablet model because user expectations do not suggest that metaphor when they sit at a desktop.
Even if the desktop metaphor was too complex to master, users still sit down at a desktop and think, "now where are my files?" because they intend to "do work in general" (have an array of their current projects and workflows available to them) rather than "complete a single task."
As was the case with a desk, they expect to be able to construct a cognitive overview of their "current work" at a computer—an expectation that they don't have with a phone, which is precisely experienced as an *interruption to* their "current work." KDE, Gnome, and most recently Windows 8, made the mistake of trying to get users to adopt the "interruption of work" mental map *as* the flow of work. It's never going to happen; they need to be presented with a system that enables them to be "at work." In practice, being "at work" is not about a single task, but about having open access to a series of resources about that the user can employ in order to *reason* about the relatedness and next steps across a *variety* of ongoing tasks. That's the experience of work for most workers in the industrialized world today.
If you place them in a single-task flow for "regular work" they're going to be lost, because they don't know what the task is that they ought to be working on without being able to survey the entirety of "what is going on" in their work life—say, by looking at what's collected on their desktop, what windows are currently open, how they're all positioned relative to one another, and what's visible in each window. Ala Lucy Suchman (see her classic UX work "Plans and Situated Actions"), users do not have well-specified "plans" for use (i.e. step 1, step 2, step 3, task 1, task 2, task 3) but are constantly engaged in trying to "decide what to do next" in-context, in relation to the totality of their projects, obligations, current situation, etc. Successful computing systems will provide resources to assist in deciding, on a moment-by-moment basis, "what to do next," and resources to assist in the construction of a decision-making strategy or set of habits surrounding this task.
The phone metaphor (or any single-task flow) works only once the user *has already decided* what to do next, and is useful only for carrying out *that task*. Once the task is complete, the user is back to having to decide "what to do next."
The KDE and GNOME experiments (at least early on) hid precisely the details necessary to make this decision easy, and to make the decision feel rational, rather than arbitrary. An alternate metaphor was needed, one to tell users how to "see what is going on, overall" in their computing workday. The desktop did this and offered a metaphor for how to use it (survey the visual field, which is ordered conceptually by me as a series of objects). Not only did the KDE and GNOME not offer a metaphor for how to use this "see what is going on" functionality, they didn't even offer the functionality—just a series of task flows.
This left users in the situation of having *lost* the primary mechanism by which they'd come to decide "what to do next" in work life for two decades. "Before, I looked at my desktop to figure out what to do next and what I'm working on. Now that functionality is gone—what should I do next?" It was the return of the post-it note and the Moleskine notebook sitting next to the computer, from the VisiCalc-on-green-screen days. It was a UX joke, frankly.
The problem is that human beings are culture and habit machines; making something possible in UX is not the same thing as making something usable, largely because users come with baggage of exactly this kind.
STOP . AMERICA . NOW
I liked the simple yet configurable Gnome interface, then they took it away from me with Gnome 2.. but I got used to it somehow. Then they took that away from me with Gnome 3 and I've been using XFCE4 since. My "simple light working dekstop" preference hasn't changed much over the years, but the GNU/Linux desktops has. KDE today isn't like KDE 2 (which was usable), it's HUGE (but atleast I can configure it to be a simple desktop if I want). GNOME3? wtf. My preferences didn't change, GNOME did.
9/11: Never forget it was a false-flag operation
The listed desktop managers had adequate functioning, standardized modi operandi, and accepted user interfaces, but decided to scramble them to fit tablets- ala one size fits all. In the Linux world, KDE went first then GNOME. People such as myself became promiscuous because they were changing things that didn't need to be changed and ignoring the things that did. I've since "returned" to my environment of choice and it's improved, but the freedom to customize may never fully recover. It matters less to me, anyway- I'm more face-deep in terminal multiplexers these days- better eye candy than compiz or screenlets.
About the same obsession teenagers have with the way their room looks: they are probably going to spend a lot of time in it (YMMV), it's the equivalent of home (the rest of the house is the parent's territory, and thus hostile) and it's also an extension of who they are.
If it is for work, you just deal with it. It might be "wrong" enough that you need to try to cause change, but ultimately it's not about you, it's about getting the work done. You'll notice that offices are usually slightly personalized. You don't get to paint the walls, probably won't pick the desk nor the chair, but you'll be able to put a photo or something to make it feel more personal, more you. Now... your own house, your own room (whether it's where you sleep or spend most of your time awake... whatever lets you feel at home), that needs to be something you like. You'll want it personal and pleasing.
Well, I guess saying it's a teenager obsession isn't fair after all... We all like to personalize the space we use and the things we value.
I don't care if I'm wrong. I only care about everyone obtaining something from the discussion.
For the most part i spend my time in browser, terminal, pdf reader, word processor and occasionally a dedicated IDE.
All i really want in a UI is the ability to switch between these apps without having to mentally switch contexts. On a non-touch computer, a menu list of installed apps+taskbar with a stacking window manager is ideal.
Linux users are not the only ones who are rejecting the new UIs. Everyone hates how windows 8 works.
There is clearly a need for new UIs for touch based machines. The mistake is trying to create one UI that works for both worlds - this is the mistake Win8 and GNOME3 made.
thanks!
A Good Troll is better than a Bad Human.
15 years ago, it was [kind of] cool to play around with config files, compile kernels and install different Linux distros the way women change their purses. Now we have other priorities in life, kids, pets, mortgages. We just want to get the job done. Sometimes I enjoy hacking some config files for fun, but it's not anymore something I'd do on a Friday night.
Open Source Network Inventory for the masses! Kuwaiba
"what about if a car came along that didn't have wheels?"
They have. It's called a "hovercraft" and proponents saw them as the wave of the future.
They are inefficient, lack positive steering or braking (good luck stopping one on a downgrade) and remain in the niche markets they suit.
If a future wheel-free car is offered, I won't need to "try" it to determine if it suits my requirements. I can infer that from what I see it do.
"This post is an artistic work of fiction and falsehood. Only a fool would take anything posted here as fact."
I was happy to switch to xfce, even though I'd never used it before. I'll use any good UI, but I won't tolerate a bad one for long.
UI designers must learn to stop changing user interface on each release. Some people try to use computers to do work, and anything that changes is a pain for them.
Once again, car analogy may be insightful: do we ever saw a car maker replacing the UI on all its model, for instance replacing the wheel by a joystick or a touch interface?
Yes, there was some backlash against 4.0 which by-the-way was meant for *developers* and not end users. But since then there has been no controversy.
So you left Linux because you demand a single interface and are not open to using Unity for netbooks and KDE/Gnome/XFCE/your preference for your normal uses. But then when Microsoft tries to unify with a single interface for desktops and tablets with Windows 8, you hate having a single interface and want to use different interfaces for different types of computers.
This space intentionally left blank
The way other people prefer to work pisses you off? Seriously? Frankly, other people telling me how to work pisses me off.
I think there should be a cross-distro standard desktop that JUST FREAKING STAYS THE SAME.
In other words, you want an eXtremely Fcukin' Constant Environment (or Xfce for short). I agree, which is why my clean PC runs Xubuntu.
All Ican figure is that either the author either believes that it's not classic if we can customize a GUI to the point that it's no longer "classic" looking
In my opinion, a desktop that can be easily customized to act "classic" is classic enough, so long as users are made aware of this customizability. But there's a practical problem with presenting too many options for customizability. See the section "The Question of Preferences" in this article.
or is judging it based on the first few releases when it wasn't fully functional as a 'classic' desktop yet
A not-yet-fully-functional GUI shouldn't be shipped as the default GUI of a GUI-oriented operating system until such time as it becomes fully functional.
Which operating system is it you're saying doesn't allow you to launch multiple, simultaneously functioning applications?
An Android phone and a PC are dramatically different tools that are used for dramatically different tasks and styles of interaction.
Are an Android tablet's "tasks and styles of interaction" closer to a laptop or to a phone, and why? Does this change when the laptop is docked to a physical keyboard, such as an ASUS Transformer?
When I was a student I used to try a different window manager and/or desktop every day. Then I had a full time job and stuck with KDE. I went along with the changes in KDE4, I really wanted to like it, I tried to for years! If everything "just worked" I might have stuck with it. Recently I decided I was tired of bothering, now I use Ratpoison and TWM. (TWM is for when I am using applications that have lot's of detached windows. Tiling window managers are awesome but not in that case!). I'm not very sold on TWM, I'm looking for a similarly lightweight non-tiled window manager to take it's place but without the funky right-corner resize button thing.
Honestly I'm not sure any of this is ideal. I think I might like a 'classic' interface, something with a task bar, start button and desktop that displays shortcuts/files in a desktop folder. But... also give it a tiling mode and lot's of keyboard shortcuts so it works well with or without the mouse.
Hmmmm... I may have to start learning about coding desktop managers....
Laugh if you want, I'm working on a product that's still shipping fvwm2 (recently updated from fvwm95...)
It's... sufficient, and the devil whose face we know - all of its shortcomings, bugs, and workarounds are well known and documented - unlike a "cutting edge" desktop that throws you a mystery quirk every so often that nobody knows about.
I agree with most of what you said, but KDE 4 doesn'tbreak the desktop metaphor. The initial releases did, either due to being unfinished or experimental, and the developers thankfully restored the traditional 'classic' desktop a bit at a time after that, with the non-classic variants as options we can choose.
Here's a quick screenshot of my desktop, showing files I've clustered together, folders (Icould move them, but haven't felt a need yet), a trash can Ican drag them into — and to add to your description, windows layered on top of each other like physical items (with shadows to strengthen the impression), and the ability to show what's inside a folder right there on the desktop.
Suggest showing them (on Windows) Right-click > Show windows side by side (formerly "stack vertically"). I showed that one thing to my girlfriend and she's thanked me again and again and again for the improvement to her workflow.
We know where leadership by an anti-intellectual "strongman" who scapegoats minorities and likes boisterous rallies goes
If you want a UNIX desktop that just works, then you get a Mac.
Indeed. The problem KDE 4.0 had was that compared to the very polished 3.10 series, 4.0 was incomplete and buggy. This was a problem with the implementation however, not with the design. This is very different to the GNOME and Unity situations, where a lot of users didn't agree with the design decisions.
I use it on my HTPC, and have got used to it. It's really not much different than OS X. I don't get the complaints, although I prefer the Gnome2 launcher compared to the weird search thing Unity has going on, but I can pin the apps I use most so it's not too bad.
Twinstiq, game news
One difference is whether you can see an application and another application at once, or an application and the menu for starting additional applications at once. This becomes important if you're doing something that requires multiple applications, such as reading in one application and writing in another. Windows 8's Start Screen can't share the screen with an application, and "modern UI" applications really want to fill the screen. The iOS and Android GUIs have the same "all maximized all the time" mentality, even on a tablet with a display well over twice the size of a phone. Or do the tablet makers expect people to buy two tablets, one for the reading and one for the writing?
When I had machines with a more modest CPU and less RAM, I used low-overhead GUI's like DWM, LXDE, Fluxbox and XFCE. Now, RAM is plentiful, and these days KDE is tweaked to operate quite efficiently. It certainly is cool to use command line terminals as a GUI, if you are nimble enough to move stuff around via command line. On the other hand, it is also a pain to set up stuff like the battery/weather/cpu/RAM/monitor on Xmonad or via conky, rss feeds and the like. With KDE, all that stuff is just there, and it is kind of pretty.
"SO we bide our time, waiting for a purer kick to bloom and the future is still bleak, uncertain and beautiful" -GSYBE
Everyone wants a classic desktop, but no vendor wants to provide one. Microsoft wants everyone on Metro so they can take a cut of sales through the App Store. The KDE and Gnome teams want to experiment because it's more fun than maintaining a tried-and-true design. Apple is seemingly holding the line for now, but all it takes is one bad VP in the UI team and OSX will become a clone of iOS.
UI designers don't like the desktop metaphor for a variety of complicated philosophical reasons. They think it would be easier for people to learn how to use computers if it was abandoned. Maybe they're right about that – iOS has been very successful among non-technical users because it simplifies things a lot more than a standard W.I.M.P. design – but once you get beyond casual use and into doing real work, multitasking becomes a necessity, and there is still nothing better than a "classic desktop" for that.
Best GUI used to be xterm. Nowadays Cygwin/Mintty with SSH does the trick. Fat clients are hardly justifiable. A few exceptions are Office suites (Web applications still haven't hacked that), IDEs and web browsers.
I sometimes cringe when I see a promising lad mucking about with copy/pasting stuff from GUIs to analyze something. Same analysis with *nix tools would take a fraction of the time and be more precise.accurate.
I hadn't the slightest objection to his spending his time planning massacres for the bourgeoisie... (P.G. Wodehouse)
"a desktop that sits on top of GNOME technology, such as GNOME3, Cinnamon, Mate, or Unity" from the reader poll linked roughly adds up to the numbers quoted, so there is no paradox. Especially since the numbers are apparently meta numbers since they only approximate what was written.
That means the author is an idiot and your explanation of what the author meant is completely wrong since it is clear that the author started with some very basic misunderstandings that you most likely did not start with.
Moreover, when a GUI is released, I'd like it to be tested beforehand. Also, it should come with the basic features any decent GUI has. How come something like Unity can be released, replacing the classic gnome, being a bug nest, missing so obvious needy features. That's beyond me. "tinkering and configuring" is the least of my problems.
Slashdot, fix the reply notifications... You won't get away with it...
I use xfce for two reasons. I read that Linus himself recommended it and Unity is slow as fuck on my 2gb of ram. I'm hoping to upgrade my system soon, though.
I use Linux because it is fun and secure. I would say I prefer a âoeclassicâ desktop but I don't see that really as a valid explanation. The latest OSX retains features of the original MacOS but I wouldn't say that the explosion of Mac users gives a damn about that. Part of their migration is disposable income, part of it is fashion, and part is ease of use (for a considerable financial cost). Windows users, I've heard, are alienated by Windows 8 -- largely because of the interface. A good Linux AdWords campaign would take advantage of frustrated Windows 8 user searches, IMO.
Personally I just don't believe in spending a lot on technology in a country of so much excess. My smart phone, tablet, and computer together cost me about the same as the Chromebook I bought my wife for Christmas. Chrome doesn't charge to updates, Android doesn't, and neither does Linux. I use them all for this and other reasons.
But I still don't think the UI is the main attraction with Linux. I think the main attraction is the lack of obligation -- you stop liking it, you can stop using it and there is no wasted financial investment. Not to mention that you can have fun educating yourself and getting it back to where you like it again.
Give me that holographic 3-D translucent panel that I can throw data at by waving my hands around. As long as it runs a kernel with a UNIX philosophy and I can compile the entire thing from source like my current Gentoo distro I'll be happy.
All I ask is that you don't F' it up. If you make the decision that I don't *also* need a keyboard and a console window because 'who uses VI anymore to program when you can wave hands around' then you're full of it. I'm the one to decide if hand-waving is better, not you. If you toss out a half-done re-write like KDE 4.0 with regressions on every major integrated application, you deserve the hate. If you break the entire metaphor like Unity or Windows 8 did for the sake of some other platform you deserve the hate. If you abandon decades of proven philosophy on a whim just because, you deserve the hate.
On the other hand, if you have something truly unique, revolutionary, game-changing, bring it on. If it is truly a step forward the world will quickly abandon the old in favor of your new, my old self included. It's when you try replacing the old way forcibly in favor of your new that you fail. That's not your job. That's my (the user's) job.
Linus is currently using Gnome 3 again.
only to see GNOME do the same thing not so very long after that, in the grand scheme of things. That's when I went to Mac OS from Linux (having been a Linux user since 1993). And that was that.
I stuck with KDE4 for at least 2-3 months. But it was a lot of meta work (i.e. work on the environment itself, rather than work *in* the environment). By the time the desktop came back in some form, I was long gone.
At the time that I left KDE, I had been a KDE user since KDE Beta 3 (1.0 beta 3 that is), having switched from TWM and an old .twmrc file from my SunOS days. In fact, I wrote one of the earliest reviews on the web for KDE (during the early beta phase) to be published by (at that time) a top 10 internet property. I was one of those early "Will Linux someday overtake Windows on the desktop?" pundits on the strength of what I saw in the new KDE platform. I was a longtime, committed user. But the total break in the workflow from 3.0->4.0 was unforgivable. KDE 3.0 was chintzy and showed its Linux/X heritage far too much, sure, but 4.0 was flat-out unusable for the first months. It eliminated common workflow assumptions *at the same time* as being so buggy as to fail to do anything predictably. The result was that you never knew whether you were looking at a behavior (an unexplained one at that) or a bug. But it didn't matter, because something different would happen the next time anyway.
When I went to GNOME I found it to be usable in ways that hadn't been true in the GNOME 1.0 days (GNOME 1 was a disaster; tough enough to keep it running, and see the "integration" at work, tougher still to actually use it). So I settled into GNOME and never logged into KDE again.
Then the "press" began to hit in Linux circles about what was coming for the GNOME 3 release, and about the adoption by major distros. I tried the stuff in the dev repository, decided to hackintosh my Thinkpad T60 on a spare partition to give Mac OS X a go, and three months later, after having been a Linux user for going on two decades, I bought a MacBook Pro and haven't had a Linux partition or installation around since (well, except in my phone and router).
Even veteran developers are relatively attached to their workflows. You have benchmarks to hit, as a general rule in modern life, and they do not involve configuring your desktop. Any time spent configuring/learning to use a GUI all over again is, quite simply, is money lost.
In simpler terms than all of this discussion, that's where GNOME and KDE screwed up. Whatever you think of the theory behind the reimagination of the Linux desktop UX/UI, the fact is that there was no demand for it. Like all open source projects, it happened without any particular concern for whether there was demand or not, or for where demand was pointing.
If the Linux world had collectively been interested in driver support, OS X level polish, and interoperability with the most common/dominant commodity and infrastructure systems in use over the '00s, Linux might be *the* dominant operating system today, running the bulk of cloudspace/serverspace, the bulk of mobile computing/phone space, and the bulk of desktop space. Instead, the fatal flaw of open source software kicked in—nobody has to think about the market. The developers had their freedom, and they exercised it.
And the result is a bunch of Netcraft confirmations that Linux on the desktop is dying. (Has died?)
STOP . AMERICA . NOW
...that pressing the super key (aka windows key) and typing is not an innovation exclusive to windows 7 don't you?
IIRC win8 retains that ability though I don't use that os. My regular desktop is GNOME 3 and it works just like that too.
The thing with Win8 and GNOME3 is that there is so much angst over what amounts to the introduction of a full screen launcher to replace a stale but familiar cascading drop down menu launcher. In both cases once you launch the same old apps all that crap is out of sight.
Of the two however GNOME 3 is clearly superior in my experience to WIN8. Microsoft went even far beyond GNOME in hiding functionality--at least GNOME retained their equivalent of a start button. Also windows is a confusing mess because it presents the launcher as the application environment...but just for metro apps. Then there is still the old desktop...but without a visible launcher (until 8.1 anyways).
At least in GNOME 3 there is still a clear division...it is clear when you are on the desktop running apps and when you are in the launcher/Switcher. It still hides a bit too much config but it is evolving faster than windows and plugins are quite useful.
In any case I spend most of my time in a handful of apps, text editor and terminal and just tab between them so the desktop environment makes little difference to me.
The one thing that actually surprised me was how much faster beginners and casual users caught onto GNOME than win8. The latter had them mystified, especially coming from XP. Despite being different GNOME was much more intuitive for the most part. Both, however, caused power users much frustration because of their instinctive desire to tweak their environment. Casual users have no such compulsion...their focus is the apps not the environment they are hosted in, and if apps are easy to find and launch that is all that matters.
I was thinking along the same lines. This should be modded up insightfull AC - oh wait, this is slashdot!
Look, I've tried setting up Linux for end user typical office use, circa 2002 and a few years down the road. Like many here, I've tried alot of distros. While Mandrake looked fresh in the end 90s, it wasn't until Ubuntu Linux vendors really started serious attempts to grok and get valueable feedback their users, as this costs serious money. KDE before that looked like crap, and while userfriendly for power users and quite "similar" to Windows, KDE wasn't offering more than cloning what people already "had" in Windows, so was not compelling enough at that time.
Now Ubuntu has screwed up in so many ways: Spying on their user (phoning home every search request) and Unity. Like everyone else they've just fell victim to their own success and have no real clue what to do next, so like Microsoft and Apple, keep screwing their talents and botching their results. The culture is not there, just a few people with the right ideas, who after a while, leaves without instilling the right mindset in the culture.
Look, UI is *hard*. To come up with something everyone understands and likes, is an impossible task. The reason for this is that you typically have FOUR classes of users:
1) Newbies have no clue and just want "IT to work". For them, tablet-interface *seems* more inviting and simpler to use. However, the danger with tablet interface is dumbing down your users! Newbies are trend-chasers (ie. always gets caught up on the wrong side of the trend).
2) Power Users have experience and know approximately where the required bells and whistles are. At one time, Power Users were newbies and had to learn it all somehow. Maybe they were lucky and have someone show them the most important bells, or maybe they had enough motivation to learn it all by trial and error. The problem with Power Users is that when you change things, they become uncomfortable and lose all that hard work learning all the warts and features. Power Users know enough to create trouble for themselves, and to confuse Newbies with erroneous information and knowledge. Power Users are trend-followers (ie. have enough time and know enough to stay current).
3) Super Users are long-time Power Users. They configure systems for themselves, and often also for other users. Super Users know enough to screw up their own systems on a regular basis, and often do because of lack of integrity and safe controls in such systems. They also know how to create stable systems, although it's boring not to expand the bleeding edge. Super Users introduce ALL THE NEW THINGS. Even though not readily apparent, people tend to remember what they recommended and get independent confirmation of their knowledge. Super Users are trend-creators (ie. have way too much time on their hands, so influence what is coming next).
4) And then you have the Programmers! They can be any one of the three classes of users at the same time, and often have NO CLUE about UI and what the different classes of users actually need!
My honest opinion is that if an UI cannot *easily* be tweaked to accomodate these 4 classes of users, it will fail one of them.
'nuff said.
As a Super User and Programmer, I just want it to work, without spending hours fighting the system to do what I need it to do.
Current setup:
Linux Mint Debian (rolling updates, no periodic reinstalls)
VMWare / Virtual Box for Windows applications
Steam for Linux gaming
dconf-editor for finding those configurations you miss (power management, screensaver options, lid events, etc. Somehow Cinnamon doesn't think you need those! There's always something missing in Linux, but Linux Mint Debian is close to perfection in my book, however, I can figure out the warts where many Power Users would give up.)
I have a 24 inch 1920x1080 LCD monitor. I only go fullscreen if I'm watching HD video, or working on a large photo in GIMP, or working on a honking big spreadsheet. I often have two web browsers side-by-each at 960x1080, or 2 or 3 xterms open.
I run icewm with multiple work areas, grouped by tasks. The taskbar enables me to launch stuff, and then it gets out of the way.
I'm not repeating myself
I'm an X window user; I'm an ex-Windows user
What you see is various software packages all reinventing what should only have to be done once, right.
Various people have invented corkboard ideas, on the mac Stickies is post-it notes and Scrivener is a research and writing system with corkboard as part of it. I have seen various drag and drop style interfaces for drawing uml or configuring networking. One package I am involved in now has a canvas you can drag and drop nodes in a flowchart.
Personally I had an idea for a tool that would draw on the desktop and define regions of it.
Currently the desktops I have seen are just a blank screen that inevitably gets filled up with crap which then has to get put somewhere, or it is just a few shortcuts. The manu bar (on a mac), the trashcan and doc are the only actually functioning items.
I would like to propose that the desktop should be an object oriented scriptable canvas with some intelligence, with storage, networking, layers, ability to transport them between instances and platforms, and something that actually helps you do your work. Smalltalk comes to mind. Anyway, my two cents. There is a lot of screen real estate but none of the operating systems actually do anything useful with it. The drawing tools that are out there in powerpoint, libreoffice or whatever are pitiful and unintuitive, so it takes a lot of work to make something useful and you don't use them in a meeting to illustrate something, you go to a whiteboard and scribble something illegible. Or you get out a big piece of paper. I'm saying a strong canvas with simple unbloated widgets in place of the desktop would be extremely useful as a standard computing component, instead of using the tons of little widgets that solve little bits of the problem.
The Windows '95 style desktop captures an ideal well. The MacOS 9 did as well, so does the original MacOs X design. The trouble is that they are far from perfect, but small deviations from the ideal makes for no real improvement, and large explorations away from the ideal tend to be horrible. Kde were probably the first to really wrestle with this horror, and they seem to be past the worst of it, though I have no idea where they're headed; gnome is trying to do the 'HUD Overlay Control / MoreThanAToyDashboard' thing which would be good if they work out how to do it well. But the Windows 95 style, with Gnome2 additions like desktop toolbars you can just stuff what you like in, are a good ideal.
I wish they would invest more effort in making it easy to repurpose what is there: make a simple Gnome Terminal a 3-line program:
w = Gnome.Window(menu=DefaultMenu,content=Terminal.Default)
w.content.run("/bin/bash",["-"])
w.show()
and make customising what the terminal window does as easy as this. Do likewise for getting a browse component running (recall the Cocoa browser demo on MacOsX?)
We should be working less to come up with new stuff, and more on making what we have both rock-solid, dead-simple, and as trivially easy to implement as possible. I wrote a little note about this idea on my Wiki: http://thewikiman.allsup.co/Im... where the idea is to write your program in your ideal imaginary language, and then build from the language you have towards that ideal, striving to make as much of what you produce reusable in other projects (and to share with others so that they don't have to repeat your progress: DRY is good, DRIP (Dont-Repeat-If-at-all-Possible) would be a better principle, since with proprietary software the DRY principle is violated whenever a different developer has to rewrite functionality because of legal or lack-of-source issues).
John_Chalisque
Further, the one full screen app at a time design is easily accommodated by a desktop system just by maximizing the apps. However, a tablet model GUI cannot accommodate those who prefer the 'classic desktop'.
So it makes no sense to rip out perfectly good functionality and leave the power users in the cold when the most they need to do is add a setting to have apps start maximized.
It's not innovation people are resisting, it's needless design churn and designers who want to force everyone to use the computer the way their grand vision says is most aesthetically pleasing (and to hell with productivity, functionality, and user choice).
A desktop has 4 main tasks
1: Allow me to open the applications I want to open
2: Allow me to switch between application windows
3: Show me the general information I need (time, battery, wireless, etc)
4: Perform 1-3 as efficiently and aesthetically pleasingly as possible.
I think DE developers screw up in thinking that the DE is the killer application, when they get this mindset they start trying to do too much and be too innovative and it comes at the expense of 1-4. I think this is what happened with Windows 8 as well as the 3d eye candy that was infecting the Linux desktops for a while. The DE isn't the killer app, it's the thing you use to open the killer app that should get out of the way afterwards.
I stole this Sig
http://en.wikipedia.org/wiki/I...
xmonad all the way! I switched a couple years back. The lack of a nice settings file* or UI dialog meant that it took me a couple days to get everything the way I wanted, but now I am completely happy. I just never realised how little a window manager really needs to do. If I see people working with actual windows (i.e. floating windows with borders, close icons etc) it just looks so clumsy.
For me, having a numbered desktop for each task I'm working on and not having to bother with moving/resizing windows, plus a WM that is fully functional within half a second after logging in, even on a 4 year old computer, is a great boon to productivity.
*) for those not fluent in Haskell...
(ducks)
Sent from my ASR33 using ASCII
Wheres LXDE in your research?
In terms of performance and usability, its the only "classic" UI out of your list.
Unless the new "classic" is slow performance and bloated interfaces?
Gnome2 was long time my desktop until I've seen what they plan for Gnome3. I also tried Ubuntu to see how Gnome3 works. Gladly, FreeBSD did not import Gnome3 long time, so I could use Gnome2 a bit longer. But at that time I also began to learn about alternatives and notice how much crap Gnome2 and Gnome3 has running in background. I started to learn how a TRUE Xorg desktop should look like and decided that KDE and neither Gnome are like cancer to Xorg.
Someone on IRC mentioned Ion as a tiling WM. I got interested and tried it. At that time I was really lazy with configuring desktop manually. Ion3 had a license dispute with FreeBSD and it disappeared from ports. I decided not to use it, because all the resulting annoyances. But I remembered tiling WMs as being useful, IF one wants to use keyboard hotkeys.
I tried to go with Fluxbox... but it was a bit annoying at some points and I changed to the sister project Openbox. I liked Openbox very much. I also noticed that a terminal is far more important. I started to ignore file managers from now on. This has been an important change. I always preferred a desktop with a decent file manager, because one thing I considered to be annoying is doing simple file operations in the terminal. It changed to the opposite now and I rather began to think how to write efficient desktop macros and scripts to make everything more fluent. I also themed my desktop in different ways and decided NOT to use any eye candy, because it is simply not improving the usability and neither efficiency. One thing I learned with Openbox is that vertical screen space is important and that I prefer to have my windows at the left screen edge. Consequently, I've put the lxpanel at the right screen edge. What also was very important for me is to manage my desktop configuration in Git, because I re-used my desktop on other PCs and merged the differences for special parts in a way that I did not need to think about what platform I use and which host it is.
I was pretty satisfied with Openbox so far, but I decided not to sit on it all the time, because I might ignore the developing new desktops. I tried Gnome2/3 and KDE again... and found them HORRIBLE(!) after using Openbox for a longer period of time. I decided to look at other WMs and found Xmonad which people mostly laughed at as a joke being that minimalist. I tried it... and FAILED horribly... I noticed that I need to learned Haskell... at least a bit.
It took a long time... I learned some Haskell... and it is quite fun, I can tell you. I tried Xmonad again and copied some of the useful configurations. I also put them in my Git repo. I found it far easier to manage different platforms and hosts, because Haskell is a full-featured programming language.
It started as a joke... Xmonad... but after configuring the details, I can see how much more productivity and usability increased. I admit that I would not recommend to anyone to use Xmonad... by why the hell do I need to tell you what to choose or convince people about efficiency of Xmonad. This is YOUR personal preference and it always ends in annoying flamewars. I just recommend one thing... take a look at the desktops... it will take some time... and you will even need to (OMG!) actually LEARN something to use it properly.
And one thing is sure... I will never say that Xmonad is my last word... I will take a look at everything I find and never give up to give a project a chance... even KDE and Gnome (when there are some notable changes).
In my particular case, I didn't become more pragmatic as consequence of my "linux needings". It still makes fun to play around with archlinux and make the system run with as low RAM as possible. However, I do not have the time or the mood to play anymore so long. Just wait until you get a child.. I swear you will become more pragmatic!
a few years back it was the case that you had to fiddle about setting the drivers up to get things to work in linux.
but have to say its now xp that you find your still hunting for disks, to install the driver yet again because, the device is plugged into a different usb port than the last time.
yea there are still a few devices that don't work in linux but not many.
system admin mostly automatic now needing almost no attention
the main reason i find linux better is from the admin and setup point of view
with linux
normally a 2 hour install including updates and most of the apps and the system is back pritty much as before usualy without disrupting any
of my previous data
with windows
maybe a couple of hours backing up old data first
having to check for viruses on that data
yea it installs in 30 mins , then the week or so hunting for driver disks and applications disks (keycodes on boxes etc )
the last time did it it was 30 or so restarts and a couple of days later that it had managed to get all the updates
i dont know about the rest of you but actually using the apps is more important than having to do sys admin and linux is way better
but i guess the guys at microsoft enjoy it so much that's the way they take you so why the slogan "where do you want to go today " when you know the answer is to the it desk as sys admin is great fun "
Why would anybody laugh at that? I've used FVWM for years, and I don't intend to switch for many more years. It's basically MWM with virtual desktops and total programmability. It basically does all the things that a window manager needs to do. New versions still come out (the last version came out less than 2 years ago), and it has all the bells and whistles for a window manager like antialiased fonts, transparency, etc., and it's highly programmable and customizable (you run scripts and issue commands at runtime even).
If you really want primitive, look at TWM. That's basically stayed the same since the 1980s, and can be seen on many old workstations of that era. It's also common to see TWM on servers that need an X11 session, but don't need a full desktop environment, like if X11 is only running for remote VNC access on a headless server.
Systemd: the PulseAudio of init systems
[Linux desktop users] know what they want — a classic desktop — and the figures consistently show that is what they are choosing in far greater numbers than GNOME, KDE, or any other single graphical interface.
To me, it makes no sense at all. In my non-native English understanding it would mean, that Linux users preferred classic desktops in comparison to either GNOME, KDE or any other graphical interface. Hell, what do these users, the majority of the Linux users, use then?? If neither GNOME, KDE nor any other GUI? CLI???
sir, the first sentence in your post just states an obvious fact.
the rest, however, of your verbose discourse is just empty, based on one single inventend assumption, that "what to do next" thing as being related to a particular desktop metaphor. i see you like to theorize but you should at least attach to reality when doing so. general industry desktop user (from office to cad through programming) does not depend on screen organization for his mental workflow at all. in fact, most use a limited set of tools they switch to over and over again, and that's it. say email + browser + 1-3 domain specific tools, most of their time being spent on those specific tools (which might even run inside the browser). it's not the layout of these tools what prompts a decision to switch, it's their current workflow. so in fact ANY window mgr would do as long as it provides any form of instant switch function. user's don't have to "think" it's time to compose an email, or make a draft, or check a balance. they know, and they also know which is the right tool for that. what they need is just a convenient way to bring it up, instantly, be it a keyboard shorcut, a click on a panel, a console command or a voice command, that is pretty irrelevant and just personal choice.
i have to add that it is precisely this kind of out of the blue theroizing what is producing the ui monstruosities we're seeing nowadays. please get a clue, or do some real work, or at least take your time watching someone doing it.
When you see a person digging with bare hands while a spade is lying right next to them, you will probably raise an eybrow.
If said person is at the same time whining about 'How slow this works' or 'How long it takes to get this done'... well, then you have your typical windows-user.
Hopefully KDE will learn from this to never do a point release until it's ready.
Time is what keeps everything from happening all at once.
When you first start Openbox there is nothing on your desk, and it takes a bit of work for anything to be put on it. Stuff on you desk means there is stuff you need to be doing. When GUIs that try to guess what work you should be doing are a no go in my book.
Time is what keeps everything from happening all at once.
I don't ever recall a time when the Linux users around me would anxiously toss out their current software stack without a moment's hesitation. Did we play around a lot with new things? Sure we did, and still do! And then we move on from the things that we're not fond of while we allow the things that we are fond of to grow on us. KDE Plasma and GNOME Shell are dramatic departures from their predecessors. I don't recall a time when the developers of either claimed any different. Not that I'm terribly fond of either, but every year they each look a little more practical. Then again, I'm a Blackbox user.
There's the "heavily inspired" *nix version called "midnight commander" or "mc" as well. I still use it at times to delete files with odd characters in them.
New people came in and decided they had to crap all over everything to show that they were there. IMHO that's why there is deliberate breakage - a DLL hell type situation where completely different things have the same name as other things so old stuff depending on other things will not work in the new environment.
I wish I could disagree, but I help so many users that run one program full screen. I just sit back and shake my head as they constantly switch from one program to another instead of arranging the program windows to see everything they need at one time.
It really start to piss me off when they have two monitors and switch between two programs, both on the main screen, both full screen. Then they wonder why it takes so long to get things done.
Nobody ever seems to mention this way: most windows I do use maximized, but I just ALT+TAB between them. Lightning fast.
And yes, I'm still a freakin' "power user". Maybe even more so, as I'm capable of simply remaining aware that other windows are open, instead of needing small windows overlapping each other all over the place.
Which level of the strata do you mean :)
there are 30 years of detailed field research on this. Again, see Suchman's "Plans and Situated Actions," Dourish's "Where the Action Is," etc., or visit the ACM digital library and look at usability research (i.e. involving observation of real people in real settings) in CSCW, HCI, etc.
You have one basic fact wrong: they *do* have to think about what it's "time" to do.
Users in computer-at-desk contexts do not have a detailed roadmap for what to do on a click-by-click basis, either from their boss or inside their heads. They have a general set of goals for, say, the quarter ("Get this project launched"), perhaps the week ("Make sure everyone is on-task and progress is being made; keep the CTO appraised of any roadblocks"), and the day ("Put together charts and graphs for Wednesday's meeting to detail progress").
But it is *these* tasks that are "theoretical" quantities. They translate into dozens and dozens of clicks, mouse movements, UI interactions, and so on, many of them interdependent (or, in Suchman/Dourish terms, indexical—that is to say, order-important and constitutive of an evolving informational and UI flow context).
The user may have "Tell bob about tomorrow's meeting" already decided, but they are imagining Bob and imagining Bob *at* the meeting. From there, activity is practical and adaptive. They emphatically do *not* have this in their heads:
- Take mouse in right hand ...
- Flick mouse to lower-left to establish known position
- Move mouse 5 inches toward right, 0.5 inches toward top of desk to precise location of email icon
- Click email icon
- Wait 0.4 seconds for email window to appear
- Move mouse 7.2 inches toward top of desk, 2 inches toward left to precise location of To: field
- Click to focus on field
- Type "Bob"
- Wait 0.1 seconds for drop-down with completions to appear
- Hit down arrow three times to select correct Bob
- Press enter
You laugh, but in fact this is precisely what you're suggesting: that users have a roadmap already. They don't. That's why we invented the GUI—to provide a visual catalogue of available computing resources and an indication of how to access them on an as-needed basis. Then, the user has to decide, in the moment, what was needed. Every single attempt to make things more "simple" or more "efficient" by presenting *only* that one thing that designers imagined to be needed at a given time—the "obvious" next step—has led to users that either feel the system is useless, that fight it to get it to do what they want, or that simply go around the system (I'll just do this task offline, on a pad of paper). You can make very telling changes to users' productive workflows and levels of productivity by changing orderings or locations of icons, etc. Marketers also know this very well on the web (google "page hotspots" to see the research about positioning of advertising and how deeply it affects CPC and other factors in online marketing).
At a less granular level, something like "Get this project launched" is also not available in a detailed roadmap to a user. Go ahead, ask them to elaborate on the precise set of tasks involved in their big quarterly responsibility. They'll come up with 20, 30, maybe even 80 split into four or five sub-areas. But getting the project launched for an average middle manager over the course of a quarter involves tens of thousands or even hundreds of thousands of discrete actions, gestures, etc., some computing-based, some not, with the computing-based ones split across dozens of applications and contexts.
It cannot be mapped out because it is contingently assembled—it has to be done on an as-we-go-basis. So the tasks in the "to do list" (and, in fact, in cognitive behavior) are theorized ("Create a new instance of the platform on test VPN, set up credentials for team") rather than existing as a detailed, moment-by-moment list of actions. This is why user docs people actually have to sit down and use the system, and int
STOP . AMERICA . NOW
We're conflating use cases and identities when we say "Newbie." As technology designers, we need to be concerned with use cases. There may be a statistical overlap between the two, but mistaking one for the other gets us into deep water for design purposes.
Rather than newbies, let's talk use cases.
Case #1: User is not "at desk, at work" but is rather "in flow, in everyday personal life." They need, for party-planning purposes, or for kid-care purposes, for example, to "send an SMS, "send an email," or "buy more diapers on Amazon.com." These are use cases that are all much better handled by tablets or mobiles, particularly if the user does not spend most of their work or personal life sitting at a computing system. The larger computing system imposes an extraordinary amount of overhead for (say) the stay-at-home parent that just wants more diapers. Leave the playroom, go to the den, power up the desktop, sit down, confront a desktop full of resources, figure out which one is the right one, start the application, and so on. All of that is overhead when we have mobiles: pull iPhone from pocket, press button, tap Amazon, type "diapers", click Buy, put phone back in pocket.
As technology folks, we have a terrible habit of taking someone's bewilderment to mean that they need more training or they're a "newbie," rather than looking at it practically: they're being told that they have to do an *awful lot of work* (moving through the house, navigating a full suite of powerful computing resources, learning to manage them) just to get some more diapers in the midst of their *real life*, the one that they actually care about, which involves diapers, not computing.
Case #2: Person new to computing is also new to the job, but it is now their *full time job* to make charts and graphs with Excel. They will happily sit down with the 600 page book, online training tutorials, and get to work learning. Why? Because this is a set of resources that are not overhead to them—it is the productive work that they will be expected to do, so the investment in time and computing use makes perfect sense. It is work, not waste.
I'd argue that in most cases, trying to marry a full-on computing environment (hierarchical file system or DB storage in quantity, multiple applications, multiple peripherals and forms of connectivity) to a rapid, task-based interface is not going to work out because they're two different use cases. Rapid, task-based use demands lightness and speed. General-purpose "big computing" resources toward the achievement of office work demands feature-richness, open-endedness, and deliberateness (i.e. the opposite of lightness and speed). One is highly endpoint-oriented, the other has no endpoint and is highly process-oriented.
The right answer is not to redesign the desktop environment. The right answer is to get the stay-at-home parent an iPad, or a laptop with everything but the web browser uninstalled, one that preferably boots straight to the web browser—in which case, the UI doesn't matter at all, because the user has no intention to use it.
The "newbies" that we commonly reference are actually a use case—people that feel that their current goals are not well-served by a high-overhead investment in full-scale computing. To serve their needs with a full-on whitebox computer, we just have to strip out the general purpose computing entirely, or at the very least, hide it entirely—which makes the system all but useless for those embroiled in "general purpose computing" use cases, particularly in comparison to those that have a full desktop UI available to them.
Make a better desktop environment *and* make a better information appliance, and both sets of users will thank you.
Try to make a desktop environment *that is* an efficient information appliance, and the computing-for-work people will find it to be inefficient and unhelpful while the casual-net-users will find it to be slow and needlessly complex in comparison to their sister's iPad.
STOP . AMERICA . NOW
Do they continue to be gainfully employed as a digger, yet still dig with their bare hands?
What do they and their boss know about their productivity and job requirements that you don't?
What are they digging for? Is it likely to be damaged by a spade? Are they relying on the tactile sensation in their hands as they dig to make critical digging decisions of some kind? What is the cost of spades? What is the urgency of this dig? Is the limited supply of spades reserved for cases in which rapid digs are needed, in order to avoid excessive spade wear? How long do they dig? Does the spade cause repetitive stress injuries or blisters that hamper their work later on, and for longer periods of time, despite the apparent productivity gains early on? Even if we go all the way to the silly end of the spectrum, are spades against their religion? Even if so, are they nonetheless the most productive member on their team even with bare hands, leading the boss to not give two damns whether they use a spade or a ball of cotton candy to do their work? If you mess with the magic sauce that makes them the most productive person on the team, are you going to be out of a job before they are, even if you believe that your orders for them to change are the "correct" ones?
It seems to me that the job of tech designers isn't to know about digging, but to listen to the diggers carefully as the experts on their kind of digging, digging needs, and the totality of their work life as diggers, and to thoughtfully provide the technical resources needed to enable diggers to do digging as they see fit. They are, after all, the diggers. We are the tech people. Our job is to make tech—which is merely a means to everyone else, not an end. Make the wrong means that doesn't help them to achieve their ends, and you will find that nobody values your tech, no matter how much you try to explain that a spade is a spade.
STOP . AMERICA . NOW
Install kscreen and you won't have these problems any more: http://www.afiestas.org/kscree...
I no longer use windows xp, but I use windows 7 but in windows classic mode which pre-dates xp but comes from windows 2K. I personally find find it to be clean and uncluttered. I also have Linux Mint Cinnamon which is very similar to windows classic and as a metaphor it works admirably. I dislike the mac interface of one common menu and the unity emulation of that interface be it a side app menu rather than a bottom app menu. In fact I am writing this on Linux Mint 14 Cinnamon. Linux mint 15 and 16 work virtually the same way. I think that operating system vendors think they they have to have new UI's in order to resell themselves every 3 years. I think it is high time that Microsoft gave away the OS and just sold extra apps to run on it. I doubt I would buy any of their apps however.
... I help so many users that run one program full screen. I just sit back and shake my head as they constantly switch from one program to another instead of arranging the program windows to see everything they need at one time.
Power user for multiple decades here. I have a big monitor and I run all my apps fullscreen. I have for the last 15 years at least. I do this because whatever I'm looking at - code, shell output, web pages - I want to see as much of it on the screen as possible. Running things fullscreen also means window-management buttons, when I need them, are located at screen corners where they're easy to hit.
When I need to switch apps, I just alt-tab. It's super fast. If an efficient workflow that's different from yours pisses you off, I feel sorry for you.
... and realizing that dicking around with what is ultimately a tool is an impediment to getting useful work done. That's the realization I had. I used to delight in building my own computers from parts ordered online, rebuilding kernels to be lean and mean, compiling all software from source, tweaking things endlessly, etc. But somewhere along the line I became more interested in what I could do with the machine rather than the machine itself. Now I just want to plug something in and go.
Profiling someone of similar ilk is nothing but mutual masturbation. Everyone feels good, but nothing productive happens.
I've played ages ago. But that was then and now
bored of that. So just want things to work. Still like to play but more interested in other things now. For me Android showed me how it can be and since then I don't want to play with interfaces.
I don't customise my phone anymore, I don't even play games.
I don't know... just prefer more interesting projects now
A blog I run for the wealth
Between Unity just plain sucking, and Linux Mint update crashing my os... I'm done with tinkering. I want something that just works. I'm back to windows
Question assumes that since we don't want Gnome 3 or Windows 8, we must be opposed to change. {Sarcasm tone=thick}Yup, that's Linux users all day. Totally. {/Sarcasm} No, Gnome 3 and Windows 8 just got it wrong and shoved it down our throats.
I think the only real difference over time, re: your question, is this: once upon a time, it was a race to test the waters with ideas and alternatives, a race to refine a range of UX choices. Most of them imitated something else, a few didn't. Most of those alternatives are still popular. In fact, XFCE is (I believe) still gaining popularity. It's too soon to declare hegemony.
I wonder constantly why "always on top", "auto-hide", and "click to hide" are our three main options for dealing with taskbars. With the possible exception of the clock and recent notifications, everything but the guts of the apps you are currently actively using, are a huge waste of screen space when they aren't being used. Just because Gnome 3 and Win 8 happen to agree with me on that point, doesn't make them any good.
It's also telling that Windows 8 is all but being recalled. Microsoft didn't ask users what they wanted, changed everything, and claimed it was based on what users wanted. Gnome 3 and Windows 8 both did this. Both can burn in obscurity for that single crime alone.
You don't unilaterally change the lives of hundreds of millions of productive people, and expect good results. A nextgen UI paradigm may need to wait for nextgen I/O devices, but it might instead just try iterative feedback instead of testing our gag reflex.
Every trollism an AC posts is prefixed, in my mind, with "A. Coward whined, in a weak and cowardly voice:"
Hate Windows 8. Hate it! Go the latest Ubuntu. Hated it. Looks like Windows 8 . WTF? Got Mint. Looks like Windows 7. Very happy now :)
I did get a life on Windows 7. Which sucks but makes me miss KDE 3.5 less than everything else.
To me it wasn't that they promised to fix everything else later, they claimed everything was good when 4.1 was released. Look at Sergio's comments on SVN's call to fork.
I disagree strongly. I hate the design. I miss kasbar. I'm not sure how to say that enough actually. Losing kasbar is like ripping out part of my soul. I dislike the model desktop editing. It isn't just the botched 4.0. It isn't just the claims made when 4.1 came out and everything was "done." I honestly and sincerely dislike the direction of the design of KDE 4. The fact that the implementation also sucks can be the final nail in the coffin.
Linux users, as the article says, spent "years of tinkering" into getting a desktop to work exactly like we wanted.
That was achieved after some time. At least in my case, since about four years I have a desktop that exactly does what I want it to do exactly the way I want it to do it. So there is no longer a need for any big changes.
It's not that we "want or don't want change", it's just that we want change where it is needed, not change just for changes sake.
I like that idea. You could have modified commands like 'ls' that would display little thumbnails of each image, and a 'more'/ 'cat' commands that would just dump each of those images at full size or scaled to fit the terminal.
Vintage computer adverts: http://www.vintageadbrowser.com/computers-and-software-ads
using too many words.
Quite amusing that your post went on to be so long. :-)
Regarding your reference to the usefulness of the desktop space for putting objects on. I think very few people have used it like that, because most people have had limited sized screens and so have used windows that are either maximised or covering most of the desktop. So actually accessing objects on the desktop would "classically" involve closing, minimising or moving applications windows(s). Later OSs have provided hotkeys for temporarily hiding application windows, but it's a poor workaround for a fundamental problem that the desktop is always at the bottom, with object icons stuck behind windows.
In reality I think most people put their object icons in a file manager window, where that is allowed, as that can be brought to the front, over the top of application windows. Thus making the concept of a desktop irrelevant.
What I meant is "disk appliances" from EMC etc took over from MS fileservers. However I didn't think of Sharepoint (it used to have a fatal flaw of embedding large files inside a database that slowed it down a lot - fixed now) and that MS SQL has been chewing into Oracle's former space.
I wish I could disagree, but I help so many users that run one program full screen. I just sit back and shake my head as they constantly switch from one program to another instead of arranging the program windows to see everything they need at one time.
Depends on the size of the screen. For small screen sizes, you need the full screen for many applications. And with laptops being far more popular than desktops these days, and 13 inch laptops the most common, that's a lot of people.
Even those that are now using larger screens on desktops will often have learned on screens of 12-14 inches.
Also don't forget that whilst your work might be constantly multitasking, other workers often spend most of their time in a single app, and don't need to transfer data between apps. Less clutter for them is better - at it's most extreme are specialist text editors for writers, that are specifically created to remove all menus and other interface chrome so that the writer can concentrate on just the words.
The desktop metaphor is probably too complex for the average developer.
They never had a desk.
A desk wouldn't fit down the basement stairs, anyway.
--
Easy to learn: Hard to use.
Easy to use: Hard to learn.
Easy to learn and use: Won't do what you want it to.
If you're productive, I don't care how you work. When you're holding everyone else up, then 'your way' doesn't cut it and you need shown a faster way to work. In one particular case, the person reports to me and they aren't getting enough done in the day. It turns out that they're wasting a lot of time by working inefficiently.
Like the Edsel, in its time: http://www.youtube.com/watch?v... Or tail fins and chrome: http://www.youtube.com/watch?v...
I laugh at the fvwm bug reports from the 1990s that still haven't been addressed - other than that, it's pretty o.k. if you never change your desktop configuration.
TLDR: i3 is definitely worth learning to use; it is an outstanding window manager.
I use/used many desktop managers (spent considerable time with gnome (2.x and 3.x), lxde, xfce, and enlightenment17) and played a bit with several window managers (openbox, awesome, xmonad). i3 started with an already excellent window manager (wmii), but then blessed it with that final bit of intellectual consistency and rock-solid performance that a power user wants in such a critical part of their workflow. i3 has been my window manager of choice for several years, and I don't feel the smallest itch to change. When I'm trying to get work done I need all my screen real-estate, and I need my windows organized. I get this for free (it just happens as I open applications). If I want windows organized in a particular fashion, it's no more than 2 or three intuitive keystrokes away--I can completely organize my desktop in less time than it would take me to grab my stupid mouse to go and begin the arduous process of grabbing window corners and resizing all the windows under a more traditional window manager. Whenever I have to use traditional window managers I am amazed at how inefficient they are!
There is a learning curve associated with using a WM like i3. For instance, you need to already know how to start via commandline all of your applications or write something to parse and organize all your system and local .desktop files like this https://github.com/jtprince/do...
- I use a desktop for a lot of tasks simultaneously - but usually one task where interruptions mean a loss of 20 min to 3 hours work, depending on how important and attention-consuming the task. This is priority #1 for me. Gnome 2 and KDE are both acceptable, as is Windows 95+ (but not 8) and MacOS. Mobile tends to be acceptable.
- I need to be able to organize files : database/file management interfaces should make sense and function. (Windows 8 does not make this easy to find, and KDE and Gnome 3 actually have the best). Mobile misses this entirely.
- Ability to organize tasks when multitasking. Task switching needs to make sense, and multiple desktops are quite handy when having to do cross referencing or coding. I am a programmer - having one desktop for browsing, another for code editing and testing and a third one for communications is pretty much my ideal basis. Classical desktops (Gnome 2 and KDE - and I find Gnome 2 slightly better) do this quite well. Newer systems have lost this. Lacking multiple desktops tends to keep me out of Windows or MacOS, for the most part - although the latter is ok at them.
- Saving and restoring of state, particularly for console sessions. A lot of what I do still takes place in terminals, particularly when managing outside servers. Gnome 3 lost this - in a fit of "bug fixing" all infrastructure was removed and there is no real support for this now. KDE is ok, MacOS is quite excellent. Android and IOS are VERY good at this and I sure wish ssh and console was more usable on those platforms.
- Recovery after failure. A number of issues can cause - with power failure coming at the top, but not excluding non-fatal hardware faults. Mobile is quite excellent, MacOS is quite excellent and KDE is ok. Without the ability to handle state, Gnome + Windows do not have this.
While I miss desktop isolated task switching and the ability to return to the last task I was working in under KDE (IMO that's a bug), this is largely why KDE is now my default desktop of choice.
I really wish I could do most of my programming tasks on mobile, but there's no infrastructure for it - they have no ability to handle multiple tasks.
I kind of miss the multiple consoles of nongraphical linux - but it's a bit hard to do things like web browsing there.
Windows 8, Unity and other such "interfaces" are not desktops and have moved away from the ability to perform routine (as in : continuous and important) tasks.
These days - they sit down with the intention of completing three tasks, the same three tasks, in order, every time.
*Check mail.
*Update facebook
*Look at porn.
And they do all three with the same application.
Unicode killed the ASCII-art *
So unless you were merely unclear as to your issues, which appears to be true for the entirety of your post, you're not anchored to one PC.
What's unclear? I asked for an FOSS Google Docs-like office suite.
Google Docs works, but just as I prefer to store my TV on an FOSS DVR I'd also prefer to do my document editing in a FOSS word processor.
As to the rest of it, what are you on about? Google docs is already working in a browser, and being non-FOSS software has nothing to do with anything you're asking for, which is "runs under a browser".
Well, sure. But I can run Windows if all I want is an operating system. The benefits of FOSS go beyond whether it does the job.
You can mount a cloud service storage on your system and use libreoffice if your issue is that you want the data available and stored in the cloud.
There aren't exactly a lot of great cloud storage services that are FOSS even if that is the only part that you want to solve. I don't really see an FOSS alternative to something like Google Drive or Dropbox.
You can install libreoffice on any desktop with linux or windows on, which probably covers you for all your needs, there's no more need for it to be a browser app any more than the browser needs to be runnable under a browser.
Libreoffice requires X11 or Windows or OSX. My phone runs none of these, and neither does my tablet. My laptop definitely runs Linux, but I'm not actually sure if it even runs X11 (certainly it doesn't support installing libreoffice or connections from random X11 clients).
Also, libreoffice does not support simultaneous editing across multiple clients, and even if it did I'm not aware of any cloud services which do (including non-FOSS options).
I'm a long time Linux user, more then 10 years, and I'm still very much in the camp of configure it until it works and until it fits your need. I have no reservation about using pre-built, preconfigured software and I don't mind if I can't fiddle with the inner workings but when I can, I like to make my software experience fit like a glove.
For instance back in grade 12 ( 9 years ago ), I wrote a sweet FVWM2 configuration and when I use FVWM2 I still use it! When I run Gnome, I have the deal, I use a series of configuration file to alter everything until I get it how I like it.
Linux is great because it can really support all the camps of software users. The first camp is leave it alone and it's fine, these are typically Windows users. The second camp should never have to upgrade the system, the server admin's. The third is rock solid interface and your computer is pretty locked down, the Mac guys and the last camp is the touch everything until you love it. To say a Linux user fits into a single sterotype isn't fair, Linux users are as varied as all the other computer users.
It's no surprise that some of these environments were way too focuses on touch technologies and tablets, and left traditional desktops in the dark. Now that feedback has been received, the environments are backtracking and reviving the classic features. Which, make sense for most as the stats demonstrate. It's unfortunate this has happened, because it has added confusion for new users. But things are coming around and for example Gnome's Classic environment, which wasn't introduced until Gnome 3 was well established, will help considerably, for those that use Gnome for example
is that why we see more and more people using OSX at FOSS conferences?
Why not just use GNOME classic or Mate?