I have tried Unity and I really like it. It is simple and it does what I need. It respects my screen estate without taking functionality and clarity. I really don't understand what other power users are missing.

P.S. I don't like that Unity has some bugs but those are minor problems.

Geeks are coding away at some application feature, managing servers, updating designs, watching some TV show, browsing the web, discussing project goals over skype with team members and other mundane things – all this on a slow day.

I disagree. I have three windows open. Google Chrome (with, unfortunately for my productivity, Hacker News open), Emacs, and a Gnome Terminal (I X forward my Emacs session from my home machine to my work machine).

Even if I did have skype open rather than an IRC session (in ERC/irssi), I don't see how Unity is better or worse than the normal Gnome set-up. Alt-Tab still works. Same goes for other applications that are distracting me from my editor.

What part of alt-tabing and application launching does Unity do so terribly compared to the old Gnome interface + Gnome-Do (I hated the menus, I tab-complete in the terminal, why can't I tab-complete in my application launcher?).

Ubuntu's Gnome set-up was never been like xmonad or even a tiling window manager and neither is Unity. It seems to me that, as far as you're concerned, Ubuntu's default set-up sucks just as much as it always has.

edit: fixed mis-wording in second paragraph.

In the end, it feels much more natural and "spacial" - rather than having a small desktop with things piled on it, I have a context-dependent desktop.

Unity took me a while to get used to, but I've found it to be fantastic and look forward to seeing the production release with Ubuntu 12.04.

With gnome 2 I can put a seperate taskbar on each monitor and switch quickly between tasks with the keyboard or mouse. I also have another panel with buttons to launch my most common apps. Small amount of mouse movement and 1 click.

With unity I have to drag the mouse right to the far left or my furthest left monitor , possibly scroll through the icons , select the app I want, click it and then I have to select the window I want. Lots of mouse movement and 2-3 clicks.

Another problem I get is when launching a GUI app from a shell script , I load the app but then cannot dock it because it doesn't understand the relationship between the shell script and the app. This means I need seperate icons on my desktop to launch some apps.

My two problems with Unity: It's too easy to get to get to the launcher when I want to hit the browser's back button and all my Super-related Emacs shortcuts had to be relocated. Now I need a Hyper key.

From top to bottom I have file manager and browser, then all my apps that I need to have open on my launcher, so I just hold down my winkey then hit 3,4,5,6 and my mail, irc, trello, calendar all launch, then I'm done.

Problem with GUI apps from shell script is bug in my opinion but I was not affected by it yet.

To be honest I don't know what the Start + 1 shortcut does, so perhaps it would help or perhaps not. If you have to use keyboard shortcuts though I don't really see how this is simpler than the taskbar where I have a clear list of everything I'm doing on the most spacious axis of my monitor.

My girl friend, who is an Apple user, was pretty impressed by the user interface after I installed 11.10 on my Thinkpad.

Thanks to the SSD I don't ever have to wait, or stop to think what to do next.

I wouldn't call this ideal, but I will call the above _the_minimal_ viable windowing environment control scheme, for what I'm after.

How is it that you're so at peace with that glitchy simpleton? (Or am I over-doing it?)

(edit) formatting

With Unity, if I am on a workspace with no terminal, and I click on the terminal to open it, I am warped back to a workspace that already has a terminal open, and I have to open a new window from there, and move it back (with several more clicks) to the workspace I really want to be on. Much different than just clicking on the terminal icon, which worked before.

I'd like unity to work out, and if they could get past the awkwardness of task-switching and workspaces, then I'll go back to it. Maybe it would be best to just treat a new workspace as though there are no applications open at all, and if you want an application on another workspace you have to switch to that workspace first.

Also, it would be nice if I had a way to go directly to, say, the third instance of the terminal that I have open, rather than trying to choose it from a picture.

Selecting the right terminal if I have a bunch open is still a little awkward, because I can't just cycle through the ones on a single workspace. I'll switch to unity for a little while and see how much that actually bothers me.

It has a few bugs, but they're not that big a deal.

I don't know. It seems common sense that "non geeks" (hear, common people) like glossy things. After all, the Miss Universe thing has some success and is pure gloss. But I think it can also be an observation bias, or the result of a self-fulfilling prophecy kind of thing.

For instance, we check what people like to watch on TV, but what the people are fed with on the TV is already the result of some presupposition over what they like.

Some counter example exist, showing that people do not always like preformatted content, or glossy interface. Two different cases come to mind:

In France, there was a movie called Etre et Avoir, about a teacher in a remote countryside village. No professional actors, no story, just some shots in a school. Who would have bet on it? Is it not obvious that spectators want action, glamour, movie stars? Well in this case it was not true: the movie has been widely successful.

Another case, closer to HN interest: Is not Wikipedia the most successful (rich) content website ever? Who could say without laughing that Wikipedia has a glossy interface?

So I think one need to avoid clichés and presuppositions about "non geeks". They may or may not like Unity. If they do, it could be because it looks "glossy", yes, maybe, but it could also be because it just works, or because it is cheap, or because it is fast, or for some other reason.

Creating a good experience has more to do with creating some good interface metaphors that look good and work well than to just emulate known platforms like Windows and Mac.

It's like if you ask somebody why they find a woman attractive they are more likely to talk about her breasts , smile or legs rather than mentioning things like distance between the eyes, nose shape or cheekbone structure.

Indeed. And the concept of dividing the world into geeks and non-geeks is flawed to begin with, because non-geek covers so many different kinds of people, like kids, chemistry professors, Belgians, housewives, the mentally retarded, taxi drivers, muslims, the poor, the rich, and on and on. The "average user" doesn't exist (and isn't called Joe).

With Ubuntu 11, friends personally asked me how to switch back to the previous system, or to dump linux completely because the system got too slow with updates. And, sadly, that's true: the experience of the Ubuntu desktop, performance-wise, has degraded to the point of being significantly slower than windows xp or 7.

I have witnessed netbooks (decent netbooks running either android or ubuntu with unity) considered as "crap" by several non-tecnical people talking in the building, simply because they don't work "as expected" and thus being almost unuseable by them.

The way I see it, most people expect a standard windows-like interface. I've long stopped being a "go-to guy for computers" so I don't enter these discussions anymore, but that's very sad for me to realize that Ubuntu had everything two years ago for both technical users and new users, while now people simply flock away.

Even at work, I'm seriously considering to dump Ubuntu LTS on servers due to the increasing (unfixed) issues we have on large systems, mostly due to Ubuntu changes compared to debian and/or ubuntu projects such as upstart. Bugs on launchpad (let alone the crappyness that is launchpad itself) don't get fixed, but are simply "waited for" from upstream (either debian or source author).

I'm disappointed, really. Having the choice of the interface would have been so damn easy.

I also agree about your non-free software argument. Plus all the useless, additional services ubuntu is trying to push (Apple-style).

Where I don't agree with the author is the image of the "average joe". The characterization is possibly correct if you are talking about non-computer literate users ages 40+ or so. But in the age group <25 there are alot of non-geeks that go from InternetExplorer->Firefox/Chrome, MicrosoftOffice->OpenOffice without considering "gloss" at all.

Unity is invisible when I'm working. It shows up only when I invoke it with the <super> key. It uses screen real estate very efficiently. It lets me open my most often used apps with only two keystrokes (I've mapped <super><0> to <super><9> to my most used apps). It makes everything else quickly reachable with a few keystrokes via <super><a> and <super><f>. It lets me quickly divide screen real estate among windows with <Ctrl><Alt> and the numeric keypad. The next version (12.04) looks set to add many features specific to multi-display setups.

It's great for people who want their desktop environment to get out of their way and prefer to use keyboard instead of mouse.

* Even if it is not very intuitive, it is universally used (Windows) * Even if it is not universally used, it is intuitive (Mac OS)

Linux is currently, IMHO, in the middle. This brings up a lot of issues from an average Joe's perspective: some hardware configurations do not work perfectly (especially laptops), time spent in learning an interface cannot be applied in other environments (like office) and some tools are not intuitively named especially for a beginner (what is GIMP? Evolution?)

Just to be clear, I use Linux everyday and as an advanced user I love the freedom it offers. For the latter reason, I don't prefer Unity. Now, when a beginner uses Unity, they will approach someone like me. But since I dislike the unorthodox UI, I might be able to help her. But I am sure a fellow Unity enthusiast would be more than willing to help.

So my feelings are mixed and not as upbeat as the OP thinks.

I've had Windows 7, Ubuntu with Gnome 2, and OSX Snow Leopard and Lion in front of her. She was ok with Windows 7...it was tolerable, but we hit our first virus soon in and I ordered it off the machine. She liked Gnome 2 better, but it took a bit of getting used. Unity was pretty straight-forward for her...she likes it very much and prefers it above all others. OS X, however, was an entirely frustrating experience. I gave her my Macbook Pro because I was using my Lenovo Ubuntu notebook far more often and felt my hard-earned Apple investment was going to waste. She gave it a fair shot but HATED it. OS X's insistence on keeping applications open in the background even after the last window is closed slowly ate up memory. If two users are logged into the system and simply running a web browser and nothing more, the system crawls...no clue why. But the absolute worst part of her experience was the completely obstinate, bizarre OS X window switching and minimizing behavior. I realize many of you are accustomed to it by now, but she still rants about the fact that command-tab won't switch between all windows, and how minimizing a window requires right-clicking the icon to find it. Anyway, I'm sure there are ways to work around this behavior, but it should give one pause when someone calls Mac OS intuitive...I'd argue it's far from it.

What is the technical difference between a program that is running or not running? As far as I can tell it depends on whether it has an entry in the scheduler , performance then should only vary based on how many O(n) (or worse) routines there are in the scheduler where n is the number of processes including ones that are sleeping?

I think that measuring swap usage is a more important metric than RAM usage.

Yes, but let's say you open an email client to receive notifications. Then you perform various tasks that involve opening new applications for a limited period of time, and they keep running after you're done with them. Eventually they'll use all the memory and the email client, since it's been idle for so long, gets paged to swap.

Now you get a new email notification and you click to open it, which forces the system to page something else and "bring back" the email client, which creates a huge delay between the click and the email window showing up.

If the temp applications had been closed as each task was completed, the email client would've never been paged out, so it'd remain fast when it was needed.

This might seem a contrived example, but if you have some applications that you tend to leave open, it can be hell.

Surely it would not be difficult to have a paging algorithm that says something like "this program has no open windows or background tasks in operation you should prefer to swap this application's memory over others", no idea whether this is how it currently works or not?

I suppose there is the overhead of doing the swap operation which you would not need to do if you'd simply dumped the contents of the applications memory on the floor. I guess the memory management algorithm could just swap stuff related to state to disk but dump library code completely and reload this as needed.

Memory is so cheap now that people will usually have more of it than they are actually using at any one time so it probably makes sense where possible to still hold recently used applications (state + libs etc) in there if there is no contention for the memory.

I think my overarching point is that OS programmers are pretty smart , I would be slow to assume that they screwed something up in such an obvious way.

I often hear people saying that the Mac is intuitive. Is there actually proper proof for this?

Also OS X is generally allot more consistent in it's interface than Linux is.

1. http://developer.apple.com/library/mac/ipad/#documentation/U...

A lot of people. It's called "network effects": if more people use my OS of choice, I get more stuff for it, from support to applications to resources to see it develop faster.

The real question is "who cares what YOU are using?".

Why there always has to be one guy in every discussion of desktop Linux, going "I use it on my desktop just fine"?

We don't care, and it's not what's under discussion. Some people also use QNX or Plan 9 in their desktops just fine. That's not the f*n point.

The advantage of using a common OS is that other people have probably had the problems before you have , often on Linux I feel like I may the first person to experience a particular problem. More users = more software generally, cars don't require "extra" stuff that is designed for a specific car in order to do their job.

It is only if you have an exotic car^W computer that you'll have trouble finding parts^W software.

So, by you own logic, nobody cares to read what you and he are using, except you and him respectively. Which was kind of, my point.

"""We don't care what they are using. """

Good for you. This thread is about adoption, though, and about caring about what they are using --and specifically whether Unity is good for mass Linux adoption.

That is your logic, not my own. I care what he is writing because his perspective is similar to my own. I don't care what operating system he is using.

So, you:

(a) "care what he is writing because his perspective is similar to [your] own", but: (b) "[you] don't care what operating system he is using.".

I fail to see how (a) can be conciliated with (b), given that he is writing about "what operating system he is using", that is the very thing that you "don't care about".

It's not the only part where you defy logic and embrace contradiction...

Responding to another comment you invoke the car analogy to invalidate my argument, but when someone confronts you you tell him that: "It is only the people with exotic cars and computer setups (Linux on the desktop for example) that have a concern."

Are you not aware that we are talking SPECIFICALLY about the second case here, "Linux on the desktop"? Is there any special logic employed here?

Interestingly, if you are unable understand that, I don't really give a shit. I believe I have now clearly articulated my position, the rest is on you.

One of the worst analogies I've seen in my life. And I've seen lots. Where to begin?

(a) There are supposed to be people that "drive expensive cars that think that everybody else should too"?

(b) And (those mythical and "annoying" people) do so because that way the would have an easier time "getting parts"?

And all this BS (that doesn't even make sense, anyone here met a Ferrari driver that believed that everyone should drive one so he can get parts easier? WTF?) is supposed to be an analogy to people promoting Linux on the desktop?

This discussion is not about someone saying everybody should run the "exotic" Linux on their desktop, it's all about people making changes to Linux (Unity, etc), in order to make it less exotic and more fit for the desktop.

Except if you think that Linux will forever be exotic. In which case you are mistaken.

"""Interestingly, if you are unable understand that, I don't really give a shit."""

Interestingly? Hardly. You have a tremendous difficulty in expressing yourself coherently.

Reading comprehension FTW.

"""are just as annoying as people who drive expensive cars and think that everybody else should too."""

So, Linux people wanting "networks effects", are "just as annoying" as exotic car driving people that DON'T EXIST.

Writing skills, FTW. Maybe next time you can invoke unicorns to back up your arguments too. You never see them, either.

So you'll get 2 or 3 nice stable releases before being baptized by fire again.

I can also say, that many developers I work with feel the same way. It wasn't as if we didn't try to get used to it.

Unity is aiming for the desktop market, not the workstation market.

My desktop is Windows, because it allows me to consume media, browse the web, shove stuff onto my iPod with iTunes, and at no point do I actually have to think.

My workstation is debian with fvwm2.

Ubuntu is "linux for human beings", not "linux for development workstations". This is not necessarily a problem.

Unity has its bugs. In the end, they'll be dealt with. It's a 1.0 release, after all.

Worth noting that Android for all intents and purposes is fighting a closed, one-device (or at least one brand of device) Apple iOS. Commanding a place on desktops (not sure the relevance of this is still the same) will be arguably a different fight than on the mobile space.

Windows runs on all hardware, its a familiar albeit crappy at times interface. What tells me most that Linux as a viable desktop for non-geeks is a fairytale never going to happen is the fact that right now, apart from a select group of pc vendors, the only way for anyone non-technical to get access to a ubuntu install is to install it themselves.

It doesn't matter if its such a simple install, users are frightened to install a simple application at times.

On this note, I always find it mystifying who all these non-geeks we are trying to appeal to are, I'm all for a slicker, easier to use desktop but why not target it at the market which have stood by and held up Linux for all these years.

My dad, a non computer expert, updated to a recent Ubuntu and wondered where did all the menus go to. He called me saying that he needed help.

Even I'm a power user, it also took me time to adapt. The first days I was trying to do things the old way. For example trying to find the applications by clicking instead of typing. After I realized, I think I'm now faster than before. I was already used to Quicksilver and Launchy, so pressing the start key, typing a few letters and pressing enter feels fast and easy.

I was also annoyed by windows maximizing when I dragged them to the top. No one told me what was that about. Today I realized that I can easily drag windows to the left or right borders for having two windows side by side, which I find very useful.

So I think what's missing is a welcome video tutorial that explains the changes and tells us how to be productive with Unity.

I wonder about one thing though: aren't geeks doing most things with the keyboard? I don't think you notice Unity much if you stay away from the mouse.

On the other hand, they might've done this too late. Android will probably be the popular solution on all devices and machines that could have been running it.

The idea that you have to somehow have an apartheid for users is silly.

If I trade an economical Toyota for a Ferrari the mission statement of the car may be very different but the basic interface (wheel + 3 pedals) is the same. Apart from knowing that the gas is going to be way more responsive on the ferrari you don't really need to re-learn driving just to switch.

There's a famous russian chinese saying: When a rhino stares at the moon - he is wasting his spleen flowers.

(Not ones hand-made for 125$, obviously, that one ran Windows anyway).

1) Get people to jailbreak their tablets and install it. Fair enough but it will be very niche and more of a novelty, will never reach critical mass to get useful commercial apps. Will probably get picked by Chinese manufacturers who will resell cheap tablets with Ubuntu re-themed to look as much like iOS as possible, probably called iBuntu or something.

2) Make deals with manufacturers and produce mass market tablets, they will have to compete head on with android and I'm not sure what their USP would be here. They would either need 100% android compatibility or would need to magic allot of app developers out of thin air.

3) Go exclusively for niche markets, for example selling to car manufacturers developing bespoke in-car entertainment systems. This would be a whole new business strategy for them and would basically remove the need for the "ubuntu project" as it exists today.

If this were to actually work, they would need some sort of huge mass marketing campaign which convinced people they actually want or need Linux on their computer, which is very unlikely. Defining your audience as people who want to save the 5% cost of a Windows license cost from their already cheapass PC is never going to amount to much.

The more cheapass the PC, the more of a difference the Windows Tax makes. $50 out of $1000 is only 5%, but $50 out of a $115 Walmart computer[1] is pretty substantial (not that Walmart is likely to be paying that much for a Windows license, but at the low end a little bit of money makes a big difference).

[1] http://www.walmart.com/ip/IBM-40GB-8183/16880171

And MS does scale their license prices based on the system costs, but I think you've demonstrated one can always find a cheaper PC without switching to Linux.

There are also large organizations that would like to cut costs wherever they can, and can't be expected to make up the Windows tax by installing crapware like a vendor can. The city of Munich is the most well-known example of a big player switching to Linux. The effort was not a complete failure but was something of a disappointment - I don't think Munich ended up saving any money when you consider the labor costs, etc. But if Linux had been ready and the transition had been more successful, you'd see a lot of imitators following Munich's lead.

But it's probably a moot point anyway. I think Android will end up swallowing the PC market like a classic disruptive technology and stealing Linux's thunder anyway.

This is the greatest thing about Linux, and its curse. In Windows you have the Windows API, in OSx you have Cocoa. Sure you can add more to the mix but most of it is the main API.

I enjoyed the direction gnome2 was going, slowly filling in the niches needed to create a complete system. Then it seems they scratched that and went with gnome3. I like a lot of things about gnome3, but I wish gnome2 would not be defunct.

The same goes for kde, slowly building a complete system of tools for most users.

Both gnome and kde are the reason people would choose gtk or qt for their development.

What Unity needs (and I am not sure if it is in the works) is a true Unity API. Build apps for Unity with Unity specifics. Of course this goes against the Linux way, but if one really wants "the year of Linux on the desktop" we have to stop being a moving target.

Accommodating this requires either multiple WMs or ones which are very flexible. The problem with Unity is that it dilutes the work available for other window managers reducing the number of quality choices but it does not replace this with enough flexibility in itself.

If I want pretty GFX and UI I'll just use my mac.

Yes, you can disable Unity... But the issue for me is a lot of man power will be spent on it. Manpower that could be put to use on other things.

For me Unity is far more powerful out of the box than GNOME 2 was. Being able to position and tile windows with the keyboard and having 10 applications easily and automatically bound to a keystroke are huge timesavers for me.

Unity isn't going to drag any joe-average users to suddenly want Linux on their PC. It's neither consistent with what their used to nor does it create any kind of superior paradigm to what already exists.

Sure you can browse the web , play music and do the odd spreadsheet etc with it , big whoop we've been doing this for years , it would have been exciting in the late 90s not now.

The barrier for Ubuntu is not really it's UI. It's problems like having a really horrible flash player, poor selection of applications for many tasks (not to mention games). Not to mention other nasties like pulseaudio,networkmanager,dependency problems etc.

Ubuntu is so desperately trying to be a cut price apple, now they want to move to tablets.. Ok so they will have the same problem they have had on the desktop for the last 20 years, lack of consumer applications.

'What comes with the PC at the store' is what joe-average is always going to use. Always. What a Linux desktop looks like and aspires to be and how well/whether it works means nothing in this context. Nothing.

Numbers-wise, people using Linux on the desktop are: geeks (an insignificant percentage of PC users) and friends/parents of geeks (an insignificant percentage the former).

We're like flies buzzing around a very small pile of elephant dung (but it's really good dung); meanwhile the elephant has lumbered off, out of site.

Linux tends to be well used in dedicated electronics (routers, GPS , phones etc etc). We really need to think about what "Linux on the desktop" actually means. What it really seems to mean is a system where the majority of the components are released under a GNU (or similar) license , so that includes the Desktop environment etc.

Let's say for example that apple decided to rip the darwin kernel out of OS X and replace it with Linux 2.6.x but kept everything else as it was would we suddenly cry "woo it's the year of the Linux desktop" , what real quantifiable difference would it make?

It seems that in order for Linux to succeed as a mass market desktop OS it needs some company to plug big gaps (as has happened with android) at what point would the "Linux people" decide they no longer wanted to be associated with it?

despite how illogical that sounds, that may very well be cannonical's game plan.

Of course 2012 won't be the year of the Linux desktop and he nails it when he says we really don't want it. Give me power and stability. Give me power-user-friendliness, not grandma-friendliness.

YY.MM of release date.

Power users like myself will always find a way to get their favourite system up and running. I agree with Canonical and their policy of "fuck you, we're doing that no matter what you tell us", since their current user base is almost 100% power users and you can't really grow out of that niche if you don't take bold decisions. We will probably leave ubuntu because of it (I'm already on Arch Linux on most of my machines) but if they manage to attract a good slice of the market it will be a great thing for Linux as a whole.

They probably shouldn't try to sell themselves as a "cheap mac", but that's another story.

So normal people have Ubuntu and Unity and the rest of us have a million other options. Hell, I even used Ununtu for development for a while. It was nice for a month but then the novelty wore off and Unity got in my way. But I hope more people realize that Ubuntu is not for you, it's for everyone (if everyone means you aren't a big power user). What's wrong with that?

Or are you referring to touchscreen usage? How many people are actually using multitouch with Ubuntu?

You can launch or switch to any of the first 10 applications on the launcher, resize/maximize/minimize a window, tile multiple windows, switch workspaces, or bring up the dash with a single keybinding for each. The only one of those I remember being able to do in GNOME 2 was switch workspaces.

(http://k12linux.mesd.k12.or.us/docs/gnome/ch01s02.html)

    Alt-spacebar    Opens the Window Menu.