"The 'usability studies' that get quoted whenever I have talked to people from Canonical about this have participation numbers that are laughably low. Having a focus group with two or three dozen hand-picked candidates to 'evaluate' complex user experience changes like these does not seem convincing to me. It seems ludicrous."
Dismissing out of hand as "ludicrous" that user research on small groups can and very often does produce useful results is enough to discredit the rest of this guy's rant. He wants to be a part of the design process but doesn't understand even the most basic practices of user research.
I am not a programmer and I do not demand representation in coding decisions and I do not criticize code choices I can't possibly understand. I wish that people who are not UX and UR practitioners would afford designers some of that same respect.
I actually think he's right in dismissing those studies as ludicrous. You can't honestly take one or two (that I can recall) "studies" with only a dozen participants and then brand your product as "usability tested." That's just paying lip service. When you're changing a complex, deeply-ingrained interface for hundreds of thousands of people, doing two studies on a dozen people--and AFTER the fact!--and then calling it a day does indeed seem ludicrous.
Though at least Canonical can say they've done SOMETHING, which (at least I think) is more than Gnome has done. Gnome Shell gives itself a handful of breathless superlatives but I don't think they've ever once actually done a study of how real people use software. (Correct me if I'm wrong.)
In the end, though, such studies are very expensive and time-consuming, so free projects would have a tough time pulling it off. Though that doesn't discount the silliness of Canonical's attempts.
> Dismissing out of hand as "ludicrous" that user research on small groups can and very often does produce useful results is enough to discredit the rest of this guy's rant.
In other news, dismissing an entire article based on what is only a minor detail in the overall argument is enough to discredit the rest of your rant, right?
On some websites, people think they look smart if they point out a single flaw in a long article and pronounce that they "stopped reading right there". Happens all the time in Reddit. But I'd expect better from a top comment in an HN thread. Yes, the article has flaws. Lots of flaws, actually. But discredit the whole thing because of one point that isn't even the main point?
The difference is that users can understand when the interface is poorly designed because they have to interact with it constantly. Sure, there are parts of UX that most users never think about, but when something has been overlooked, like when you can't move the launcher sidebar to the bottom, or the settings are confusing to get to, there's a usability problem.
I think the real thing about Ubuntu, however, is that it's Linux. That means if you don't like Unity, you can install GDM, KDE, Fluxbox, Awesome or any number of other window managers. You aren't stuck with the (subjectively) poor design decisions Canonical made, and you certainly have more choice than you do in OS X or Windows.
It makes very little sense to pick a distro if you dislike their default interface. There are too many choices, or you can pick a distro like Arch that is WM agnostic.
It does if you value other aspects of it: Debian package management, Ubuntu PPA's, Wayland (when that arrives), popular apps like Spotify target Ubuntu and Fedora and don't 'just work' on other distro's, etc.
Default interface is just one aspect of the decision, and some of the others may outweigh it even for those who hate the interface direction.
The difference is that users can understand when the interface is poorly designed because they have to interact with it constantly.
Can users understand the difference between poorly designed and different than what they're used to? Given that studies have shown that people think keyboarding is faster than mousing even when it's not, I doubt it.
more accurate title: "ubuntu: i don't like what you're doing anymore."
ubuntu isn't doing anything wrong. they're not doing things the old way, nor are they pretending to. it's top down control with unilateral decisions, but that's what's necessary to move forward. for a product to come out well, decisions have to be made. if you listen to everybody, you get a product that sacrifices quality in favour of not pissing anybody off, like linux mint with it's three distinct pre-installed media players.
I think Canonical's new direction is a breath of fresh air in the world of Linux. We already have so many Linux distros, and what has that gotten Linux? 1 % of the consumer market? Can we just have one Linux distro that is consumer-oriented, and tries to shake things up a bit?
Yes, it's called "OS X". Okay it's BSD, but that is exactly what Apple came along and did ten years ago. The desktop Linux movement has a much bigger mountain to climb (insurmountable imo) than it did when we were having this exact same discussion in the late 90s.
He already rebutted this in the article with the example of GNOME 3. But I agree with nextparadigms; if "Linux is about choice", why not have one distro that does things differently? The only real problem with Ubuntu IMO was the fact that they built a community of volunteers and then disregarded their input, bait-and-switch style.
i'm not sure how contrasting ubuntu with gnome3 is really a rebuttal, when gnome3 gets almost all the same criticisms that ubuntu does. the bloggers start frothing at the mouth every time gnome3 introduces a change the same way that they do when ubuntu introduces a change. gnome3 is progressing as rapidly as they are precisely because project leaders are exercising exactly the same sort of unilateral decision making process that canonical is.
I agree completely. Design by community is what every distro does. As a result, Linux got nowhere on the desktop. Maybe the problem lies in how Ubuntu communicates. It is very much a "we'll do it how we want to do it". And as a result, it's been very successful. If Ubuntu was designed by the community, it would be just another distro. The fact that it has top-down directed design means it can differentiate. Ubuntu's measure of success can be directly tied to the amount of outrage from the Linux community, people who don't seem to understand that Ubuntu is Ubuntu and Debian is Linux and they can both successfully coexist.
Linux on the desktop was always a non-starter until Ubuntu gave it a kick in the ass. Now there's even a successful Ubuntu-only PC maker, System76 making (not quite) Apple-style laptops. Because of Shuttleworth, we have a nearly-ready viable third option in the market.
Do I agree with all their choices? No. Am I using Ubuntu on a daily basis? No. Am I looking forward to the day when they announce Ubuntu is "ready"? Hell yes. If you have a point, make it. If you just want to complain, get a cat.
But the person making the decisions not a designer. When you load Unity for the first time you have to be taken aback, or at least I was, but how glaringly ugly it is.
Is it, though? Canonical states quite clearly that they employ a TEAM of designers to handle the UX aspects of Ubuntu. Unless you're suggesting that they're micro-managed, for which you'd need to supply some proof.
Longer-than-page multi-column layout without even a anchor link back to the top under the first column AND almost invisible link styling - why does this designer hate his readers so much?
You can make multi-columns look good if you make sure to place all the content above the fold. I made an experimental Readability hack which did this and expanded the page horiztonally with a bunch of javascript intercepting navigation - hooking into stuff like mouse scrolls across platforms is a nightmare.
It might have worked better if the columns were more segmented so that you didn’t have to scroll down and up and then down again. That way you could get more content for less scrolling and maintain comfortable line length.
They may work, but you have to keep all of your content above the fold and paginate the article. Comments can be implemented as sidenotes and can point to specific parts of the article.
What I meant is that nearly all internet articles are optimised so that I only need to scroll down. Scrolling up only is necessary if I need to reference something that I already read.
Both your case and a single-column layout fulfill that need. The linked article does not.
> While Canonical is happy to accept thousands of hours of free work in bug fixes and quality assurance of their code, they are not interested in letting the community have a meaningful say in its conception.
Which solves the problem of design by committee and results in a great desktop experience. I love how Unity is preserving space on my 13.3" screen and how Ubuntu team made all the choices for me and then made it all work together -- I don't have to customize a thing (I use all the default applications and don't even change the background).
I remember the times when it took a day or more to install an operating system, all the necessary applications, and move your data and settings. With Ubuntu nowadays it takes less than an hour (including some hacks around the still rough edges, like reducing power usage) to have a completely up and running Linux installation. Hopefully, Mark Shuttleworth will stay on that course, because it's working great.
Disclaimer: I'm not a core member of any team, I filed two bugs overs years of using Ubuntu. Everything above and below is written from my very subjective perspective and it's entirely possible that I'm just not a very sophisticated user -- give me a browser, a terminal, hide all that other stuff and I'll be happy. Though Unity did convince me to start using Gwibber, Empathy and some other apps from their integrated tray.
> Gnome is a community-driven project through-and-though, yet it has recently come under fire for the way it has developed the latest incarnation of their desktop.
Yes, because their community-driven approach to design led them down the path of reinventing the wheel:
> Instead of picking existing components and giving them the final polish like Ubuntu did before, the GNOME project started developing things from scratch without any apparent reason to do so. And even worse: incompatible to existing solutions. It started with the rejection of the appindicator specification implemented by Ubuntu and KDE. At that point it was not clear to me whether the specification was broken or whether the responsible people at GNOME were just ignorant.
> Then came systemd. And it started to be apparent that unfortunately it was the latter.
I'm glad Canonical is trying to bridge the gap between Linux and an Apple-like design aesthetic. The Linux desktop experience has been extremely unfriendly to less than expert level users, and it'd be great if that changed. But if history is any indication, that means adopting a more top-down decision making process. Making tough choices, and streamlining design isn't something that comes out of forum discussions.
Of course, not liking the direction of Ubuntu is your prerogative. However, the great thing about Open Source is that you and like-minded buddies can move to a different distribution or remix Ubuntu to your preferences. In that case, everyone gets what they want.
I don't mind the some of the directions they have taken with Unity (I've switched to Gnome 3 at this time, I feel more comfortable it, even with no shutdown/restart option in menu ;p) but the shear lack of ability to customize is really annoying. I understand them choosing defaults but why take away all ability to customize?
Also, some changes still make no sense to me, the buttons on the left still annoy me. The dock sometimes interferes with those buttons when the windows aren't maximized.
I guess I'm no longer their target audience. I feel sad about that as their really isn't many alternative I find appealing. I could use Debian directly but I still find that Ubuntu was doing a great job at cleaning up tedious issues with Linux. Their new installer that asks questions while writing files to disk is such an obvious thing it should have been done years ago. There is plenty of other areas that they've truly improved in the Linux experience. The last UI stuff is just not there yet.
If you don't like the direction that the current default Ubuntu desktop is going, I suppose you could always try one of the variants, such as [Lubuntu](http://lubuntu.net/). That's where I'll probably head, since I'm running an older Ubuntu release and I'm pretty sure I'd rather not switch to Unity upon upgrading.
I can't blame Ubuntu for wanting to make their default desktop experience the way they want. It's their project.
My problem isn't so much with the innovation. My problem is with the lack of choice both Unity and GNOME 3 give to their users.
What Ubuntu and GNOME 3 are doing is not giving their users an option. I run Ubuntu on a work station, which is not a tiny screen gadget. I need to do work on this thing and I'd like to get work done.
Both Unity and GNOME 3 are departing from life time engrained metaphors and replacing them with supposedly better ways (whereas beauty is in the eye of the beholder I guess). No start menu, or menus at all. No icons on your desktop (wtf do I have a desktop for then?), Massive icons sticking out everywhere taking over valuable screen real-estate and be in your face at all time, etc.
If those new ways of doing things would be optional, at least for now, and either be able to be turned on or off rather than forced down your throat than in my opinion the resistance would have a lot less. Also, if those new ways of doing things would really be better it's something the end users can decide, not 20 people in a focus group.
On top of all this, timing couldn't be worse. Both the introduction of Unity and the discontinuation of GNOME 2...
I currently run Ubuntu 11.04 simply because it's the latest Ubuntu version GNOME 2. I have no intentions to upgrade and I'm looking for an alternative. OS X?
OS X is no alternative--you'll have just as little choice or less there.
You should consider KDE--it's a really nice option these days, and easy to configure however you like. Fedora is a great distro if you want to run KDE with minimal setup.
I also think KDE has come a long way. You can configure it all you want and if you ignore the plasma-stuff widgets, then it doesn't get in the way and is very space-efficient (just a panel really?).
I keep hearing good things about Fedora's support for KDE but have not tried it personally.
I run KDE on openSUSE and am very happy with it, as well as the community there.
Are we using the same Gnome 3? It takes up far less screen real-estate than than Gnome 2 did, only a small black bar on the top. There is a "start" menu, in the top left (called Activities).
My issues with Unity, as of Unity 5 (in Ubuntu 12.04):
1) Alt-tab should switch between windows within a workspace. This is HUGE. What is the point of workspaces if your navigation buttons/application launchers ignore the boundries?
2) Clicking on an application launcher should open a new instance of the program, not retrieve a window already open.
3) IF item 2 is to remain true, then the possible windows to re-focus should ONLY be the windows in the current workspace. See item 1.
4) Unity's taskbar should be movable to the bottom of the screen.
---
Credit where it's due: Unity 5 allows resizing of the Unity bar, and you can now pin it to the desktop, so it's not constantly popping over the left side of the window when you try to close windows, access menus, etc.
One must still install gnome-tweak-tools to edit the font configuration and other things, which is inexcuseable.
Everyone has hit on the same issues I found with the article, but these parts bugged me.
First:
>Apple’s software is hard to customize and can be extremely clunky and annoying to use — that is, if you use your own brain instead of blindly buying the hype.
Am I reading too much into it, or is that very insulting to Mac users? I'm not sure where I'm buying into the hype by preferring the look and feel of OS X 10.7 to Windows 7. The author seems to subscribe to the "if you like something I don't like, you must be inferior" school of thought.
Second:
>A very good example of this is iTunes and the way the iPhone forces you, even today, to plug in and sync everything through one horridly clumsy application.
iTunes is in no way a great application, but the author has made it apparent he comments on things with which he has no experience: iOS devices haven't required users to sync through iTunes since iOS 5. You do it all on the device a la Android.
With those issues and the various points other commenters raise, I don't see how this article is anything more than a rant by a person who sees simplicity and usability as an encroaching evil.
As we get closer to the next big version (wherein most of the apocalyptic changes take place) I'll be reading HN for useful lists of alternatives. Anyone with useful suggestions can chip in now of course :)
Linux Mint 11 and 12 are great alternatives. The Mint team has also made Cinnamon[1] which is also a great alternative to the default GNOME3 interface. If you don't want to reinstall Linux, Cinnamon is available for lots of the major distros, or you can compile it yourself.
You can always stay in the current version until you feel safe. Plus, if you don't like Unity, you can use Gnome, KDE, XFCE or whatever makes you comfortable.
I really don't see the fuss about it. If you don't like the new version, skip it. It's not like the next Ubuntu release will be more than 6 months after the current one.
I've been trying out Arch Linux (http://www.archlinux.org) for a few weeks now and am really happy with the results. I installed using ArchBang (http://archbang.org) which I think for a first time install might be the best way to go. (I think I'm almost ready to try a from scratch Arch install)
Now that it's installed and I'm used to the package manager (pacman - which I like better than apt-get / aptitude if that's possible) I find that it's been refreshingly easy on a day to day basis. And I feel like I have a great deal more confidence that I know what's actually going on as far as installed programs and configured behaviour.
Arch is great, but it targets pretty much the opposite type of user than Ubuntu. Could you imagine telling your grandma to switch to Arch from Windows?
Of course, nowadays, I would suggest Mint, because Ubuntu is just weird. Their original motto, "Linux for Human Beings" should be changed to "Linux for Mark Shuttleworth". He's off in his own world now.
Debian is fantastic. I tend to stay on testing (wheezy atm) because it is fairly modern yet stable.
Stability is the biggest problem with Ubuntu I believe- people will generally adapt to the UI changes that canonical is making, but breaking stability is a complete no-no. Even today, on Ubuntu 11.10, Unity shows wierd bugs time and time again. Apps like VirtualBox and KeePassx will show multiple times (or just disappear) in the dock. And sometimes the whole unity interface will close and restart itself.
Compare this to the old gnome2, or xfce, openbox etc; You can go for months without the UI doing anything unexpected.
The main difference is Debian isn't quite as user friendly out of the box. Sudo isn't set up for you for example so you'll be in for a surprise the first time you try to use it. Also, Debian isn't quite as pretty. The fonts aren't as good and the desktop looks quite a bit dated. The upside is Debian at least for me is much faster. I don't know what Canonical did to Ubuntu 11.10 but it drug for me. Also, Debian doesn't crash at all at least for me whereas Ubuntu did suffer some random instability.
That having been said, I like the idea of Ubuntu. I like the fact that they are trying to move the Linux desktop forward and I will happily return if they can get Unity straightened out. As it stands though, I'm a content Debian user for the forseeable future.
Interesting, I've been using 11.10 on a netbook since last fall, and on a new laptop since January, incluing daily use of VirtualBox and KeePass(2, not x), two of my top 10 most frequently used apps. Zero stability issues, I've been very happy with it.
I wonder if there's some hardware issues, or something else you've got installed that's causing those problems.
Hmm. This was a fresh vanilla install that I tried recently on the same hardware that runs debian wheezy. I wanted to have all the benefits of ubuntu's large community (PPAs for a lot of sofware etc, plus I do generally agree with the direction that ubuntu is heading in), but the lack of stability was just killer.
I think the gloom around Ubunutu is a overly pessimistic, but if you're really intent on switching you should consider KDE. The current version is really elegant and usable--I really like it. Fedora is a good distro for KDE.
I'm happy to see a big player in the open-source world take a design-focused approach to software development. The author may not like Apple's design, but their top-down approach has led to products that are loved by many. It would be great to see this approach succeed in open-source, so that we end up with well-designed products that preserve the freedoms of open-source.
Canonical wants to create somethingm the OS Mark dreamed of, just like the GNOME and KDE people, what's wrong with that?
“If I had asked people what they wanted, they would have said faster horses.”
― Henry Ford
Sometimes you need to decide whats best for the user. Most users do not want choices, most users want something consistent that works and that is what Canonical is trying to deliver.
I've always been confused by the "choices" argument: it seems to be a false dichotomy. Why not have good defaults but then let determined users change whatever they want? If you don't care, you just use the system as is; if you are consistently annoyed by something, you can dive in and fix it.
That is what Canonical is pushing for, you get Unity and GTK, if you don't like it, you are probably one apt-get install away from changing it. To accomodate Ubuntu to my needs I hack the hell out of it, it takes me hours to get it right, to make OS X work the way I like it's just an hour or less, but I have more power over Ubuntu than OS X.
The problem however with choice is maintainability and duplication.
Do we really need so many window managers? Why not have only one that actually works?
While it's cool to have multiply desktop environments for almost every distro, it has it's toll, somebody has to maintain and test those.
What would we have if we had combined the effort that was put into Libre Office, Abiword, KOffice, Open Office and the rest?
Don't get me wrong I love having control over every bit of the OS and I love being able to choose, but I feel we could sacrifice some of it it for the greater good.
> that hides the base system as completely as possible
That is the "doing it wrong" part. But, I can see/admit my "doing it wrong" is someone's grandma's "doing it right". Although, I still think (wished) grandma had to learn how a car/computer/foo worked "grossly" before she was allowed to operate it.
This person has some other version of open source in mind in which it means something other than 'I can download the source code with the program.'
I don't take community suggestions for features on any of the projects that I have on github;
A.) because no one uses any of my stuff.
B.) because I don't care if they use it, I'm doing it for myself.
If the makers of Ubuntu have a vision for their idea of a perfect desktop experience, they should go ahead and do it. Sometimes it does take a dictatorial visionary to get important stuff done. We already have gnome 3, yes, and it is fine, but there is plenty of room for others. Particularly successful others like Ubuntu/Canonical.
Yes, Apple is successful. But so was Microsoft, and you wouldn't want to emulate Windows either!
The point is that while Apple products are selling, the author completely reasonably doesn't like their software and does not want Ubuntu to have the same issues.
Dismissing out of hand as "ludicrous" that user research on small groups can and very often does produce useful results is enough to discredit the rest of this guy's rant. He wants to be a part of the design process but doesn't understand even the most basic practices of user research.
I am not a programmer and I do not demand representation in coding decisions and I do not criticize code choices I can't possibly understand. I wish that people who are not UX and UR practitioners would afford designers some of that same respect.