Slashdot Mirror
GNOME Usability Study Report
pdiaz writes: "Here is a report made by Sun Microsystems people about GNOME usability. They collected a bunch of professionals (lawyers, engineers, Graphic Designers, etc..) and put them in front of a Gnome desktop. They were asked to perform some tasks and tell what some icons, menus, etc., do. Some quotes are really funny, like when they asked what does the terminal emulator icon launch." Very interesting stuff, and this approach is necessary because once you've gotten accustomed to the system it's no longer possible for you to evaluate how it appears to an inexperienced user.
Instead of being irate about people who can't understand everything as fast as you (or did you understand it all that quickly?), think about what they said. One complaint often made about Microsoft Windows was how confusing it was to find the settings you wanted, and how many places you could make the same change. Microsoft has listened to people and improved their interface. Windows 2000 is much easier to configure than Windows 95, because related settings are grouped together. I am not a Windows fan, but I think we can learn a lesson here. We do need a unified architecture for overall settings on both KDE and Gnome. Icons can be confusing, and tailoring them to make it easier for people to understand is not a BAD thing to do. If we want non Unix/Linux types to use our system, we have to at least try talking to them in terms they understand, instead of calling them stupid.
To drag out an overused quote: "The only intuitive interface is the nipple..."
Seriously, there are no natural interfaces, merely familiar ones. Also, the easiest interface to learn isn't usually the most productive interface for an experienced user. When those goals conflict, designers should choose the productive interface over the easier one, since users will spend a lot less time learning the interface than they will spend using it.
> I think you'd have to agree that a button labeled "start" or "run" makes a lot more sense than a foot
I suppose that makes sense pressing that "START" button to shut the machine down? Brilliant.
Some quotes are really funny, like when they asked what does the terminal emulator icon launch."
It is exactly that attitude that makes Gnome unusable to the average PC user. Until that changes, Linux won't make inroads. Taco has already addressed this though. MS spends lots of money doing this kind of thing. The best we have done so far is a cheap rip off of that. We constantly rip on MS for bad software, yet when push comes to shove, we're just trying to give away free versions that look like theirs. It isn't working. This is an important first step in usability testing for Gnome. KDE would be wise to do the same, though they are probably about a decade (maybe a little less) ahead of Gnome in usability. Gnome could/can catch up by doing things like this. KDE can't remain idle, and MS won't remain idle.
i do use a scheme that has this changed, but it wasn't the default. flexibility is nice, but it's even better when things just work.
--
it only solves the first issue, and that still depends on what theme you use. you can still logout with one click (or be presented with the 'do you wish to logout' dialog if you have confirmation on).
--
Not all /. users hate macs. I have all my audio equipment hooked to a mac, and my laptop is a mac.
--
way to provide a link, slick. even knowing it's name, i couldn't find it in an acceptable amount of time (less than 2 minutes), seeing as you gave no useful information.
--
The problem is that almost nobody takes the time to think about what they're doing to decide if it's a good idea or not. Try navigating on a windows machine without a mouse; after a few minutes it's pretty easy. Try that with GNOME. Switch between apps and see if the keystrokes are consistent. see if the focus goes where you think it will in a complex form. most of the time, GNOME keyboard shortcuts are implemented as a complete afterthought, and it shows. If there's a GNOME standard for this, it's followed poorly.
Additionally, most of the original desktop themes are just plain useless. they're:
- pretty but useless -- how do you maximize one of those windows?
- geeky and useless -- okay, tell me where i click to minimize, maximize, close, or stick the windows.
- just plain dumb -- who the fuck knows how this works? i sure as hell don't, and I have better things to do than figuring it out.
And the scary thing is all these themes have 10k+ downloads, and it took me about a minute to find these examples. I'm sure if I really cared I found find much better examples of what I'm talking about.I just spent 15 minutes looking for a truly good theme somewhere without success. that's a tragedy. that will hurt linux's mainstream acceptance far more than the fact that cmdrtaco was too dumb to buy a supported scanner.
--
For example, why does everybody copy the design that the 'window kill' button should be right next to 'maximize'? That's horrible design, put window kill on the left, maximize and minize on the right.
Why is it possible to click down on the 'K', move the mouse a few pixels up, release the mouse, and log yourself out. If you have a fast computer, and you use KDE, you've probably done this before.
GNOME allows an application to use the entire task tray, then when you have two applications, it uses half that size.... and it squeezes down. It's efficient use of space, but it's inconsistent and makes it harder to tell with a single glance what's running. KDE makes good use of the space without this annoying inconsistancy.
What the hell are these icons? Stop being cute, start being useful. If you're running KDE, hit the K menu now and tell me what the following icons mean 'quick browser', 'bookmarks', 'toys', 'system', 'preferences' (these last two are way too similar), multimedia or graphics. None of those icons gives you any intuitive notion of what you're about to launch.
Additionally, I doubt I'm the only one who has taken the less-used apps in the menu for each level, made a folder called 'sewer' and stuck them in there. Yes, we're all proud that there are lots of applications now. No, we don't use 90% of them, and having them in our menus just slows us down.
Things are improving, but it's still terribly ironic, the way copying Microsoft is referred to as innovation, yet when Microsoft copies, that's just plain wrong.
--
To use your example of a friends car: I once drove a friends car, and while there were places where that was an issue, one statnds out that wasn't: The turn signal wasn't a lever on the stearing column, it was a switch on the dash. Yet I used that switch 10 times before I realised that it wasn't a lever! Whoever designed that interface made a major change, yet it was completely transparent to the user.
Interfaces should be natural. It should be no problem to go from unix to windows to mac to cpm to OS/390 to... It is, and that is a problem
A large portion of computer users are quite serious -- they spend a large portion of their time using computers. It is reasonable for them to invest time into improving their performance and expanding their abilities. They don't need to learn how to move files around or whatever -- they've figured out all that -- but they need to learn how to be the true masters of their computer.
No one pays much attention to that step however. You are on your own -- some make the leap, some stumble. Not much effort is put into making this easier, though.
Yes, like a shortcut, MS's filesystem shortcuts get you someplace quicker, but half the time don't work. That's fine if they think shortcut is a better word there. But using it for hypertext links is stupid -- there is nothing except the link, there is no long path that a link is a shortcut to. It makes no sense, any metaphor that it implies is incorrect.
The whole point of usability is not to do anti-intuitive, anti-conventional things like that (no matter how small a detail the word might be). So I would hope that no one would adopt that term in an effort to match MS.
MS uses the term "shortcut" where everyone else in the world uses "link". Not just those lame "shortcuts" in the filesystem, but IE uses that term for HTML links. This is from usability testing? Yeah, right.
In FrontPage instead of having templates, like most HTML editors, FrontPage uses shared borders and themes, while "template" is used for something different. And it uses the term "web" where everyone else uses "site" -- the way they use "web" is simply stupid.
I can't recall any others at the moment, but everytime I use some MS product I notice these minor, strange namings. They often forgo convention to use their own odd words. They want to invent a lingo so everyone is confused when they try something new, just like these people were.
Trying to immitate that would be like using the Word .doc format for file saves -- it's hard enough just to import the crap, you can't expect to become the crap.
I'm not sure "shell" would mean any more than "terminal emulator"....
It should be easier to tell people to click "shell" or "run shell" than to say "run the terminal emulator program".
Having tried both I greatly prefer hitting a keystroke to add characters.
However "My Documents" is a good indication that MicroSoft is not studying things much either. In fact "My Documents" is ON one of the disks. Removal of the C: drive will cause "My Documents" to become empty, which would be somewhat confusing to the user who can clearly see from that display that they are NOT on the C: drive!
It should initialize with the *actual* "My Documents" directory already opened and highlighted.
Unlike either Windows or Linux it would make sense for this directory to be immediately under the physical disk seperations.
Another idea would be to hide structure that cannot be changed without turning the machine off anyway. So the top level should be something like this:
/cdrom
/floppy
/My Documents
/Joe's Documents
/Sally's Documents
/The web
Of course both Windows and Unix make it very slow to get at the data, while fast to get at the less-important "filename". This is all backwards.
Please don't mess up the file system even more with "attributes", thank you.
Bullshit. This is exactly the type of "windows blinders" that people here are complaining about.
The average user does not want to "install" a program when they double click on it. They want to use the program! In fact the average user would be overjoyed if they could throw the program in the trash can if they don't like it and it is gone.
It is unbelievable that people complaining about the mysteries of Linux can blindly spout crap like "install" and think they are describing real non-computer-expert's thoughts.
Of course I should point out that Linux is as bad or worse than Windows with this "install" shit.
GTK is LGPL you idiot. You can make closed source apps and charge all the money you want for them.
I have never seen a person who learns point-to-type (you can get this on Windows and NT by messing with the resource manager) switch back, and they quickly become frustrated when encouterint click-to-type. Where I work more than half the NT machines have been switched to point-to-type.
The fact is that point to type is, without question, superior. It is as close as possible to the ideal way to direct keyboard input to several objects using current hardware (the ideal would be to somehow read your mind or track your eye movements to see what thing you are thinking about).
The fact that this is not the default on all Unix systems (or on new versions of Windows, for that matter) is a good indication of how harmful the engraned user expectations are to advancing the design of machines.
Otherwise there is no reason for the beginning user to see a shell, I agree.
I still feel that "shell" is a much better word for people to see, and it means as much to a typical user as "terminal emulator" or other such verbage.
So I guess my complaint is that "My Documents" is shown by MSoft at the "root", rather than it's actual location. This defeats the whole advantage of forcing the physical disk structure to be part of the hierarchy: if a disk fails or changes it is clear what portion of the tree is affected.
I think a solution for MSoft would be to have the Explorer and file chooser start up with "My Documents" preselected, but in it's actual place in the hierarchy. It would look exactly as though you navigated down to it in the current one.
Of course as they have named the directories down there now, it makes the Unix directory naming look like a work of genius... Maybe they could fix it so it is not nested in so many levels. Perhaps C:/People/username (or /home/username ???)
"Someone might be overwhelmed by the amount of options." (P7)
"There are too many features and icons for sombody new." (P7)
"Some things were accessible that an entry-level user would not want or need, but other things were buried deep." (P11)
"If someone showed me it would be okay, but if I was a new employee without help, I wouldn't get it." (P6)
While that is what we all do, it is what the useability experts are paid to figure out, not the test subjects. We all think we know what a new user would do but we don't, we have to watch them.
While their opinions and insights are great, some nice hard metrics (time to accomplish a task, clicks / keystrokes required, number of false selections, etc) are also great and leave the UI designer with some goals to shoot for. (In typical slashdot fashion I've not finished reading the study in question so that may very well indeed be in the study).
Chris Cothrun
Curator of Chaos
Bleh!
I believe Slackware still ships it, but I could be wrong.
> If you don't know what post-modern rationalism is, do philosophers assume you are dumb?
Some of them will rip you to shreads for not having a basic understanding of philosophy, yes. People in every field don't understand how the general public can be so stupid as not to understand the basics of their subject.
> If you don't know Russian, do Russian professors mock you
Maybe not just Russian, but if you revealed you knew but one language, probably. You've never read people trashing Americans for being monolingugal before?
> Computer nerds are the last bastion of
> unadulterated bigotry
Yeah, whatever. Any discipline are going to have people who think themselves superior for knowing that discipline. English majors don't understand how people can be so poorly read; geography majors can't understand how people don't know where countries are. I complain about many of the customers at Homeland, as do most of my co-workers. Computer nerds are just human, doing what humans do.
Where's that blue-screen with that unreadable data I keep seeing when I use Photoshop in Windows all the time.... Hrm. GNOME must be broken.
Really, how can you argue with behavior-based experimental data that "this isn't how people behave"? Oh right - with unfounded 3l337 opinion.
I completely agree with this (see also my post in the "Why Linux will never be mainstream" comments). However, some of the recommended fixes made me sit up for a second.
As an example, the researchers recommended replacing "Halt" with "Suspend (Halt - stops the processor)"
IMHO, this is a bad idea. "Suspend" a) already has established meaning in the context of computers (go into power-save or sleep mode), and b) implies stopping something in the middle to return to it later (as in the phrase "suspended animation")
My personal recommendation would be something like "Shut down" or "Power off" (and if you can detect soft vs. hard power-switch, use an appropriate term in each case, e.g. "Turn off computer" vs. "Shut down the system").
The experimental data are pretty solid, but there's nothing gospel about the recommendations, except "Consult an experienced technical writer" -- preferably one who wrote docs for absolute novices and got direct feedback on it from them.
-- Old Man Kensey
OTOH, it makes perfect sense to have shutdown and reboot behind a 'foot' symbol.
To expand further on your perceptive realization that "...its not stupidity, its simply lack of experience.", sewing machines, electronic keyboards, and other items often have an approximately hand sized object on the end of a cable that's intended to be placed on the floor and operated with the foot. Someone who's unfamiliar with the concept of a mouse and is confronted with one for the first time has only their previous experiences to call upon in trying to figure out how to deal with this new unknown. Perhaps the lack of intelligence was on the part of mouse designers for not making it more obvious what the mouse isn't.
I see even classic Slashdot is now pretty much unusable on dial up anymore.
Well then, what operating system *will* make me special? :-)
I see even classic Slashdot is now pretty much unusable on dial up anymore.
Interesting attitude...
If you don't know what post-modern rationalism is, do philosophers assume you are dumb? Or do they attempt to explain it?
If you don't know Russian, do Russian professors mock you, or do they try to convince you to take a Russian class?
Computer nerds are the last bastion of unadulterated bigotry, doing whatever they can to encourage a new digital apartheid and engaging in the worst forms of de-humanization rhetoric.
Potato chips are a by-yourself food.
You are arguing from the specific (you) to the general (everybody else). What you find stupid may not, in fact, be stupid at all.
However, my point still stands: a philosopher or Russian prof would not call those who don't understand their field of expertise stupid. You did.
While people drive cars everyday, they don't interact with the car beyond a simple interface -- an interface that they took classes to learn and years to perfect. Some secretary who can type 90 words a minute, who's had a computer with a mouse and icons and such foisted on her only sees the thing as an impediment to her abilities. To her, it is simply a fancier typewriter, albeit one that randomly erases work she's done when the app crashes, and has a jillion options she barely understands (and will never use) that are constantly hyped by a ridiculous talking paperclip.
Just once, I'd like to see a comp-sci nerd, instead of "Users are dumb", say "We don't program for our users, therefore we are poor programmers".
Potato chips are a by-yourself food.
Don't let the simplicity fool you, though: It may be very hard or even impossible to access this interface
:)
Well sure, it is for me too, but that's because my license was for a different installed base and expired at least 25 years ago
--
I've finally had it: until slashdot gets article moderation, I am not coming back.
"Where's Clippy?" (P183)
Trolling is a art,
All interfaces take, well, getting used to in the beginning, this isn't exclusive to computer interfaces. I know that every time I get into one of my friend's vehicles I have to ask, "Hey, how do I turn on the lights? Where is x y or z?".
That being said interfaces which are reasonable can be adjusted to within a reasonable amount of time. Gnome is certainly something that those of average intelligence with the right amount of time should be able to get down... If they are interested in it and there is some reason to adjust to it.
So find a reason for people to use Gnome, and they will. (I am not saying Gnome doesn't have a use...)
prosebeforehos.com
"I want to access my email, but when I open up the Access icon my messages aren't there!"
"Excel? I'd think a serious software package wouldn't be named after chewing gum."
No, not real quotes, but they could be. *Everything* has context. These Microsoftisms are only in place because that's what these people have been taught (yes, taught) to recognize. Just because something is different doesn't make it "wrong" or "less intuitive" - you have to look past the first impressions and see how the GUI works once the user gets some basic familiarity with it.
At my place of work, courses on Word and Outlook are de rigeur. Do we really want to base a user-friendly GUI on a system that requires training to use? That's what most of the comments provided seem to indicate ("where's the start button?", "why settings and not control panel?", etc).
Then why is MS dumping the Win95 interface in XP?
...this coming from the people who brought you the Sun WorkShop? :)
Sure people hate what they are not familiar with. They don't like change, but some of the end comments by the users does not neccesarily support the argument that a windows like interface is needed. I like P10's comment the most:
"It is like a cross between Windows and a Mac. It looks like it is designed for everybody."
People familiar with Windows are able to just use KDE (debateable, but no studies have been performed on it so I will give it to you), but that doesn't mean that they like it. It is what they are familiar with and what they know. So they can be lazy and not have to learn anything new.
Many users will probably be like that, but there are a lot of users out there that hate the Windows interface but don't even know it because they have never used anything else. I absolutely hated the Mac interface until I used it for several days in a row. Once I became familiar with it it was very nice, I started to dread going back to my windows desktop.
Basically what I am saying is I want BOTH to thrive, I want both to be put through these kinds of tests repeatedly. People should be able to use what they like, one does not have to win over the other.
Q.
The only 100% intuitive interface is the nipple...
--
*Condense fact from the vapor of nuance*
25: ten.knilrevlis@wkcuhc
*Condense fact from the vapor of nuance*
OpenLook has been effectively dead for a long time. Sun switched to CDE when they ditched OpenLook in favor of Motif, which had basically won the battle at that point.
As for open sourcing it, Sun did that before they gave up on it... it was kind of their last ditch attempt to outmaneuver Motif. Unfortunately it was too late. Had they done it about a year sooner it might have made a difference.
I used to use olvwm on Linux back in the 1993 to 1995 time period... I imagine the source code is still out there for it, but I don't think it ships standard with many distros these days, let alone is part of the normal installations.
I had a user have a fit because he felt he shouldn't have to click the "New" button in Outlook to compose a new email message. He felt it should be labelled "Send." Your quotes are probably closer to reality than "they could be."
--
--
"Outlook not so good." That magic 8-ball knows everything! I'll ask about Exchange Server next.
Tying the logo with the name might have helped on the terminal emulator problem too, although they really ought change that to "Command Line Prompt" or something similar.
-sk
If only keyboards had a sensible layout....
I find it incredibly ironic that we try to learn to type fast on the QWERTY layout, one which was specifically designed to SLOW TYPISTS DOWN. In the typewriter days, keys stuck together if one typed too fast, so a new layout that impeded the typist was devised.
We should have switched to a sensible layout (one like Fitaly for Palm comes to mind) long long ago. Now, I fear, it is too late to switch...
In designing systems and user interfaces, it is
fundamentally important to not confuse two distinct concepts: usability and transparency.
Usability is directly related to the efficiency of
performing tasks and the ability to anticipate the
user interface for new tasks.
Transparency is the "intuitiveness" of the interface or system. It is primarily a measure of
how easy it is for a naive user to come into the
system and get a something done.
Transparency is intimately related to the experience of the users being examined. In a certain respect, it is a measure of familiarity.
Unfortunately, you will get high transparency
scores nowadays if you simply look and act like
MS Windows.
Usability is a whole other bag of onions. Some of
the features of a transparent interface are relevant in assessing usability, but only to a point. While transparency is something critical for new or casual users, it can be almost completely irrelevant to an experienced user. Once a certain level of familiarity is acheived, usable systems are those that make the most common tasks the most efficient to access and provide easy means of aggregating and controlling common
sequences of tasks. Emacs is an immensely usable
system that has a very low transparency score.
It is interesting to note that the Usability Principles in this study seem to be correctly labelled: they *are* related to interface usability. However, the assessment methodology seems to be primarily measuring *transparency*.
I'd say that this is a basic flaw in the study and
colors the recommendations highly.
It would be nice to see someone do a similar study
but concentrate on the power users and address the
issues around high performance usability.
Unless you've sat down and observed your interface getting tested with a usability professional or two who work with regular folks to see how the application works *in the real folks non-geek world* then you don't know what you're talking about.
Really, how can you argue with behavior-based experimental data that "this isn't how people behave"? Oh right - with unfounded 3l337 opinion.
Sure, there are other things we could do to better test usability - like have them spend a week or two with Gnome after this test, then test again to see how much they picked up.
but until you're doing testing with your own projects, until you appreciate that these are real people in the real world (that same world you think should use Linux as a desktop OS) then you're really missing the point.
cz
see www.usability.gov
IBM Ease of Use
The perennial Jakob Nielsen
Usability Professionals Association
Webword Usability Blog
.. it isn't the _first_ step. Eazel also did some usability-testing with Nautilus (which is an important part of GNOME), and a lot of talk in the development of Nautilus is different from other projects, because the question "but would this be intuitive?" is asked all the time..
Otherwise I agree totally with what you're saying.
KDE is currently superior in some fields, but one field GNOME is superiour is usability-testing.
There are going to be a lot of posts like, "What lusers! They need to RTFM so that they know that 'terminal emulator' actually means command line prompt!"
This is not what you should take away from a user interface study. This *is* what the users see when first presented with the program. It really doesn't matter what the programmers/designers of GNOME think. If the user doesn't like it, then he doesn't like it! If he can't understand, then he can't understand.
A long time tenet of communication is that if there is miscommunication, then it is usually the fault of the communicator who hasn't adequately taken into account the audience. If we as programmers/designers aren't using the interface to *communicate* then it is *we* who are failing to communucate, not the audience who is failing to understand.
Why do you think that MS has slowly moved to simpler and simpler language? People don't need techo-speak to understand what is going on with the computer. Understanding phrases like "illegal operation" requires a bit of underlying knowledge about why such an analogy is being used. So why use it. Just say, "your computer just crashed, but it's okay. Just press that little button on the front of the computer so it can restart. Have a nice day!".
Mad Software: Rantings on Developing So
It can be hairy at times - but I get satisfaction that I know what does what, where, and when. It may take me a bit longer to set something up, figure out its dependencies, etc - but in the end, for me, it is worth it.
To be honest, this update from SuSE 6.3 to 7.2 has been less painful than I thought it would be. I can't say painless - but not the huge bear I was imagining. I was half expecting to end up moving my home area to another partition, then reformating the root, boot and swap partitions, and reinstalling (then moving the home area back), but so far, it has worked out great. Even the stuff I compiled under 6.3 still works (AFAIK - but I am going to recompile it in the end).
No - Linux definitely doesn't hold your hand for that kind of an update.
One thing I wonder though, and I hope to be able to try it someday soon - if I can gather the hardware together. I wonder how a "virgin" install of SuSE 7.2 would go - the update went smooth, with the installer and everything being very, very slick - better than what I remember from the Win95 or 98 installer. I just wonder how easy it would be with a fresh machine. It looks like it would be super simple, from what I can gather...
Worldcom - Generation Duh!
Reason is the Path to God - Anon
In Mac OS X, the Apple menu has at once been deprecated (no longer used as a system launcher/customization tool), and brought forward (important system commands stored there).
The one was 'cause customization was Byzantine (dig around in the System Folder for a Folder to use to customize it), the latter 'cause it's supported by the menu hierarchy they're putting forward
(System-level stuff -> App -> situational)
In UI tests, failing to realize the Apple menu was a menu was a common stumbling block for naive users on Macs.
William
--
Lettering Art in Modern Use
Sphinx of black quartz, judge my vow.
Build it, and they will come^Hplain.
That's pretty silly when you think about it. A C: drive, the syntax (C:) etc... It's as weird as anything under Linux. It's just that users have learned this one since the beginning of time, er, MS/DOS epoch, so now they expect the same kind of sillyness.
We need to corrupt our youth at an early age so when they are exposed to the Windows world, they'll be like "Drive letters? How fucking primitive!" :-)
"Who gives a shit?"
People who are worried that users taught bad habits will actually force them away from being able to write elegant, intuitive systems. Easiest is not always best, etc. (Hardest is not always best either). Seems like GNOME needs a "novice" mode, which like Windows 98 hides all the advanced stuff (like scary black windows my god!), but can be displayed with a click of a button (or by a permanent setting). This way both novices and advanced users can be happy. Maybe every GUI feature can have an experience rating, and the user can set what experience level the GUI should display itself up to.
It's 10 PM. Do you know if you're un-American?
Mozilla, XMMS, Gimp, Compupic, Gphoto, Balsa, Freecell, ...
-- Pure FTP server - Upgrade your FTP server to something simple and secure.
{{.sig}}
...and even babies, check my OpenBSD baby .
-- Pure FTP server - Upgrade your FTP server to something simple and secure.
{{.sig}}
People familiar with Windows or MacOS will be lost with Gnome, or any new user interface. Because they already know icons, locations, shortcuts, etc. Working on something different needs time.
My girlfriend never used computers before we met together. And on my computer, there's only OpenBSD, FreeBSD and Linux. I showed her how to log in, browse internet, paint, play music, print photos, etc. Her desktop has icons for main applications.
And she's not lost. She can use the computer without any help. With Linux, FreeBSD or OpenBSD.
The last week, she had to work on Word 2000 at her daily job. She was totally lost, found Windows slow and ugly, didn't understand why the word processor had so many complicated buttons and menus, etc.
So no interface is more intuitive than another. It's just a question of what you are more familiar with.
-- Pure FTP server - Upgrade your FTP server to something simple and secure.
{{.sig}}
I can't wait for all the idiots to start posting, "But they're going to make it just like Windows!!!"
While the reality is is that not only are 'regular' users familiar with windows, but MS has spent significant resources studying exactly these issues and they are common even to those who are not familiar with windows. Many of these concepts (which the Linux community has shunned for years trying to avoid being like Windows) are going to have to be embraced (and extended) by the Linux community if they are going to gain any mindshare in the population.
-Adam
This sig 80% recycled bits, 20% post user.
Back when I was administering a Linux network, the most common questions from users were "Where's my C drive?" and "How do I make all those weird directories dissapear?" (referring to the stuff in /, *after* unsuccessfully trying to delete them). Good thing they didn't have root access...
This is really great work. Bravo.
It's sometimes painful to watch the average user struggle with things that seem obvious, but this is exactly the kind of feedback that is difficult for a programmer to get.
For linux to succeed on the desktop, Gnome (or KDE or something equivalent) must do well at meeting the needs and expectations of this kind of user.
I hope people will see this report as a very valuable insight into what goes throught the minds of ordinary users. I did. Hell, I've even thought the same thing as in some of the comments (I shouldn't have to click on the login box to type my username!)
Indeed. The really, really funny thing is that a lot of Windows users can't even use a very slightly modified Windows system.
For instance, I used a hex editor to change the word on my start menu from "Start" to something useful (i.e. the name of the computer, Astarte).
Every single person who sat down at it saked, "Where is your start button?" immediately. Even though it was in the exact same place. Nothing else changed except the word on it. I'm not talking about just my mom; I'm talking about engineer types, people who have used other operating systems (primarily SGI Irix) extensively.
If the people in this survey had clicked a nappy foot, or a K on the top of the screen, or even some "useful text string in the lower left hand corner" every day for 10 years, they would sit down at a MS windows box and say "Where are the programs at?" or, "'Start'? what a stupid name for a computer!"
Neh
... and there is no doubt, that one day he will be
where the eye of his telescope has already been
Someone forgot to click preview (and Slash won't allow me to resubmit)
/usr/share/appmenu ?
:1' is not intuitive
All comments based upon Ximian GNOME 1.4 updates as of 20010722 and KDE 2.1.1 wit the KDElibs 2.1.2 patch applied.
* Gnome 1.4 can't make icons (launchers) on the desktop unless dragged from Nautilus. One can't modify those properties. Which is bad. GNOME 1.2 and KDE allow this.
* If a launcher can't find a program, I don't get an error message (must less GNOME having the brains to find the program). KDE allows this.
* If a launcher runs a program that spits out text, I can't see the text. I think the launcher shuld wait and see if any windows are being launched by the program, and if they don't pop up within a given time limit, show their text output.
* Nobody selects their apps based on toolkit. My mom doesn't ask for a GTK app. She want's soemthing to read her mail. So can GNOME and KDE start both start using a directory like
* Indeed, Programs = Applications. Both GNOME and KDE suffer from this bug.
* Ximian GNOME 1.4, with all updates, is still very slow on an Athlon 900 w/ 384MB RAM. Or, more specifically, Nautilus is bad.
* Implementation of things (semitransparency) which should be in X into GNOME is a bad technical decision.
* The GDM bug which allowed entry only when the mouse was over the dialog has been fixed a while ago.
* If I try something in Control Center, I shouldn't need to commit the changes.
* Any app that ever tells me I should be root should be shot. If I have permission to su, ask me for the password. Otherwise, tell me I'm not allowed to run the program.
* Windows XP GUI multiuser capabilities are unfortunately better than GNOME and KDEs. I.e., its possible to go back to the login manager and log in as someone else while the other user keeps their GUI session open.
Changing to a VT, logging in as another user and running `startx --
* I want to change the layout of the window buttons. How do I know what NextStep / macOS / Windows used? I just want the X on the left!
* Red Carpet is great, and should replace GNORPM as the standard software installation method. For that matter, what exactly is a GNORPM? Let's rename Red Carpet to `Software Installer or `Installer' and put it on the default desktop (or high up in the menus)
* GNOME and KDE require other apps to be launched to modify their menu structure. Its not achievable via drag and drop.
* Like the study says, a foot isn't immediately apparent as a launcher. Neither is a giant K. KDE call this button the go button, and should fix the imagery to be more obvious (a `Go' street sine that pulses when people first log in?). GNOME should also do something like this.
I'm not sure "shell" would mean any more than "terminal emulator"....
Agreed. Hijacking of the term shell to mean CLI is simply wrong. A shell is something designed to protect users, or something that covers up what is beneath. Everything is a layer of abstractions at some sense (you use bash? Real men who know what they're doing write to their disks by hand with magnetised needles).
My shell is KDE. My friends shells are explorer.exe and GNOME
I think `command prompt' would be an appropriately specific title.
All comments based upon Ximian GNOME 1.4 updates as of 20010722 and KDE 2.1.1 wit the KDElibs 2.1.2 patch applied. * Gnome 1.4 can't make icons (launchers) on the desktop unless dragged from Nautilus. One can't modify those properties. Which is bad. GNOME 1.2 and KDE allow this. * If a launcher can't find a program, I don't get an error message (must less GNOME having the brains to find the program). KDE allows this. * If a launcher runs a program that spits out text, I can't see the text. I think the launcher shuld wait and see if any windows are being launched by the program, and if they don't pop up within a given time limit, show their text output. * Nobody selects their apps based on toolkit. My mom doesn't ask for a GTK app. She want's soemthing to read her mail. So can GNOME and KDE start both start using a directory like /usr/share/appmenu ?
* Indeed, Programs = Applications. Both GNOME and KDE suffer from this bug.
* Ximian GNOME 1.4, with all updates, is still very slow on an Athlon 900 w/ 384MB RAM. Or, more specifically, Nautilus is bad.
* Implementation of things (semitransparency) which should be in X into GNOME is a bad technical decision.
* The GDM bug which allowed entry only when the mouse was over the dialog has been fixed a while ago.
* If I try something in Control Center, I shouldn't need to commit the changes.
* Any app that ever tells me I should be root should be shot. If I have permission to su, ask me for the password. Otherwise, tell me I'm not allowed to run the program.
* Windows XP GUI multiuser capabilities are unfortunately better than GNOME and KDEs. I.e., its possible to go back to the login manager and log in as someone else while the other user keeps their GUI session open.
Changing to a VT, logging in as another user and running `startx -- :1' is not intuitive
* I want to change the layout of the window buttons. How do I know what NextStep / macOS / Windows used? I just want the X on the left!
* Red Carpet is great, and should replace GNORPM as the standard software installation method. For that matter, what exactly is a GNORPM? Let's rename Red Carpet to `Software Installer or `Installer' and put it on the default desktop (or high up in the menus)
* GNOME and KDE require other apps to be launched to modify their menu structure. Its not achievable via drag and drop.
* Like the study says, a foot isn't immediately apparent as a launcher. Neither is a giant K. KDE call this button the go button, and should fix the imagery to be more obvious (a `Go' street sine that pulses when people first log in?). GNOME should also do soemthing like this.
If the topic is user interface design, nothing important will be said until everyone involved has read 'About Face: the Essentials of User Interface Design' by Alan Cooper (here on amazon). I've been re-reading it again, and am amazed at the insight and the prescience.
.... Ok?'.
Fundamental is the difference between good software engineering, and good user interface design. The automobile industry recognizes this gap - most users don't see what engineering is embedded in the engine, drivetrain, etc, but everyone sees the smooth lines of the body. And the second you sit behind the wheel you can tell if the designers intelligently arranged the controls to be easily accessible and clearly read. NONE of this has ANYTHING to do with the engineering of the car - it could have a revolutionary new suspension system, but that's not what you see. The same gap exists between user interface design and software engineering. An elegant use of pointers is invisible to your user. The relative elegance of software engineering techniques means nothing for user interface design.
Programmers tend to be BAD at interface design - BECAUSE we understand more about how the machines and software work than the average user does. We know the box, so we don't think outside the box.
The book opens with a great discussion of a user's goals, which are usually NOT to recompile a kernel. Cooper says users want: to not look stupid, to get an adequate amount of work done, not be too bored. These goals are clearly not addressed by error boxes that pop up saying 'library x caused a page fault at
I've been thinking about this one. One idea I've had is "liberating" some of Jacob Berkman's code from the gnome panel (simply as a matter of laziness, since they already have a "top of the screen widget" and have dealt with all the Xlib root window cruftiness) and then bonobize it, and then hack Gtk at the lowest level possible to communicate with the bonobized menu widget. Of course, you'd have to deal with updating that menu every time you switch to a different application.
KDE does the global menu bar (from what I've heard) through giving the WM a few clever hints. I really should check out their code when I have time. It's probably a much more sane way of doing it than my previous suggestion.
Go wolfpack!!!
look like Windows"
Apple spent far more resources than Microsoft developing their UI, and many of the choices they made were because they were shown to be effective in the usability lab. Microsoft made many of their UI decisions not because they were well thought out and well tested, but because they were different from apple and less likely to get them sued (though Apple filed suit anyways). If you try to do the opposite of what is well researched, you'll often end up shooting yourself in the ass. Of course, if you've got a monopoly, it doesn't matter how unusable the damn thing is.
If the linux community is to succeed in the market, they will have to accept that the people who designed and used what they consider to not be a "real computer" actually know a hell of a lot more than they do in a certain area. If they are smart, they will accept their guidance.
Yah. It's called "Bash." It's a little esoteric to use, but it's damn fast. You can extend its functionality with plug in modules called "programs." For instance, for a lot of flexible multi-directory operations there's the "find" program. Hundreds of such "programs" have been written and many of them are probably available on your system already. You should experiment a bit with the fast, flexible bash file manager.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
I have Gnome installed on my Linux box, but more often than not, I use the command line to naviagte through my directories. Why? Because it's faster. The Gnome file manager takes about 3-4 seconds to open a directory on my 350 MHz box. If I'm trying to find a directory 4 layers deep, this adds up to annoyance.
Does that mean the command line is better? Definitely not! I'm a Mac user and I love Apple's interface. But in Mac OS, opening a folder is instantaneous. I can find a file in Mac OS at least as fast as I can using a command line.
What it comes down to is this: In designing a user interface, there are tradeoffs in speed vs. functionality. I would like to see more speed, specifically in the Gnome file manager. I'm quite willing to trade some features for more speed, because without the speed, the features aren't useful to me.
Does anyone know if there is a lightweight, fast file browser out there?
Is there a mirror anywhere else?
Even if it wasn't that humor has a full right to a place in the community. 90% of humor is about the failabilities of others, the other 10% is about ourselves. The submitter isn't hurting anyone by finding that concept amusing, especially not in an environment that is primarily *nix users in the first place.
I'm the big fish in the big pond bitch.
Well you're kind of ignoring my post, and directly attempting to make me feel bad. Fortunately I know I make mistakes and it doesn't bother me at all, especially when I have a limited time to get a post in and have to type quickly.
I'm the big fish in the big pond bitch.
Dare me huh? You got it, but then I run XFS. I can't reccomend running a journaling file system highly enough.
"Prefiero morir de pie que vivir siempre arrodillado!"
Take this ball and run with it, someone! Don't just bitch about how "Microsoft has conditioned everyone to look for a Control Panel!" Who gives a shit? If your intent is to write software for use by the masses, you'd better be damn sure and write it so that the masses will like it and want to use it!
--SC
You read fiction? I write it! Lemme know what you th
Why ? because the strength of the KDE look-and-feel is that it's a close copy of Microsoft Windows
But the funny thing is that I also feel that KDE's greatest weakness is that it is a close copy of MS Windows. As a battle-hardened Unix user who started with austerity of TWM, has been scarred by OpenWindows, scratched by CDE and finally found salvation in FVWM before the time of Sawfish's appearance on the scene, KDE's default window manager feels too constraining. It's not configurable enough - I can't push into working in the way I want it to work. I can't program it in Lisp. The list goes on :-)
That's not to say I don't like the KDE stuff - I just run it from my Gnome desktop. The apps are fab, but you can keep the window manager.
Thankfully we have a choice :-)
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
I think this may have been a flaw in the test. The users were simply plopped in front of terminals and told there was a new system on it. In the real world, there would be some exposure to advertising. There would be e-mails sent to employees. There would probably be a meeting about it or somethig. Usually people see the product logo before it ends up on their system.
You may be right. But ideally, it would be nice if they were plopped in front of the system and everything made sense to them. It's a goal to shoot for, even if it can't ever be reached. The foot may actually be a poor choice for a logo since it isn't immediately recognizable as one, but I don't think GNOME is likely to change it based on one usability survey. That'd be way too much work for too little benefit.
I would certainly say that the meaning of that one button is not as big a deal as many of the other things they revealed in the test, since once they know it is the logo it makes sense and they will remember. The settings under programs thing, on the other hand, I could easily see confusing them over and over again. The same with many of the other points raised.
Reading it, the comments seemed to be a lot of things like:
"This is ridiculous! The start button is a foot? What does a foot have to do with a start button?"
Read a little further. They gave the participants the very important hint that the foot is the GNOME logo and then:
Their guesses were all dead on. If you didn't know that the footprint was the GNOME logo, you'd be confused, too. Think of all the associations you can make with a footprint. Traveling, history...exactly what they guessed.
"Whoa? How come the settings are under something called "Settings"?? Where is the control panel?"
Your paraphrase lost the meaning of the original. Try this instead:
They were not confused that the settings were in something called "Settings"; they were confused that the settings were in "Programs". Sounds like a pretty valid complaint to me.
Microsoft has succeeded in making their own screwed up naming conventions the "standard" of computers everywhere.
No. The users' expectations you've quoted were reasonable and not centric to a Microsoft desktop. You found what you expected to. You completely ignored all the information contrary to it.
There are advantages to testing usability with Windows users. If you want to convert people from Windows to X, you need to make sure that the Windows people will not be entering a hostile environment.
For linux to work on the desktop (which I personally think won't happen but that's not the point), you need to make sure Windows people can migrate over.
Now this elite 'smartier than thou' attitude towards end users is precisely what is going to keep linux away from desktop users.
Even if you take 30 minutes to learn the interface, doing something because you were told to do it that way is different than doing something naturally or instinctively. If my car had you push down for a right turn signal (instead of up on the lever), it'd take you merely a minute to learn it. However, it would still feel wrong and usual, and you'd probably hate driving the vehicle.
Sure, you can learn a UI. But the more natural it feels, the more instinctive it is, the better. There is a lot of money spent on user interface research and evaluation. If it was as simple as saying "spend half an hour learning how to use it", then companies would of been wasting millions of dollars.
Agreed... if they wanted to use a word 'Start' was not the right one. I've been using computers in some, way, shape or form, for 10 years now. When Win95 came out and I tried using it it took me a while to figure out what 'Start' was really for. I would click on a desktop icon to select it, then hit 'Start' thinking that would launch it. Took a good half an hour for me to figure out how to use and customize the menu. Very frustrating really.
Something like "Main Menu" would have been more logical, or, heck, just "Stuff".
I'm running Ximian on my desktops now, and I like the ability to break Programs, System, and Help off onto their own little menus on the pager. Takes away that extra layering of menus w/out cluttering the pager all up.
I think this may have been a flaw in the test. The users were simply plopped in front of terminals and told there was a new system on it. In the real world, there would be some exposure to advertising. There would be e-mails sent to employees. There would probably be a meeting about it or somethig. Usually people see the product logo before it ends up on their system. Actually, when you run startx the logo comes up then, but they didn't have them do that. I'm sure that would bring back a lot of memories of having to type WIN every morning. I'd say most Linux desktops are about where Win3.x was in terms of useability.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
>"Terminal emulator program" means NOTHING to the "average" user, but "Shell" "Console" or "GNOME Prompt" would.
I don't think so. When I help users over the phone and I tell them to open a "DOS prompt" or "Command prompt" about 3/4 of them say 'What!?!? How do I do that?'
>Linux will probably NEVER get the drooling idiots, but do we want them?
Yes, we do. There is strength in numbers due to something called "economies of scale". Look it up. Also it would be in bad taste to abandon Grandma just because she has a hard time with Windows.
I have heard of people referred to as Linux 'elitists' before, but I always thought they were a myth or just Microsoft devotees pretending to be Linux enthusiasts to give us a bad name. If you are neither of those things then please think hard about how to include those who are less adept with technology.
I think that Linux distro's will be far easier to use than Windows in a few short years, like Palm Pilots on steroids, yet the advanced user will be able to get to all the powerful tools which are tucked away out of sight.
how is asking a question about what a certain unfamiliar icon does "funny?" i should hope you aren't ridiculing the user (haha, laugh at the user, he doesn't know what that button does, what a moron) such "humor" has no place in the community -mike
"I would kill everyone in this room for a drop of sweet beer."
--
I'd rather be lucky than good.
This is really, reallly funny. I am in slackjawed awe, can you imagine where this method of argument could take us?
The possibilities......
What exactly does GNOME have to do with argonomics? The last time I checked, a desktop environment has no bearing whatsoever on soil or plnat sciences.
Conformity is the jailer of freedom and enemy of growth. -JFK
I find it somewhat ammusing that the Sun study basically recommends, where applicable, that the GNOME features be made to work more like windows. Not because windows is better (or worse), just because most novice users think the same way and Windows has already taken most of that into account through extensive usability testing. The fact that some of the users in the test had already used other, well-tested systems, like Windows and Mac obviously shows through. But I believe that if you took computer 'virgins' and ran them through the same tests, they'd probably come up with similar results, especially in the areas of error-elimination and better up-front user education.
It's interesting that Linux and Windows are tending towards similar user experiences, but from completely different angles: Unix has traditionally been for the expert and is moving towards the novice. Windows and the Mac have traditionally been for the novice but both are providing more customizability and advanced features for the expert.
And that's why you set their shell to Word, Outlook, or whatever application they need to run. That way, there is only one application, and you don't even have to train them to "click on an icon." It loads automatically. But they probably padded their resume anyways when they said they know how to use Excel.
That also has the advantage of stopping them from browsing the web or otherwise wasting the company's money while they're at work doing personal things.
This would be necessary, of course, to include with a very restrictive system policy to lock down non-approved applications.
And before you "free-everything" loud mouths pipe up and say that infringes on people's rights or is not proper, try running a company. See how productive your employees are if you end up paying them to surf the web 2 hours a day, talk at the coffee machine another hour, and end up working 5 hours (and probably not working very hard either).
These are the kind of people you fire immediately. They have no "right to the web" at work. They can do this at home on their own time.
You know these users. The ones where you say, "Right click on Outlook" and they first move the mouse around while searching for the Outlook icon on the desktop (hopefully it hasn't moved yet or be prepared to wait 5 more minutes, unless you have the audacity to point it out for them), and then glance at the mouse to make sure they click the "right" button.
What's really sad, is that these people are often the sysadmins as well.
Hmmm. Yet Apple, who are known for actually giving a rat's ass about the Real Usability of their interface have had an Apple menu for the last 16 years and no one was so confused that they stopped using it. The all-purpose, one great menu really doesn't have a good name (although I agree that start makes at least as much sense as any alternatives) and the logo of the desktop maker is as good as any. Personally I prefer they not put a word on such a menu, since it's meaningless.
I do not have a signature
Sun's new studies on how the GNOME interface works in my opinion is a bit too ad hoc if they want to improve its look and feel.
They're several years behind what Microsoft has done with their famous Usability Lab. Look at the way Microsoft has designed Windows 95 interface that went into Windows 95/NT4/98/ME/2000; this is due to lots of hours of extensive scientific research into how people use their computers. It's also the reason why the interface of Internet Explorer 4.0 to 6.0 is way, way more polished than the interface of Netscape 6.0x (Netscape blew it big time by abandoning the interface used in Netscape Communicator 4.x versions, which was actually quite good).
What all the Linux supporters need to do is to fund the equivalent of the Microsoft Usability Lab for both KDE and GNOME. I'll guarantee that both KDE and GNOME will be much easier to use once some decent research is done on making the interface easier to use by end users.
Of course it makes sense! You're "kick starting" the system to turn on!
That makes more sense than a big assed "K" any day.
--- Jump!! Fire!! Bullet time!! - Lego version of the Matrix
1. Try Red Hat 7.1, it is much much better.
2. As far as windows being sucky being a form of job security: yes, but what you can do with Linux will amaze people. My boss and I are constantly amazed about what we can pull off replacing costly licenses with free stuff, that does the job better too.
-
I've had enough abrasive sigs. Kittens are cute and fuzzy.
Why ? because the strength of the KDE look-and-feel is that it's a close copy of Microsoft Windows, and this is good for 2 reasons :
People who can use Windows are not disoriented by KDE (rah rah, old argument, I know ...)
Microsoft being all about "first user experience" (read glass and chrome on a desktop anybody can use more or less intuitively), they probably spent a ton of money on the design of the Win95 interface, so why not reuse it ? It's far from perfect, but you can be pretty sure it'll be accessible to the mass thanks to M$ money, and KDE reuses all that R&D for free.
M$ is not stupid, and they've been reusing the same old clunky Win95 interface for years now. IMO, that's because they know for sure it's what flies with the users. So, I like Gnome, it seems solid and well built, but I'm sorry to say, it'll probably lose the GUI battle because its single biggest flaw is its non-M$ look-and-feel.
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
"This is ridiculous! The start button is a foot? What does a foot have to do with a start button?"
and
"Whoa? How come the settings are under something called "Settings"?? Where is the control panel?"
Microsoft has succeeded in making their own screwed up naming conventions the "standard" of computers everywhere.
In the future, will all UIs have to have start buttons, control panels and taskbars to be considered usable?
There is one big problem with having the menubar at the top of the screen.
If your window manager is in point-to-select mode, when you try to go to the menubar of your application you would inadvertently switch which window is the active window. Then the menubar would be for the wrong program.
Got friends?
None of the comments are all that suprising...
GUI enviroments simply aren't all that intuitive, period. There may be ways to make them more intuitive however this study, while interesting, appears to be more a measure of how similar to MS Windows, Gnome is.
This is not to say the study is without value. Certainly it is valuable, but agronomic design just isn't at a level where a user can sit down and intuit the functions in such a complex devide as a computer operating system. It just isn't possible. This is not a reflection on Gnome so much as a reflection on the study of agronomics.
That said, the study was a good read and did make valid points in it's recommendations. It's just important to keep in mind what was actually being analized.
--CTH
--
--Got Lists? | Top 95 Star Wars Line
I often come in contact with users who are moving from old unix based terminals to windows based software. Many of these users have absolutely no GUI experience whatsoever.
There are similar "stumbling" blocks that various users hit when switching platforms -- either from a text-based unix terminal to GUI or from a MAC to WIN32, or whatever.
For most "work" environments, if the user knows how to "click" an icon to run his software, thats about all they NEED to know -- outside of how to operate their software package. It's silly to expect a 50+ y/o client services secretary from company X with no GUI experience to "master" any windows/mac-ish interface. Their "job" is to be able to navagate their software, launch it and shut it down. Beyond that is really expecting too much.
-jhon
If the same study was performed on the win32 desktop, by the same group, it would pass with flying colors.
This was a study of usability and familiarity. Like it or not, the win32 desktop is usable and familiar to Joe Average-computer user. It is Gnome that, for most users, needs the changing to make it friendly.
Personally, I like Gnome (and KDE) much more than the win32 interface, but it's because I'm used to them. To appeal to the wider audience, changes will have to be made to make it more like the familiar desktop, which in this case is Windows.
"Why do I need to press start to shutdown."
... mean?"
"What is the any key?"
"What does Fatal Exception in
"Press yes if you want to reboot."
"No, I didn't send you that email."
"Die clipply die."
"Click on this, right click on that, double click there, oops sorry press back."
"Why am I out of memory, I'm not running anything."
Users are dumb no matter what the OS.
What is pirate software? Software for inventory of stolen treasure?
A lot of it is common sense, but many X developers would do well to go through the site. Fortunately the GTK pushes developers in the right direction (build the tools, and you can implicitly enforce the standards), but we still have a ways to go for GNOME to be as consistent as, say, Macs were in the late '90s.
Oh, and M$ bashers will have many opportunities for chuckles here. :-)
I'm a bloodsucking fiend! Look at my outfit!
- Look.
You see files here. Also, a Trashcan and the Internet
- Take Internet.
You can't do that!
- Drop files.
Where?
- Drop files in Trashcan. /
rm -rf
Done. 261792K deleted.
- Ah! Undo! Undo!
I don't understand that.
- Get files from backup!
I see no backup here
- Get backup from Internet
It is getting dark. You are eaten by a grue.
C:\
I'm a bloodsucking fiend! Look at my outfit!
But in my opinion, KDE is a bit more usable for two reasons: the file manager is great, and there is a nice usable feel to it. GNOME is cool too, though.
For people coming from a life of windows, linux can be a very scary world. I had problems from day one, (day one being not so long ago), just getting around to things. I wanted to set up my NIC and I was trying to figure out where i could go to Network Neighborhood, finally realizing there is no network neighborhood. Then there was changing the resolution...I rightclicked expecting to see Properties and go through that easy Windows process. NOTHING! That took some time too.
Granted I am not the smartest person in the world, but I am a CS major and computers are my life! I was so used to the Windows way of doing things that shifting over for just the easiest things took some time.
The worst part is that apparently unclean stoppage of Linux is a much worse prospect than Windows...I suppose it really isn't an issue because of Linux's supposed reliability...but if anyone hasn't yet simply turned their computer off accidently while running Linux....I dare you..