Slashdot Mirror
Shuttleworth Suggests 1-Way Valve For User Experience Testing
darthcamaro writes "No surprise but Ubuntu's Mark Shuttleworth has come out swinging in favor of the Linux desktop. Speaking at Linuxcon yesterday he detailed the things that he thinks Linux requires in order to win the desktop wars. Those include: co-ordinated software releases, better quality and design, some user experience testing and oh yeah, a dose of 'shut the f*** up' too. During his keynote, he extended an invitation to any open source application to submit their software for testing by user-experience experts. The sessions would be recorded for posterity, and the developer would not be able to interact with the user. "'If the developer is in the room, they have to say nothing. It's the shut the f*** up protocol,' Shuttleworth said. 'You sit and watch someone struggle with the software that you've so lovingly produced.'"
I've done a bit of software dev here and there, and I've never had the luxury of being near the users when they first prop it open.
For that reason, I've developed a habit of showing a beta to a nearby co-worker, or a friend, and ask them "Check this out."
And when they say "What is it?" - I haven't done my job right.
We -need- RMS though. Without RMS we just have a bunch of people wanting to get stuff for free. Heck, without RMS and the GPL, Linux would not exist, Linux as in the kernel itself. Chances are it would have been licensed under an obscure license and died due to a non-commercial or other clause. It was only due to the GPL that the kernel was released under a typical license.
Taxation is legalized theft, no more, no less.
Yup, Linux needs more hype.
:) is ask a new installer if they want to use a "Windows GUI compatibility mode" and then have the system laid out as close to a Windows system as possible, out of the box.
/dev/sda1 stuff. Streamline Aptitude and enable all repos by default - a Windows convert dosen't give a shit about licensing issues, they just want those graphics drivers. Etc. etc.
Gnome's U.I. is already easily navigable. What Ubuntu needs to do to attract more Windows users (non-gamers, of course
One "start" menu inline with a single taskbar, on the bottom. Combine the items in the "preferences" and "administration" menus into a single "control panel". Alias the filesystem items to things like C: and none of that
They need to embrace, extend, and extinguish. Beat Microsoft at their own game. Maybe even make all that a seperate "Windows compatibility distro" so the purists won't bitch and moan.
"Jim! I'm on the Lynooks now, and I printed off 500 envelopes for the newsletter, but they're all rotated! I put the envelopes in this way, but they come out all wrong!"
You seriously overestimate the ability of a standard plebe to adjust to any change.
If you fucking switch out their pen from a twisty pen to a clicky pen, it's not a difference, it's not a preference, it's a problem, and the new way is wrong, and it's your job to get them the damned pen they like.
Humans are great at adapting, but only when forced. Then they'll never stop bitching about how good it was in the past.
And then be beaten with sledgehammers until they understand that the goal should not be 'unconfigurable' but 'no configuration needed 90% of the time, and configurable the remaining 10% of the time'.
Amen to that. I've ended up swapping my desktop environment to KDE4 for my personal computers. It actually *makes sense* out of the box. I was absolutely thrilled to see that installations and management was actually 'easier' than on Windows. Normally, when I start running a Linux distro, I end up on google, but not then. (For the record, I'm running on Kubuntu.)
"Sorrow is better than laughter, for by sadness of face the heart is made glad." [Ecclesiastes 7:3]
You can still perform plenty of validation testing irrespective of what platform a person has used. If a given person can't figure out what your trying to do with the tools provided than the software needs work. When I used to do work for a manufacturer we took people off of the assembly line or equivalent, made sure they knew nothing about computers and used them to perform the testing. If they couldn't figure things out on their own, the test was considered a failure. Blaming the platform or the userbase is a sign of a poor developer, it's no different than blaming your tools.
The entire point of such testing is to remove assumptions and find out what happens in the real world with people that don't have the programmers base line knowledge. Developers have a way of assuming a given level of knowledge that users simply don't have. That's why people like Mark Shuttleworth have done so well, they've presented Linux in a way that simply doesn't require that baseline knowledge. The issue is not whether or not a given tool is capable, well written, more efficient or otherwise. The point he is trying to make is that the issue of acceptance and use by the masses comes down to usability by the masses (documentation can and should heavily influence this).
If your making software only intended for highly trained users that will go to school to learn how to use it (SQL, CAD etc.) usability may not matter as much because you can assume the user has a baseline of knowledge. If your not making software that requires specialized schooling to use, than you should be testing software in a manner as suggested.
I'm a coder, gamer, and all-around power-user. I've been using Linux for years, including 2-3 years of using Gentoo exclusively, back before it had any sort of gui installer.
In all that time, I've only had Gnome not let me do something (or make it overly difficult) twice: once was when they went to "spacial" (I think they called it) handling of folder-opening in Nautilus, which was only a slight pain to fix and which, AFAIK, has been switched back to a not-retarded default anyway, and setting each virtual desktop to a different background, which I'd still like to be able to do but which really isn't that big a damn deal.
What exactly do all these "Gnome won't let you configure anything! KDE 4Evar!" people want to be able to do with Gnome that they can't?
No, I think it is because Apple has made a big difference between a "PC" and a Mac when people think about them. For one, the hardware is different. You are generally typing on a different keyboard, using a different mouse and looking at a different monitor. With Linux you keep all your same hardware. Plus, because the idea of an operating system has been lost in culture with the exception of the mostly-hardware locked Mac operating systems, the different versions of Windows and hardware-specific OSes. People don't understand that Linux is not a free version of Windows even though it does the same tasks.
Taxation is legalized theft, no more, no less.
I'm confused.. weren't you talking about word processors just two lines back? Now it's operating systems? What aspects of that operating system, exactly? Are you talking about the desktop management or the CLI?
That said.. I understand what you're trying to say.. that people are biased from their experience with a 'competing' product.
On the other hand, that bias may not be a bad thing.
Just as an example... The GIMP vs Photoshop.
If somebody had never touched a graphics editing application before and just got around to copy/pasting something and would be wondering how he or she might rotate and enlarge that bit they just pasted.
In The GIMP, this is very straightforward. There's a rotate tool, and a scale tool. Almost couldn't be easier. The GIMP developers rejoice.
Yet, they don't, and are seeking a unified transformations tool. Why? Because people's experience with other graphics editors shows them that users realize the added value of a transformation tool that can do rotating and scaling (and sometimes more) at the same time via on-screen feedback; both in terms of workflow -and- in terms of the quality of the result.
But if you only let people who never touched another graphics editor test the existing tools, you'll have to wait for that one-in-N person who goes "wouldn't it be easier/better if..." to get to that "wow. that's so obvious when we look at it now, why didn't we think of that" point.
That's why you want a diverse set of testers, and that includes testers intimately familiar with competing products. It's your task as a developer (or usability expert) to find out whether their bias is justified (i.e. the expectation isn't odd, it's simply logical) or not (expectation -is- odd).
( fwiw - though gui.gimp.org is not responding - http://gui.gimp.org/index.php/Transformation_tool_specification , http://www.mmiworks.net/eng/publications/2009_03_01_archive.html , http://gimp-brainstorm.blogspot.com/search/label/transformation%20tool )
I wanted to tell NetworkManager to do something specific (IIRC, use a specific DNS server rather than the one handed out by the DHCP server on my DSL gateway, but it's been a year or so) and couldn't. When I opened a ticket about it, it was closed WONTFIX with the notation that the idea behind it was zero-configuration and adding the ability to configure it to do this was therefore unacceptable.
I want gnome-terminal not to eat my right-clicks. People have been asking for that for *years* and are constantly told that the Gnome developers know better than they do about what they need.
True story here: dad's computer had OpenOffice, not MS Office. My sister's experience with OpenOffice's Impress was terrible: she needed to print all slides from a .ppt file, and couldn't find this option. As she had a tight deadline, and I had nearly zero experience with presentation software anyway, I shrugged and installed MS Office. She ran Powerpoint and found her way very easily.
Just a bit later, I tried to find out how one prints all slides from a presentation.
Guess what? It's done EXACTLY the same way in Impress and Powerpoint. Same function, same name, same location. See, this is not a "Photoshop versus Gimp" style comparison; interface-wise, they were nearly identical (that was before the "ribbon" thing). If she found her way in Powerpoint, she should have found her way in Impress. Yet, she somehow panicked with the new program.
What can a developer do about users that won't even TRY?!
Circumcision is child abuse.
... as a developer.
They basically have labs with one-way mirror. User is left alone in a sound-proof room and given a set of tasks to perform. Everything is recorded (including facial expressions and sound), and any developer can take a look at the test either from the adjacent room or from his/her workstation (using Windows Media Player). The only input the user gets is when he gets so confused he can't accomplish the task from the list. In which case the person conducting the test just says "next task" and that's it.
The experience is really humbling. You just realize that people out there are FAR, FAR less experienced with computers than you thought, and even working their email client is a challenge for most.
You make your assumptions on the basis of what's convenient for you. Guess what, people out there are not you, and what's good for you is torture for them (the inverse is often true, too).
We ended up redesigning the entire chunks of the UI sometimes, some features got cut, some scenarios overhauled. And in the end we still didn't do enough of usability testing (IMO), but such is life in commercial software development - you work against an arbitrary schedule.
Why should software go through Ubuntu to get validated by UI Experts?
I'm guessing the *primary goal* is to get developers to have UI experts look at their software, PERIOD. I'm sure Shuttleworth would be happy if it were someone else's UI experts.
The sad but true fact is that today, the vast majority of open source software *never* has any usability testing done.
Read it like this: "Linux software needs usability testing done. The Ubuntu project can provide resources to help accomplish that."
If he wants to make Ubuntu financially self-sustaining, Linux desktops that play well with media conglomerates aren't going to get anywhere.
Huh? What the hell are you talking about?
Bottom line, I get the feeling he sees himself as the great entrepreneurial hope for all of Free Software and that it, in general, will be successful when his company is successful. Well, Mr. Shuttleworth, they were doing fine without you.
Not in the realm of developing usable applications and OSes, they weren't.
*A consistent user interface doesn't exist. Mac's Finder UI looks remarkably similar to the Disk Utility, it doesn't help you work with either one! If anything, one builds expectations the other fails to deliver.
So, since a 100% consistent user interface doesn't currently exist, we should therefore give up and not even attempt to make one? If everybody thought like you, nothing would ever happen.
Comment of the year
I do usability studies as part of my job.
We do a one-on-one facilitated session with a user in one room, have an observer session in another room watching in real-time.
You want to have developers in the observer session, and part of the point of this is to change developer minds, and give them unfiltered feedback on what users are doing with their work. I've watched this in action many times, and it has a profound effect on developers.
Most developers write UI and processes for other developers to use. One example: 'you have to create a row or data entry object for a database table before you fill it out with data values' - developers and DBAs think like this, but most other humans think that the filling-out of data creates the row or object. for them it mimics the real-world concept of writing a note on paper; they don't think about creating the paper first. If developers want people in the real world to use their programs, they need to make them work in the way that regular humans expect, and the best way to convince them of that is to show them humans behaving normally...that is, not like developers.
It was odd to see Shuttleworth quoted as wanting "User Experience Testing". This is almost certainly a misnomer or misunderstanding of "Usability Testing" - which is part of (some would say tangent to) User Experience practice.
One important thing to know about usability testing: It's reactive. It's not generative. It can tell you what's wrong with your project, but it can't create new ideas about what project to create.
The latter goal is the domain of User Experience practice. User research, surveys, ethnography, rapid prototyping, shadowing studies at customer sites, JAD, search, site and other analytics (and yes, Usability Studies) all go into the User Experience (UX) practice. It's bigger than usability testing.
"Sorry, the simple fact is there is no need for another desktop OS.."
"Instead, how about focusing on being a workstation OS and a server OS?"
The subtlety between being a desktop OS and s workstation OS is lost on me. So is the need to differentiate.
If this is the thinking going on behind Linux desktop development, then I understand why it is still almost there, but never quite. No surprise.
ps- the SFTU protocol is truly needed in the Linux community. I still get the predictable responses to requests for help with Linux software issues:
1. "RTFM!"
2. "Did you install it correctly?"
3. "Doesn't Winblows suck?"
4. "Did you test your RAM?" (I like this one a lot)
5. "If you don't understand the documnentation, perhaps you shouldn't be using this"
6. "You should be using {insert another application name here, it need not be for the same purpose} instead"
7. "You should try {insert another distro name here} instead" (I get this a lot less nowadays)
8. "Go back to Windoe$"
More helpful advice from the Linux community is not what I wanted. I just wanted help. Of course, I expect a lot from the nonprofessionals that respond the most, I know. I hardly ever snap back anymore. I'm hopeful that one day Linux will indeed command a significant portion of the desktop market. I may even be alive then.
deleting the extra space after periods so i can stay relevant, yeah.
Well if a million users expect a certain UI widget at a certain spot doing certain things, what's there to stop someone from fulfilling this expectation?
If the goal is mass appeal to Microsoft fanbois, well, make it appealing then. It's much easier to change a bit of code than try to evangelize some million users. Improving any one's deep ingrained wrongness can backfire when everyone is used to it and has to adapt to everything new at once, that's life, always has been.
Car analogy: all car makers seem to have different layout of their reverse gear in stick shifts. We can't rip out all stick shifts, we cannot standardize, because people who've always driven a particular will lament for weeks when something changed. So we have a status quo for decades which nobody quite wants to change.
Microsoft got heavy flak, no, nuclear artillery, for every single change they did to the Windows UI in the last 10 years. People actually seem to like the "Windows standard"-mode of XP and all users at my company fought tooth and nails to keep that when we migrated to new terminal servers - they like it so much that people constantly ask if they could somehow revert Vista or Windows 7 to that look.
So Microsoft get's their own dose, really. Since XP, GUIs (and their userbase) have come to a point of maturity where progress can now only move forward very good reasons. We may use other window managers, different layouts or whatever, but to the general public, the Windows XP non-kiddy GUI mode has been the definitive gold standard for most regular people - for now more than half a decade.
When Microsoft could copy over the descriptive buttons from MacOS ("Overwrite:" [yes|no] and "Keep this setting [yes|no]" to File exists: [overwrite|don't overwrite] and "[Keep setting|abandon setting]" etc.), we're actually finished building a UI metaphor.
STFUnix
Saying "I'll probably get modded down for this" in a post is the best way to get it modded up.
Car analogy: all car makers seem to have different layout of their reverse gear in stick shifts. We can't rip out all stick shifts, we cannot standardize, because people who've always driven a particular will lament for weeks when something changed. So we have a status quo for decades which nobody quite wants to change.
Well, cars also have different lights controls, wiper controls, radio controls, gas trap side, instrument cluster layouts, etc... Yet I dont think that most people think of those features, or the position of the rear gear, as critical when purchasing a car.
So, bad car analogy. (even though the rest of your comment makes sense)
As long as the pedals and the steering wheels are in the same position everything else pretty much goes, everybody can still use the car.
I understand your point but it sort of misses the point of the article. There is a difference between "Can do" and "Can do without wanting to slam my head through the monitor." This is the difference that is being pushed here.
After all, there are legitimate usability complaints and complaints about how bug fixes are handled. Keep in mind that these are complaints heard by current users of Linux, a group that is likely well above average in terms of technical savvy as a whole. I don't know if you do/have done desktop support (not trying to be condescending...there are all different sorts of geeks on Slashdot) but average users get totally bent out of shape about really mundane things. If they delete one of their displayed columns in their EMAIL program they flip out and run around yelling "MY EMAIL IS BROKEN!" Things have to work quickly and easily or you'll lose them in a second. This means no "Just recompile your Kernel" or even "Just use Wine". Definitely don't tell them to "Debug the code yourself", as they won't even know what code is, much less know how to fix it.
Also something to consider, if you started a huge Linux Marketing campaign and it managed to secure a pretty big market share developers would find themselves with 10 times more bug reports/feature requests than they have now. As another complaint is the slow/non existent response time on some projects this problem would be compounded. A few more developers might come with the rest of the huge crowd but I'm guessing that for the most part, anyone who wants to develop FOSS software is already doing so. You might see a flood of for-profit companies jump into the mix with a bunch of closed source software to fill in the gaps, but then you are actually losing a lot of what makes Linux special in the first place.
Now, why is it that for Linux the totally clueless get their computer set up by an expert (don't give them the root password), whereas Windows XP is just set up like crap?
If you take Windows XP Linux and a blank computer and gave it to a totally clueless user, asking them to install it and get online, neither OS will result in "secure in the "kiosk" sense", as the user will inevitably have the root passwords needed. The only way Linux scores better security there, is because there isn't really any malware targeted at that platform. As we all know, once the user gives admin rights to malware, the computer is fucked.
However, if you have an expert lock down XP, the clueless user won't be "pwned" anymore than when using Linux. And they'll be limited in the same sense.
Sounds easy enough? It isn't. In the age when Compiz is THE default manager running over Gnome, we have the ancient long-standing problem: you have the cute "desktop cube", you have 4 desktops and you have only one wallpaper for all of them. Of course Compiz allows you to place 4 separate wallpapers, the problem is they will be obscured by the default Nautilus wallpaper.
And now Nautilus allows you to switch the whole desktop off (wallpaper + icons), it allows you to set transparent wallpaper (through which Nautilus default background will be seen), it allows you to set background gradient style and colors, but it doesn't allow you to tell it "don't draw background, let some external program do it."
I think this problem is as old as the "desktop cube" and possibly older. There are 3rd party patches but they haven't been accepted into Gnome.
45 5F E1 04 22 CA 29 C4 93 3F 95 05 2B 79 2A B2
I always employ the 'Shut the fuck up' protocol. Unfortunately for my testing team it is usually me who is shouting it . . .
Right. I work(ed) as a software tools developer in a group of such people. Most of our users were other technical people, a high percentage with MS in CS.
They needed to know how to use many Unix tools, plus the domain knowledge for whatever project they were working on.
Unfortunately, each tool developer had their own idea about what was an "intuitive" user interface, thus our users needed to learn them all...
When a "Brain dead User"(tm) tried to explain why they had difficulty using the tool for whatever their job was, they'd say something like "But then I need to do XXXX, and the tool won't do that" -- And the developer would say, "Sure it does, you just need to do YYYYY". As a BDU for some tools myself, I got really tired of that answer. Imagine a mechanical tool -- that response would be something like "Sure it can, you just need to stand on one leg and hold it behind your back with your other hand" (Geee, why didn't I think of that?).
Eventually I realized this was another place where "The Customer is Always RIght" applies, except many of us don't understand the proper meaning of that phrase: If the customer has a problem, you have a problem to solve, even if they don't properly describe it. It may be a matter of educating the customer, but it may be better to educate yourself -- how do they think it should work?
I don't know about where you work, but where I worked, it was hard to learn everything I needed to know about the product domain -- needing also to learn a couple of dozen idiosyncratic user interfaces was just extra work. The result was that even if I finally realized what the perfect tool was and made it, I couldn't get people who were already in overload to take the time to learn to use it. Then I started trying to understand what they wanted to do and how they wanted to do it, and then made something they didn't need to learn to use, it just worked the way they did, and did more than they expected.
Of course the next step for the company was retirement incentives for the experienced employees with layoffs to follow.
Double-click, to open executables.
Users have been throwing this one at Gnome/KDE devs for years. It seems it is too Windows for the idealists, to the point that it's become quite a fun troll in the bug reports.
In reality, its a stupidly simple feature to save uncessary repetitive steps. Working with a graphical file manager is more efficient, and for some tasks necessary if you want to be productive, having to pop a terminal open for something a file manager *should* be able to do is silly and should have been fixed by KDE 1.0. A silly example itself, there are much better (longer) examples which demostrate how Linux groupthink raises my hackles. 'Free' as in gratis/beer, sure but at times non-Free in terms of thinking. (The hell with karma)
After logging in slashdot still does not take you back to the page you were on. It's been that way for 20 years.
It seems reasonable to expect gnome-terminal to pass to ncurses applications all the mouse events described in curs_mouse(3x), including BUTTON3_PRESSED, etc.
But why spend time and money teaching when continuing on with what you have works?
It doesn't work. There's an entire industry built on the fact that it doesn't work. There are entire job classifications based upon the premise that it doesn't work. And we (The People) are no longer in a position where we have the luxury of continuing to throw money at this woefully broken piece of shit.
Hey, I finally got my first freak! Took you long enough!
You've gotta be joking. Every windows install I've ever done comes with a generic driver and even a nifty message along the lines "Your display settings are at a lower resolution ....." it won't give you directx (sililar to compiz) and the is gives you no help installing the manufacturers web site. You still have to know the manufacter and model of the card. On the other hand Linux just give you a story about how evil the industry is and still do it all for you. So with the ever popular time = money ratio Linux has the best ROI.
A loop, by its nature, continues. If that didn't make sense, start reading this sentence again.
For me, "successful" in regards to my own code means it has benefited me in some way, and that is:
A. Getting paid for it
B. Getting code contributions
Since I see it that way, I GPL my code. I consider it successful when somebody comes to me to pay for an improvement, or when somebody takes my code and improves it (since they have to release the changes).
Getting my code used by millions but not getting anything back is of no value to me.
I'm a developper, so I'm pretty computer-savvy, but have been using Ubuntu like a standard user (for tasks such as email, Internet, chat) for several years. That means I don't want to have to type console commands to set up standard things (like being able to share my files with my Windows PC with Samba...Ubuntu 7.x was a pain for that, Ubuntu 8.04 made this disappear and made me happy). Therefore, I don't read Ubuntu forums, and don't follow planned evolutions but I do update to each new Ubuntu version on release.
All this to say that I didn't know about this bug report/controversy for automatic update version. However this change was one of the first things I noticed on 9.04 and was very happy about it. Although I always clicked on the little icon to update my softwares in previous versions (i know how important updates are), this change makes updating easier, therefore it makes my life as a basic user easier and it's one more step in the right direction (and an important one that increases the likelihood i will one day recommends Ubuntu for less computer-savvy people, because it will help them keep their software updated).
So IMHO Shuttleworth was right on that one, and you should have used the STFU testing method before complaining.
You're not going to completely cover all the minute details, of course. But there's a ton of established de-facto standardization that you can emulate or reproduce without doing damage to your user interface.
To stay in the analogy domain: make sure the left hand control is for lights, indicators, horn etc. (= signalling) and the right hand side for wipers, air conditioning, defrost etc. (= climate adaption), no matter whatever side your steering wheel is on. If you need to turn, twist, click or move a paddle, switch or knob is less important, because at least the user is going to look in the right place and will probably figure it out quickly.
Some things cannot be standardized, especially the gear shift position, because hardware limits prevent you from mounting the gear selector on the right side of a right hand drive or left side of a left hand drive - but the position is obvious enough in the first place.
Problem are the myriads of possible reverse selector positions, because
- people usually need it in a hurry
- it obviously cannot be worked around
- inappropriate settings are highly dangerous
- it's possible to not notice a wrong setting at all (hence the sound most modern cars make when reverse is selected)
- it's possible to be underway for a while before noticing that this setting is ambiguous
This carries over to GUI apps and desktops. Desktop Linux apps are crammed with features, no less than apps for Windows and OSX. The problem is that the designers don't consider what tasks users actually do, and it what order. That's where extensive (and expensive) UI testing pays off for Apple and Microsoft.
Consider Blender. Its UI is much maligned for being a nightmare of contextual button panels, but the UI's density is not the underlying problem. What's wrong is that the designers don't take into account the workflow of actual users. The software's jam-packed with functionality, but much of it is mired in multi-step processes that make the workflow painful. It's almost literally impossible to use the app without tutorials.
Go read the docs and try to figure out how to apply a decal to a model in Blender. It's reasonably straightforward once you know how to do it, but it's weirdly complex. It's a perfect example of the mindset of a designer who's trying to implement a feature--"We can do it with an empty object!"--without considering the task the user wants to perform with it.
You're not going to win over any users by saying, "The function's in there! It's on a par with professional apps that cost thousands of dollars!" If a user runs into teeth-grinding frustration, the app's functionality doesn't matter. It's not what the app can do; it's what the user can do with the app.
Ok, that's true. NM's better than the awful GUI network config programs that came before it, and the alternative remains the same as before it existed (i.e. fire up a terminal and fix the shit by hand) but there's a lot of room for improvement. I'm with you--NM could use more custom-configuration options. It's fairly new, though, so hopefully someone will come along and fix it. If not, I fully expect it to be forked, sooner or later.
Only, that it's not. I've been able to set my own DNS server like FOR EVER in Gnome. Don't believe me? Right click Network Manager icon > Edit Connections > (Select connection to edit) > Edit > IPv4 Settings > Method: Automatic (DHCP) addresses only. Voila!
Though personally I much prefer the older gnome-system-tools network interface, being much simpler and easier to navigate, I find the latest Network Manager GUI needlessly complicated. And I hate how it abuses the tray area, clearly against the HIG, for something that should be an applet, but I have yet to see an OS that gets this right.
Huh? The only thing I can find about this via Google is that some people want right-click to mark end-of-selection rather than opening a menu. Is that what you're talking about?
Now that I think of it, this would certainly be a useful feature to have for some people. But in 10 years of using Linux, it has NEVER occurred to me that I might want to pass my right clicks to a terminal application. Then again, in what way does Gnome force anyone to use gnome-terminal? Following the same way of thinking Gnome should bundle Emacs instead of Gedit, or Octave instead of gcalctool.
I don't see why the basic utilities bundled with a DE shouldn't be, well, just a basic suite which can cover 99% of use cases. I, for one, am happy that my desktop offers me an easy to use terminal application which doesn't suffer from feature creep.
Now if we are talking about core desktop elements, like gnome-panel or Network Manager, it would be another thing entirely since you wouldn't expect even advanced users to switch them out. But installing their favorite tools, is exactly what you would expect from more demanding users, and while this is no excuse for not adding some simple features that would benefit everyone, it certainly means some obscure feature requests will sometimes not be implemented.
I agree though about setting different wallpapers on each workspace. This is the one single feature I have been missing since switching from KDE, back in the gnome 2.6 days.
I found that I quickly found how broken and how resistant to logical thought the learners are. I have never actually tried helping but my mother worked at one of these places teaching IT so sometimes I would be there. I will jsut highlight some of the worst things for people who have not tried this.
People forget the most basic things from one week to the next. One person basically started from scratch every single week and never got anywhere.
People cannot switch on computers, there is one large button (plus on small one) on the front of a PC, which incidentally has the standard universal power symbol on it. They need to be shown how to press it.
People do not read. For example in Outlook Express one user wanted to know how to find an email, there was a large toolbar icon (there was only one toolbar with maybe 8-10 buttons on it) with text saying something like find/search on it. When it was pointed out they say something like 'oh that was simple', they simply do not bother reading the text on the screen.
People cannot repeat the same process they learnt in one program in another. My mother is adapting material for a Dreamweaver web design course, for which a requirement is to do one of the beginners courses. There is a page explaining how to save the file. It is literally a list of instruction saying click file>save select the location from the dialog box, enter a name and click the save button. Naturally I have abbreviated the instructions because they actually take up a side of A4 what with all of the excess detail and screenshots. When I asked if this was really necessary my mother said that it definitely was for about half the people taking the course and this was a web design course which is categorised as advanced.
Some software is obtuse and difficult to use of course but it is not just the softwares fault.