aysiu
May 24th, 2006, 09:24 PM
A lot of times in the discussion of Windows and Ubuntu or even KDE and Gnome, I hear certain terms thrown around that are not well-defined or examined:
intuitive
The concept of intuitiveness does exist. For example, in Windows, as has been pointed out by many critics, it is not intuitive that one would click on the Start menu in order to shut down. Nor is it, in the drag-and-drop world of Mac OS X, intuitive that dragging an icon from the Dock to the desktop would delete it (it goes up in a puff of smoke). And, to be fair, it is not intuitive in Ubuntu that the /usr directory (supposedly short for user) would be off-limits to users and modifiable only by root.
But, as can be seen in the examples above, intuitiveness is overrated. After all, I see Windows users every day clicking on the Start button in order to shut down. They don't give it a second thought. My wife no longer tries to drag icons off her Dock to the Trash in order to remove them--she knows, almost instinctively it seems, to drag them to the desktop so they go up in a puff of smoke. Likewise, I always use sudo when copying files to or modifying files in the /usr directory in Ubuntu. I don't try to modify them as user and then smack my forehead, "Doh! I forgot that was owned by root. It tricked me again with the name /usr!" I think you'll find people quickly get fairly used to certain interfaces' behaviors, no matter how "counterintuitive" those interfaces are.
Of course, all the examples I gave are, in some sense, empirically counterintuitive--Start is a positive action and Shut Down a negative one; in a metaphor for the physical pushing and pulling, dragging and dropping indicates moving, not deleting, unless the object is being dragged to the trash; the word usr looks like user, but it can't be used by what it looks like. There are other counterintuitive behaviors that are not empirically counterintuitive but contextually counterintuitive--in other words, they run contrary to what the user is used to, whether from other parts of the interface or from a previous interface.
For example, it's not any more empirically intuitive to download a setup.exe from the internet and then double-click it and click a bunch of next buttons and the finish than it is to click on Applications > Add/Remove Applications and then click to install the applications you want and have them downloaded off the internet behind-the-scenes and then installed. But for someone who is used to installing software the former way, the latter way will seem counterintuitive because she has a certain workflow in mind (open web browser, search for software, download software, install software, delete installer) instead of a different workflow (open software manager, search for software, mark it to be downloaded and installed).
Bottom line: what you're used to always trumps intuitiveness or counterintuitiveness.
user-friendly
When people say user-friendly, they usually mean discoverable and able to be done with a mouse, point-and-click style. Even if you must go through twenty different menus to find what you're looking for, you can eventually stumble upon it without further guidance.
There is a value to this approach. It is valuable for one-time tasks. If I perform a task once in order to get my computer set up, and then I never perform that task again until three years later, it's probably not worth any memorization in order to perform more efficiently. It's more important that I'm able to figure out through menus or wizards how to do the task. For example, to set the time (which, presumably doesn't change unless you're changing time zones, and which usually adjusts for daylight savings time automatically), one should be able to point and click. If one has to memorize some cryptic commands in order to set the clock, that's not user-friendly.
The reverse is also true, though--repetitive tasks should not be guided but have the most efficient means available for performing them. Well, actually, ideally, you would have both available. I'm a bit peeved because I do mail merges a lot at work, and we just updated to a newer version of Microsoft Office that forces you to go through the mail merge wizard every time you want set up a mail merge. Yes, that's all good and well if you do one mail merge every two years, but if you do one every week, you don't want to have to answer all those stupid questions every time and then click next to be asked more stupid questions. In the ideal world, every task should be able to be done in all three of these fashions:
1. A simplified graphical environment--point-and-click a few options
2. A wizard that assumes you're an idiot and need to be guided every step of the way
3. A command-line interface that allows you to do a lot of tasks with one or two commands
The problem is that certain environments or software programs offer only #1 and not #3 or only #2 and not #1 or only #3 and not #1 or #2.
Without all three available, though, I'd say it's far more "user-friendly" for me to learn one command to resize a bunch of .jpg files to be 48x48 pixels instead of not learning any commands and then opening every single file in an image editor and then pointing and clicking my way to getting them all to be thumbnail size. Is it more "user-friendly" to learn nothing and spend hours accomplishing a task, or to spend five minutes learning one thing that helps you accomplish that same task in two seconds?
Along with user-friendly, of course, is easy. People tend to use them interchangeably, but generally easy denotes the ability to accomplish something quickly. No one spends hours pointing and clicking only to declare afterwards, "That was easy!" If they point and click for five seconds and accomplish a lot, they will, of course, think the task easy. I've found this generally to be the case, regardless of whether it's point-and-click or command-line driven--any task that seems rather hefty but can be accomplished in a short period of time is considered "easy."
In the end, though, user-friendliness, intuitiveness, ease of use--they all boil down to an assessment of balance: Is it worth my time to learn and memorize how to do this more efficiently or not? That's the bottom line. Yes, it takes time to learn a new desktop environment, a new operating system, a new set of programs. Yes, it takes time to learn commands you type in a terminal. The question you have to ask yourself is "Is it worth it?"
I can't speak for anyone else, but I've found it more than worth it. Learning more about Ubuntu has actually helped me learn more about Windows at work. Learning keyboard shortcuts and commands has saved me time at my work at home and my work at the office. Yes, I was pissed that I needed to resort to the command-line to change the time in Xubuntu (since that's not a task I do often enough to want to learn how to do), but I would much rather apt-get install twenty programs than try to install twenty programs in Synaptic or download twenty setup.exe files and walk through twenty wizards.
I've made my assessment. Have you made yours?
intuitive
The concept of intuitiveness does exist. For example, in Windows, as has been pointed out by many critics, it is not intuitive that one would click on the Start menu in order to shut down. Nor is it, in the drag-and-drop world of Mac OS X, intuitive that dragging an icon from the Dock to the desktop would delete it (it goes up in a puff of smoke). And, to be fair, it is not intuitive in Ubuntu that the /usr directory (supposedly short for user) would be off-limits to users and modifiable only by root.
But, as can be seen in the examples above, intuitiveness is overrated. After all, I see Windows users every day clicking on the Start button in order to shut down. They don't give it a second thought. My wife no longer tries to drag icons off her Dock to the Trash in order to remove them--she knows, almost instinctively it seems, to drag them to the desktop so they go up in a puff of smoke. Likewise, I always use sudo when copying files to or modifying files in the /usr directory in Ubuntu. I don't try to modify them as user and then smack my forehead, "Doh! I forgot that was owned by root. It tricked me again with the name /usr!" I think you'll find people quickly get fairly used to certain interfaces' behaviors, no matter how "counterintuitive" those interfaces are.
Of course, all the examples I gave are, in some sense, empirically counterintuitive--Start is a positive action and Shut Down a negative one; in a metaphor for the physical pushing and pulling, dragging and dropping indicates moving, not deleting, unless the object is being dragged to the trash; the word usr looks like user, but it can't be used by what it looks like. There are other counterintuitive behaviors that are not empirically counterintuitive but contextually counterintuitive--in other words, they run contrary to what the user is used to, whether from other parts of the interface or from a previous interface.
For example, it's not any more empirically intuitive to download a setup.exe from the internet and then double-click it and click a bunch of next buttons and the finish than it is to click on Applications > Add/Remove Applications and then click to install the applications you want and have them downloaded off the internet behind-the-scenes and then installed. But for someone who is used to installing software the former way, the latter way will seem counterintuitive because she has a certain workflow in mind (open web browser, search for software, download software, install software, delete installer) instead of a different workflow (open software manager, search for software, mark it to be downloaded and installed).
Bottom line: what you're used to always trumps intuitiveness or counterintuitiveness.
user-friendly
When people say user-friendly, they usually mean discoverable and able to be done with a mouse, point-and-click style. Even if you must go through twenty different menus to find what you're looking for, you can eventually stumble upon it without further guidance.
There is a value to this approach. It is valuable for one-time tasks. If I perform a task once in order to get my computer set up, and then I never perform that task again until three years later, it's probably not worth any memorization in order to perform more efficiently. It's more important that I'm able to figure out through menus or wizards how to do the task. For example, to set the time (which, presumably doesn't change unless you're changing time zones, and which usually adjusts for daylight savings time automatically), one should be able to point and click. If one has to memorize some cryptic commands in order to set the clock, that's not user-friendly.
The reverse is also true, though--repetitive tasks should not be guided but have the most efficient means available for performing them. Well, actually, ideally, you would have both available. I'm a bit peeved because I do mail merges a lot at work, and we just updated to a newer version of Microsoft Office that forces you to go through the mail merge wizard every time you want set up a mail merge. Yes, that's all good and well if you do one mail merge every two years, but if you do one every week, you don't want to have to answer all those stupid questions every time and then click next to be asked more stupid questions. In the ideal world, every task should be able to be done in all three of these fashions:
1. A simplified graphical environment--point-and-click a few options
2. A wizard that assumes you're an idiot and need to be guided every step of the way
3. A command-line interface that allows you to do a lot of tasks with one or two commands
The problem is that certain environments or software programs offer only #1 and not #3 or only #2 and not #1 or only #3 and not #1 or #2.
Without all three available, though, I'd say it's far more "user-friendly" for me to learn one command to resize a bunch of .jpg files to be 48x48 pixels instead of not learning any commands and then opening every single file in an image editor and then pointing and clicking my way to getting them all to be thumbnail size. Is it more "user-friendly" to learn nothing and spend hours accomplishing a task, or to spend five minutes learning one thing that helps you accomplish that same task in two seconds?
Along with user-friendly, of course, is easy. People tend to use them interchangeably, but generally easy denotes the ability to accomplish something quickly. No one spends hours pointing and clicking only to declare afterwards, "That was easy!" If they point and click for five seconds and accomplish a lot, they will, of course, think the task easy. I've found this generally to be the case, regardless of whether it's point-and-click or command-line driven--any task that seems rather hefty but can be accomplished in a short period of time is considered "easy."
In the end, though, user-friendliness, intuitiveness, ease of use--they all boil down to an assessment of balance: Is it worth my time to learn and memorize how to do this more efficiently or not? That's the bottom line. Yes, it takes time to learn a new desktop environment, a new operating system, a new set of programs. Yes, it takes time to learn commands you type in a terminal. The question you have to ask yourself is "Is it worth it?"
I can't speak for anyone else, but I've found it more than worth it. Learning more about Ubuntu has actually helped me learn more about Windows at work. Learning keyboard shortcuts and commands has saved me time at my work at home and my work at the office. Yes, I was pissed that I needed to resort to the command-line to change the time in Xubuntu (since that's not a task I do often enough to want to learn how to do), but I would much rather apt-get install twenty programs than try to install twenty programs in Synaptic or download twenty setup.exe files and walk through twenty wizards.
I've made my assessment. Have you made yours?