Pro user interface and one mouse button
I apologize in advance to my dear readers for the difficult style of exposition. Very much here is attributed to a particular theory of interface design.
What I am discussing here is my reflection on Jeff Raskin’s idea that a computer mouse should have one button. “Appendix A provides a rationale for why a mouse should have only one button. "(Ruskin, p.27)
We must assume that the idea to create a tablet-type computer comes precisely from these positions. After all, in fact, it has one button and gestures.
')
Let's absolutize this idea to the level of any device. We have a conventional computer that has a keyboard, a screen, and a pointer (such as a mouse or a pen).
Also, we presume that the user should not enter the mode. That is, to create an additional locus of attention on the performed action, on the object (conditional second layer). An example is when you click on a file, and then enter the menu through the right mouse button.
From here you can make the statement that when clicking on something, the user should only have an object-action behavior variant (Raskin calls this a “noun-verb”). What does it mean? For example, I select the text and drag. Here I perform an action on an object. The object is text. Action - to drag.
Let's make one more absolutization. That the device has only one program. We do this in order to show logic. This means that in other programs - applications, there will be a similar logic.
What is the context menu? This is the mode. That is, by clicking on the right mouse button, I enter the mode of selecting actions on an object. Not the action itself is performed, but the mode of action. It now seems obvious simply because you know about its existence. Someone told you about this opportunity. And if not? And if I do not know about this opportunity? This is a cultural pattern, if you want.
I am now writing this text in the Timlab Document Document editor. When I began to study it only, I did not know that by pressing the right button, something would change. But I guessed it. I was taught this.
I still can not believe it. Try to take your grandmother, let her over 60 and the computer she had not seen before. And now try to explain to her that in order to move a file you need to press the right button, click cut, etc.
What is the thing? It is that a person needs to see the actions performed. Do not hide them from him.
My father is 57, he is a journalist. For more than two years using Linux. When he goes online, he uses only those functions that he sees. He may not remember that a thousand steps must be taken to set up the system. He has me for it. But he knows that if he clicks on this huge pictograph of Yandex in Chromium, he will fall on Yandex. The highlight and commentary partly encourage him to commit these actions.
Here we get the same picture: object -> action.
Let's imagine the same situation if we have one pointer button (mouse, stylus). Everything is exactly the same, in this case we do not use the right button at all.
How can you implement the same function that gives the right button?
We are missing an important point here. The user selects the file in order to do something with it. Not just like that. So this is what the user now needs only to suggest something to do with the file.
This means that possible actions on the file should be displayed immediately by selecting the file with a single click of the pointer. In our case, the menu is displayed above or next to the file. This is a hidden interface. What is the difference from a right-click? The fact that here I click - and the menu is displayed regardless of whether I want it or not. With the right button you need to know about this feature. Further, it can turn into a stimulus - a reaction, like with Pavlov's dog. We do not give the user a thousand options. We give one that will work. We reduce the amount of code. We reduce the number of potential errors.
Leave a double click to open the file, as is customary now. This is also a pattern of behavior, it also works as an incentive - a reaction.
And if you select 32 files? And all in parts? Now do not forget about the keyboard. Yes, this action is harder. He needs to learn. It belongs to the category of shortcuts, since the same can be done individually with each file. However, the menu is displayed in the same way after the selection is completed. In order to give the user time to make an additional choice (for example, to drag these files into a folder), we will tie the opening of the menu to release the Shift or Ctrl button. This is again a reference to the fact that people do not just select files.
Why this rake? After all, it is easier to right-click on a file and ... lose the selection by making a mistake of 5 pixels or relying on the definition area of ​​the icon (God, what the user should think about. Poor man ....) It's not nice, right?
Let's take a look and count.
Suppose that one such allocation, loss, recovery takes 20 seconds. (Something smart)
Let the number of Internet users in Russia is about 60 million. (The figure is approximate and rounded off from statistics). Each of them in one way or another uses a computer.
We consider:
60,000,000 x 20 seconds = 1,200,000,000 seconds.
1,200,000,000 seconds / 60 seconds per minute = 20,000,000 minutes.
20 000 000 minutes / 60 minutes in an hour = 333333,333 ... hours.
333333,333 ... hours / 24 hours in days = 13888,888888889 days.
13888,888888889 / 365 days in a year = 38.051750381 years.
In other words, only in a single operation, the 60-millionth army of users loses about 2/3 of the already short life of the average Russian. I make five such mistakes a month. Now calculate how many human lives we live on only one such error. In one month…
It is clear that once such an error does not occur for everyone.
So what is the essence of a one-button mouse? The point is to exclude errors from the possible rank and provide the user with only one solution to the problem. The task of the virtual interface is to shorten the path before committing the action. We do not do thousands of interfaces to achieve the same result.
Citizens-developers, do not complicate the life of the user. He has a boss for that.
The literature that served as the basis for the thoughts expressed in this post:
Also grateful to all those who wrote texts about usability and interface design on Habré, as well as many others, whose words, actions or texts pushed me to reflections on this topic.
What I am discussing here is my reflection on Jeff Raskin’s idea that a computer mouse should have one button. “Appendix A provides a rationale for why a mouse should have only one button. "(Ruskin, p.27)
We must assume that the idea to create a tablet-type computer comes precisely from these positions. After all, in fact, it has one button and gestures.
')
Let's absolutize this idea to the level of any device. We have a conventional computer that has a keyboard, a screen, and a pointer (such as a mouse or a pen).
Also, we presume that the user should not enter the mode. That is, to create an additional locus of attention on the performed action, on the object (conditional second layer). An example is when you click on a file, and then enter the menu through the right mouse button.
From here you can make the statement that when clicking on something, the user should only have an object-action behavior variant (Raskin calls this a “noun-verb”). What does it mean? For example, I select the text and drag. Here I perform an action on an object. The object is text. Action - to drag.
Let's make one more absolutization. That the device has only one program. We do this in order to show logic. This means that in other programs - applications, there will be a similar logic.
What is the context menu? This is the mode. That is, by clicking on the right mouse button, I enter the mode of selecting actions on an object. Not the action itself is performed, but the mode of action. It now seems obvious simply because you know about its existence. Someone told you about this opportunity. And if not? And if I do not know about this opportunity? This is a cultural pattern, if you want.
I am now writing this text in the Timlab Document Document editor. When I began to study it only, I did not know that by pressing the right button, something would change. But I guessed it. I was taught this.
I still can not believe it. Try to take your grandmother, let her over 60 and the computer she had not seen before. And now try to explain to her that in order to move a file you need to press the right button, click cut, etc.
What is the thing? It is that a person needs to see the actions performed. Do not hide them from him.
My father is 57, he is a journalist. For more than two years using Linux. When he goes online, he uses only those functions that he sees. He may not remember that a thousand steps must be taken to set up the system. He has me for it. But he knows that if he clicks on this huge pictograph of Yandex in Chromium, he will fall on Yandex. The highlight and commentary partly encourage him to commit these actions.
Here we get the same picture: object -> action.
Let's imagine the same situation if we have one pointer button (mouse, stylus). Everything is exactly the same, in this case we do not use the right button at all.
How can you implement the same function that gives the right button?
We are missing an important point here. The user selects the file in order to do something with it. Not just like that. So this is what the user now needs only to suggest something to do with the file.
This means that possible actions on the file should be displayed immediately by selecting the file with a single click of the pointer. In our case, the menu is displayed above or next to the file. This is a hidden interface. What is the difference from a right-click? The fact that here I click - and the menu is displayed regardless of whether I want it or not. With the right button you need to know about this feature. Further, it can turn into a stimulus - a reaction, like with Pavlov's dog. We do not give the user a thousand options. We give one that will work. We reduce the amount of code. We reduce the number of potential errors.
Leave a double click to open the file, as is customary now. This is also a pattern of behavior, it also works as an incentive - a reaction.
And if you select 32 files? And all in parts? Now do not forget about the keyboard. Yes, this action is harder. He needs to learn. It belongs to the category of shortcuts, since the same can be done individually with each file. However, the menu is displayed in the same way after the selection is completed. In order to give the user time to make an additional choice (for example, to drag these files into a folder), we will tie the opening of the menu to release the Shift or Ctrl button. This is again a reference to the fact that people do not just select files.
Why this rake? After all, it is easier to right-click on a file and ... lose the selection by making a mistake of 5 pixels or relying on the definition area of ​​the icon (God, what the user should think about. Poor man ....) It's not nice, right?
Let's take a look and count.
Suppose that one such allocation, loss, recovery takes 20 seconds. (Something smart)
Let the number of Internet users in Russia is about 60 million. (The figure is approximate and rounded off from statistics). Each of them in one way or another uses a computer.
We consider:
60,000,000 x 20 seconds = 1,200,000,000 seconds.
1,200,000,000 seconds / 60 seconds per minute = 20,000,000 minutes.
20 000 000 minutes / 60 minutes in an hour = 333333,333 ... hours.
333333,333 ... hours / 24 hours in days = 13888,888888889 days.
13888,888888889 / 365 days in a year = 38.051750381 years.
In other words, only in a single operation, the 60-millionth army of users loses about 2/3 of the already short life of the average Russian. I make five such mistakes a month. Now calculate how many human lives we live on only one such error. In one month…
It is clear that once such an error does not occur for everyone.
So what is the essence of a one-button mouse? The point is to exclude errors from the possible rank and provide the user with only one solution to the problem. The task of the virtual interface is to shorten the path before committing the action. We do not do thousands of interfaces to achieve the same result.
Citizens-developers, do not complicate the life of the user. He has a boss for that.
The literature that served as the basis for the thoughts expressed in this post:
- Jeff Raskin. Interface: new directions in the design of computer systems: Symbol-Plus; M .; 2005
- Donald Norman. Design of the usual things: Williams Publishing House; M .; 2006
Also grateful to all those who wrote texts about usability and interface design on Habré, as well as many others, whose words, actions or texts pushed me to reflections on this topic.
Source: https://habr.com/ru/post/158113/
All Articles