Re: The path of least blame

[António Fernandes <antoniojpfernandes gmail com>]

No dia 4 de Agosto de 2011 23:04, Akshay Dua <dakshay gmail com> escreveu:

Your thesis bases itself in the assumption that "the expected result" of "clicking the terminal icon" is "start a new instance", but you don't provide evidence to found that assumption.

If you read my assumptions again, you'll see that I only assume that clicking the terminal icon starts a new instance. Not that it is the "expected result".

Ok. Then, instead of an unfounded universal proposition, your assumption is a particular proposition. That was just your assumption, but it was wrong. That's ok; there is no one to blame.

The Terminal application icon represents the Terminal application itseft. It does not represent a "New window" command [1]. The application icon might be used to launch a new instance of that application (as is the case with Gnome Panel launchers), but it does not always necessary, because there is no compulsory association between the Terminal application icon and the "New window" command.

Therefore, if the user clicks on the Terminal application icon, he/she can only expect to get the Terminal application. That expectation can be fullfilled either by raising an existing window or creating a new one. Both are valid ways to give user what he asked for; so, there is no reason for the user to specifically expect one of them. You can only expect to get a terminal. Everything else is bias caused by being used to a specific user experience.

 

Maybe it's users' expectation that is wrong

This is why users get angry. How can an expectation be wrong? It is always right. It is only that expectations are different. For some, they expect a new instance should be started, for others they expect to be taken to the existing instance. My point is that if software does not have enough information to perform as expected, it should choose the option that does not allow users to blame it.

Even if both expectations were equally valid, that would be the wrong criterion. Software exists to help users, not to save itself from blame. 

As far as I know, "Do not get blamed" is not a design principle of Gnome Shell. "Take responsibility for the user's experience" and "Be considerate and forgiving" are [2].

If, in its mission to help users keep track of their activities, Gnome Shell is unfairly blamed for it by those who are strongly used to a different workflow, that's not a reson to give up on its principles.

 

 From the user perspective, it is one and the same: it takes me to the application

But, the point is: its not the same.

The path is not the same, but always lead to the same result.

Take this example: When you ask for a cheeseburger in a fast food restaurant, there are 2 different possibilities from the employee's point of view: 1) take a cheeseburger if one is available; 2) ask the cooks to make one, then take it when it is ready. However, from the client's point of view, both meet the client's request. If the client does care about it and wants a hot cheeseburger, he/she may make a special request (I usualy ask for a cheeseburger without ketchup if I want a hot cheeseburger).

The same happens with Gnome Shell: the user requests a Terminal, so a Terminal window is focused; if there is no Terminal window available, a new one is created. If the user does care about it and wants a fresh new window, he/she may make a special request:

- CTRL+Left Click for new window in current workspace

- Middle click for new window in new workspace

- Right click for menu, then left click for "New window"

- Left click to focus, then "File"->"New terminal"

- Left drag and drop for new window on target workspace

- etc.

Tons of options available for special request. One, simple, direct left click for default request. Both in one single icon to interact with the application.

This is not about wrong or right. Its about making users less angry with software and technology in general.

Footnote: Users might want the feature to raise existing windows, but they will be understanding of the fact that Gnome cannot possibly know if they wanted to raise the existing window or create a new one.

The same can be said to users that want the proposed behaviour.

Sure, but I think if you've missed my point. In the new instance scenario users blame themselves or are happy with the outcome, in the "take me to the application scenario" many get angry. Which one will you choose?

The second option. Get users to blame themselves conflicts with design principles. Users getting angry at change is human nature, not a Gnome Shell bug.

[1] This icon does: http://upload.wikimedia.org/wikipedia/commons/b/b2/Window-new.svg

[2] https://live.gnome.org/GnomeShell/Design/Principles