Freshest comment

Awesome, a very buitifil parody of the "Think Different" poem.

Subscribe to updates

Syndicate content

Understanding Macs and PCs: Closing Apps

Mac OS X close buttonOne oft-misunderstood difference between Mac OS X and Windows behavior is the way users close applications. Many people misunderstand the specifics for one or both of the operating systems, while other aren't aware that there is a difference. 

Case in point: Even tech reviewer extraordinaire Walter Mossberg goes a wee astray in Some General Tips for Switch to Mac From Windows, an article I expand upon to correct Mac vs Windows inaccuracies. With additional input from smart readers, some good stuff gets scattered throughout that page's comments. Here I'll summarize what you need to know about closing applications the Mac way and the Windows way:

Windows close buttonClosing a window in Windows

Here's what Mr Mossberg wrote. It's a common take, I believe:

Quitting programs: In Windows, you can quit a program by clicking on the red “X” in a square at the upper right corner of the window you’re using.

Close, but not quite. Clicking the red "X" in the upper right corner of a window in Windows doesn't necessarily quit a program. It only closes the window, the same as in Mac OS X.

For example, if you have five Internet Explorer windows open, clicking on the red "X" in the upper right of a window will only close that window. Internet Explorer will still be running, now with four windows open. 

However, here's the catch: If that window was the program's last (or only) window, then Windows does quit the program entirely, as described above. So if you click the red "X" on your last Internet Explorer window, you do shut down the program.

Closing a window in Mac OS X

What's different on a Mac? Here's Mr Mossberg again:

But on the Mac, if you click on the equivalent button — a red “X” in a circle in the upper left corner — you are merely closing the window, not quitting the program. To quit the program, you must either select Quit from the leftmost menu or press the Command and “Q” keys together.  

Again, that's only true sometimes. As Mr Mossberg describes, if the Mac OS X program has multiple windows, or can have multiple windows, then closing a window only closes the window, and nothing more – even if it's the last window! The Mac program remains active, as is indicated by the "active application" blue dot under its icon in the Dock, and its Menu bar at the top of the screen. So, despite no open windows, you can still use the program's Menu bar to perform usual functions, such as calling up a new window, opening preferences, or finally shutting down the program via the "Quit" menu command.

The above is certainly a difference from Windows' behavior. However, here's the kicker that Mr Mossberg missed:

If a Mac program has only one window and can have only one window, then closing the window will typically close the application as well. Just like in Windows.

Is the Mac way weird?

So, closing the last open window on a Mac OS X application sometimes shuts down the application, and sometimes doesn't. Isn't this terribly inconsistent?

No, there's a logic behind the Mac's ways – though many experienced Mac users themselves don't know it. It's this:

If a program can have multiple windows, as a web browser or word processor can, then closing the last window is not necessarily a sign that you want to shut down the program. You might very well want to simply close the single open word processor document, and then start a new one, for example. Mac OS X makes no assumptions about what you want here, and lets you actively shut down the program when you're ready.

Think of it this way: if a program can have any number of windows open, then zero windows is just another valid possibility. The Menu bar – sort of an "honorary window" itself – remains available for giving new commands to the program.

But if a program can have only one window, like Mac OS X's Calculator or System Preferences programs, then in all likelihood, you're closing the window because you're just plain done with the program. True, you could think of contrived exceptions to that, but it's logical for Mac OS X to assume that by closing a one-window application, you're indicating that you're finished using it.

So we've got the the basic rule:

  • If a Mac OS X application can run with one or multiple windows open, then it'll also happily run with zero windows open, awaiting your further commands via Menu bar or Dock. You need to give it a "Quit" command to shut it down.
  • If a Mac OS X application only offers one window, then closing that window suggests to the program that you're done. It shuts itself down to save you the trouble of sending a "Quit" command. 

That's it.

Or for Apple's own words to its developers:

In most cases, applications that are not document-based should quit when the main window is closed. For Example, System Preferences quits if the user closes the window. If an application continues to perform some function when the main window is closed, however, it may be appropriate to leave it running when the main window is closed. For example, iTunes continues to play when the user closes the main window. 

Document-based vs non-document based. I'll continue to use my explanation of "single window" vs "multiple window", as it's helped me stay on top of things, and has worked as an explanation to newbies. But above we have the official word from Apple. 

Human logic

Why do Windows and Mac OS X handle this matter differently? One reason: each system's choice of app-closing behavior is affected by other aspects of the interface. For example, users of Mac OS X can easily work with an application with all windows closed: as mentioned earlier, the Menu bar itself is effectively a final, special window that remains available until the application is closed. The application's Dock icon provides yet another interface element that hangs around to allow control of the windowless program.

Windows, on the other hand, couldn't easily offer users the choice of running windowless programs, even if it wanted to. Menus are embedded within regular windows, so when the last window is closed, there's no Menu bar or other interface element remaining to let users control the application. There's no logical choice but to shut down the application.

Yet whatever the interface and system details, it'd be misleading to view the above behaviors as purely technical differences. In the end, they're human decisions – intentional choices made by the designers of each OS's interfaces, in expectation of how users will likely want to interact with programs. With some interface revamping, Windows designers could find a way to allow interaction with a program even when all its windows are closed; for example, the running application could leave some element in the Task Bar, allowing access to a new window or a menu. But the designers presumably decided instead that the current behavior is better. 

It's enlightening to look at computer behavior from the standpoint of what human logic lies behind everything. Take Mac OS X apps like Mail and Safari. As applications that allow multiple open windows, they keep running (as you should now expect) with all windows closed. But they also stand hair-trigger ready to toss up new windows: just click on either app's Dock icon, and it'll give you a fresh window to work with. Even though you didn't ask for a new window via the Menu bar.

Why do they do this? It's a matter of human logic: the programmers thought through the scenarios, and decided that someone who goes out of the way to call upon Mail or Safari in the Dock most likely wants to interact with the program in the usual way – that is, wants to see a Mail Viewer window or a Safari browser window.

Again, it's possible to think of contrived exceptions. Like this: "No, I don't want a browser window, I want to close all Safari windows, then go to another app, then use the Dock to go back to Safari, and then open its Preferences, all without seeing another browser window!" If that's your plan, then it'd be nice if Safari would indeed keep its windows closed until you specifically tell it, via the Menu bar, to create one again.

But the programmers apparently decided that such an oddball wish has to take a backseat to the newbie who's closed all his Safari windows and, not hep to using the Menu bar to create a new one, forlornly clicks on the Dock thinking, "What happened to my Safari?" It's these real-world, likely scenarios that bring many of the quirks we know and love (?) to our software – at least, to good software by developers that actually think through such issues.

Human illogic

Of course, human logic means human inconsistencies, too. The above rule for whether closing a final window closes a Mac OS X app explains a lot of behavior, but doesn't make everything crystal clear. First, it's by no means always obvious to users whether a program is a one-window application or not. For example, System Profiler looks like a classic one-window application – but close that window, and things stay running. A look at its Menus explains why: System Profiler is secretly a multi-window app. That's unexpected; who knew?

Other applications will have one main window, but allow some smaller panels as well; do those count as windows or not for the purpose at hand? The Activity Monitor app is a good example: its main window stands unique and uncloned, yet closing it leaves the program running. Why? Well, inspection of its Menus shows that it's able to spawn lesser-known windows like CPU History; further, with or without any windows visible, it can also display graphs in the Dock. Are those the reasons why it stays running? In light of the earlier explanation by Apple to developers, that would make sense.

Still, there are yet other apps that misbehave for reasons unknown. Take that rebel Disk Utility: it's a multi-window app (surprise!), yet closing its last window does shut down the program. Why? Is this yet some other twist to the Mac OS X rule that needs to be noted? Personally, I think the explanation is simpler: its programmers made a boo-boo. Oops!

Think like a human!

I often hear newbies told that they have to "learn to think like the machine" to master computing. That is so wrong. The key to mastering a computer is thinking like a human – specifically, thinking like the human programmers who designed software behavior in response to expected human user behavior (and let in a dash of the programmers' human failings, too). That's the way to grasp the difference in Mac OS X vs Windows behavior in closing applications – and the way to computer mastery overall!

Resources

Tech heavyweight Ars Technica just weighed in with a detailed look at Mac OS X's Dock vs Windows 7's Taskbar, a topic that touches heavily on the two OSes' respective approaches to closing applications. It's a nicely detailed look at Mac OS X's emphasis on windows as representing documents, and Windows' emphasis on windows as representing applications, which leads to interesting consequences (such as the Dock being a better application switcher, and the Taskbar being a better windows switcher).

If you've got an interest in interface design, or just want to use your Macs and Windows machines more productively, the article is a great read. 

 

Average: 1 (2 votes)

Post new comment

  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • You may quote other posts using [quote] tags.
  • Lines and paragraphs break automatically.
  • Web page addresses and e-mail addresses turn into links automatically.

More information about formatting options


Things people say

One of the greatest myths about Macs today is that they cost way more than their PC equivalents, when a direct spec-to-spec comparison between the two often proves that is not the case. But the myth persists because that was exactly the situation in the 1990s when Apple churned out a succession of indifferent computers that costs hundreds, if not thousands, more than their competitors.