James Sparenberg wrote: > > The average person does not press a Save button after turning the volume > > knob on his stereo. If he'd be wondering whether his computer settings are > > going to be saved or not, then he has been mentally mutilated by > > incompetent software designers, and should be cured ASAP. ;-) > > You are correct but in the case of a volume button the user has > immediate feedback as to whether or not his input has been received (the > volume went down or up.) I think the question is not whether the button > should say save close or ralph. But rather feedback as to the affect > that the users actions have had is needed. Remember the days of typing > rpm -Uvh on an rpm and then wondering ... did it install? Where did it > install? How do I launch it. From this perspective Buchan's point is > valid... Unless of course you prefer that you box leave you in the dark > about what it just did. > > James
I'm sorry, but I think that this and Buchan's point (reiterated in his previouw message) miss the intent of the "close" button in this context. The user is checking checkboxes to activate or deactivate sources; that's it. There is no destructive behavior, nor confirmation needed. Should the user wish to reverse his decision, he has only to select/deselect a checkbox. The checking action is immdiate (or at least preceived to be, which is the important thing), so an "abandon" action is confusing. This is not a case of an install; the user is simply choosing which items in a list are activated. Nothing outside of the context of this list is affected. Please refer to the beginning of the thread (or what I pasted below). The original poster was confused why a "quit and don't save" button didn't recover a source that had been deleted. This is because the "remove" button takes immdiate effect; as stated by Guillaume, there is no way to back this operation up. Therefore, a single "close" button makes sense in this context. Buchan's "fire" button analogy (in a message that preceded this one) is correct -- but *not* in relation to the "close" button. The point is, by the time the user closes the window, the damage is already done. Therefore, you *cannot* give the impression that the action is undoable. Guillaume stated that making things undoable is too difficult (at least for now?). I proposed the close button to remove the ambiguity, not to address the base issue which is that there is no undo available. For everyone's edification, I present the original post here: jokerman64 wrote: > nothing serious just a few annoyances that you may want to take a look at. > First off in the software sources manager, whenever you remove a source and > then quit (not save & quit), your changes are saved anyway. This is a real > pain in the @$$ if say, you accidentally deleted the entry for CD1 or > something else important. It should cancel all changes made if Quit is > selected and save all changes if Save & Quit is selected. Otherwise the Save > & Quit button is both useless and redundant. It also gives a false sense of > security and is damned misleading.
