On Sat, Sep 17, 2011 at 06:00:35PM -0400, Benjamin Johnson wrote: On Sat, 17 Sep 2011 17:08:52 -0400, Roger <rogerx.oss at gmail.com> wrote:
However, how would you like a prompt on your vehicle every time you pressed the brakes? Do you really want to stop? Or, how about every time you start your vehicle and you're within an emergency situation and get, "Do you really want to start your car?" With real world applications, security of preventing accidental quitting is application response to users intentions.
I don't know much about cars, but I do know it's pretty hard to accidentally start one. And computers are, in terms of physical interactions, much more difficult to use than most real-world objects (and less predictable, as well). That's why usability is so tricky to get right, and why most programs' interfaces tend to follow established standards and conventions.
By the way, it's worth noting that vim prompts you to prevent accidental quitting, BSD vi prompts you to prevent accidental quitting, even freaking *ed* prompts you to prevent accidental quitting. I find it pretty annoying sometimes, too, but there have been many times when that prompt has saved me from far greater frustration.
You answered this one yourself. The default option for VIM on quit, is to only ask if it should quit when the file in the buffer has been modified. Other then this, if the file buffer is unmodified, no prompt before quitting is issued. As such, should a browser do the same; only when forms or HTML composer/editor has a current editing session, or associated email composition has been edited, should a browser ask when to quit? The only thing likely editable within Dillo are bookmarks and some forms. When you think about it, all data being viewed within the browser is static to it's original location. The only information lost is the original entered URL, which usually can be found via Google, hence, no information is at threat of being lost. -- Roger http://rogerx.freeshell.org/