On Thu 14 Aug 2008, Jamie Zawinski wrote: >> Hmm, it does; but the delay is just long enough to be irritating, but >> just short enough that it almoast takes me longer to reach for the >> mouse >> and click. If the message disappears just before I click, I might end >> up >> closing a window or so if I'm unlucky enough to click the X... > > Well, I can make the delay shorter, but then I'm not sure it'll be up > long enough to notice/read...
As I suggested, make it configurable. >> The timeout is implemented within the xscreensaver code, right? >> So surely xscreensaver "knows" whether there's a timeout. Hence it can >> act appropriately in such a case. I'd have thought that the PAM stuff >> would only be accessed after the user hits enter, so PAM shouldn't be >> involved unless it was a real login attempt. At least, that's how I >> would have programmed it... > > Unfortunately, PAM doesn't work that way. To authenticate, you drop [snip PAM desc] Hmm, you're right, PAM sucks :) >> Anyways, I'm sure the warning message is put there by xscreensaver. > > Sure. But if we assume for the moment that the message does have merit > for the case where someone really did type the wrong password, there's no > way to tell the difference between that and "didn't try". I remember now that the message used to be displayed inside the password prompt window in earlier versions. That was a *lot* better than the current behaviour. And what's wrong with letting someone (who knows how to read the manpage) configure that window away? > Also, those incorrect attempts are logged in syslog too. That doens't interfere directly with me interactively, so I don't care about that. Paul -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
