>>>>> "Martin" == Martin Vermeer <[EMAIL PROTECTED]> writes:
Martin> Here is a patch that makes the behaviour more managable. It is Martin> still based on page-at-a-time scrolling, but this time Martin> 1) the time between scrolls is a more humane 0.5 s, and 2) it Martin> is _guaranteed_ to be 0.5 s, because now only the 'heartbeat' Martin> synthetic mouse event can cause scrolls, not genuine mouse Martin> motion events. Why is 2) true? Martin> I am not happy with the arbitrary constants 20 and 40. Any Martin> better ideas? What is the 20 for? I guess this is related to my previous question. JMarc
