On Wednesday 16 February 2005 12:56 pm, Maurizio Colucci wrote: > Of those 50%, how many people do you believe would press CTRL+F without > first making the window fully visible again, KNOWING THEIR SEARCH > RESULTS COULD VERY WELL BE IN THE INVISIBLE PART? The search results in a browser window would almost never be in this part, because the browser scrolls the page so that the results are close to the top. And if you have never used this 'search-bar-at-the-bottom' concept before, and you expect CTRL+F to bring up a search dialog like almost every other app on the desktop, why would you be worried about moving the window? > Of those 5%, how many people, AFTER hitting CTRL+F and not seeing > anything, would not think "maybe something happened in the invisible > part of the window? ". Let's say 20%. > > Multiplying the fractions, we have (50*5*20) / (100*100*100) people who > would be hurt by a search toolbar placed below. That's 0.5%. Hardly > something we should worry about. This is not to say I have something > against a search bar on top. To me, the point isn't that 'this or that fraction' would be inconvienced by having the bar on the bottom, it should be the oppposite - what fraction would be inconvienced by it being on the *top* ? I would think none - the top is where it should be. There was one point made above about the page flickering, but if the bar was designed properly (so that it came down 'above' the page relative to the z-axis, rather than pushing it down), then this is a non-starter. If the situation arose where the page had no margins and the found text was actually below the bar, you could*then* move the page down below the widget. This scrolling action would be no different than the current auto-scrolling that happens while you search in a browser. -- If you wait by the river long enough, eventually you will see the bodies of all your enemies float by. - Sun Tzu _______________________________________________ kde-usability mailing list kde-usability@kde.org https://mail.kde.org/mailman/listinfo/kde-usability