I am also keeping an eye on an http log the corner.Now, if these windows were tiled, I'd have a hell of a time trying to see the paper I was working on, or the web info, or the log. I have an Appleworks window open, as well as several browser windows pointing to relevant information. I second gnosis.For example: I am writing a long research paper for my university-required writing class.
Am I correct in my understanding? If so, would it be better for Apple to use such a method in OS X, given the power of Quartz? Or would this simply be too much of a hassle given that we users (and programmers) are largely creatures of habit? Thanks for your input.Greg The reverse occurs, when a window is closed.This kind of windowing scheme seem to me to be a very elegant way of managing windows, and seems to be not only more efficient with respect to memory usage, but also in the terms of maximizing the amount of real estate used. When the maximum number of windows per track is used, an new vertical track is created, and all of the previous track a shrunken in size accordingly. When the user opens an application, its window fills up the remaining space in the current vertical track. When no applications are open, there is initially just one vertical track. Each vertical track is then subdivided into a set of horizontal viewers (his term for a window). If this is true, this seems to cause a further degradation of performance simply because of inherent algorithm complexity.In his tiling scheme (he refers to it as hierarchic tiling) the screen is divided into a series of vertical tracks. The algorithms involved in keeping track of the overlapping layers of windows and restoring the data are significantly more complex than the algorithms involved with a tiling scheme. It seems like this could be avoided by using a tiled window approach rather than the overlapping windows prevalent in the desktop metaphor.Another issue is raised by Niklaus Wirth in his book Project Oberon. If this is true, it seems like the issues concerning the double buffering would be compounded because of the window managers needing to keep track of a theoretically a infinite series of overlapping windows, and maintain a great deal of information as a result. In one of the recent discussions concerning the performance of Quartz, many people lay the 'blame' on the fact that the windows are double buffered (I could be wrong but this what I gathered from the discussion). Here is a question to all of the experts in UI design: Which is better tiled windows, or overlapping windows?I am asking for several reasons.
0 kommentar(er)
