tag:blogger.com,1999:blog-7729093380675162051.post7388490619540673285..comments2024-03-28T13:56:47.604+11:00Comments on moyhu: Trends of gridded BEST and GISS shown with WebGLNick Stokeshttp://www.blogger.com/profile/06377413236983002873noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-7729093380675162051.post-64123891790404443092014-07-20T14:14:30.154+10:002014-07-20T14:14:30.154+10:00Thanks, Carrick. On common color scale, I rely on ...Thanks, Carrick. On common color scale, I rely on automatic scale setting (locally, in JS). The range varies a lot with length of time period. To make them comparable, I'd have to preset. But the data I store is currently cumulative sums of yx and y, and it only gets converted into trends at the user end. So it's awkward.<br /><br />The simplest way to do side by side is with two browser windows. I use absolute picture setting, so you can contract the window to leave just the picture, and resize with Ctrl- etc.<br /><br />On pos and neg, I have sometimes made 0 a special color. It's too messy for general use, but I could offer it as an option. In fact, I guess I could offer complete user control of color scheme. That might fix everything, for sophisticated users.<br />Nick Stokeshttps://www.blogger.com/profile/06377413236983002873noreply@blogger.comtag:blogger.com,1999:blog-7729093380675162051.post-9231951181660542712014-07-20T13:52:40.436+10:002014-07-20T13:52:40.436+10:00Nick, thanks for another great visualization tool....Nick, thanks for another great visualization tool. The only things I wish were different were <br /><br />1) a common scale for the two series.<br />2) A visual way to distinguish positive from negative trends.<br />3) having a way to compare them together <a href="https://dl.dropboxusercontent.com/u/4520911/Climate/Trends/Nick-Best-GISS-1924-2008.jpg" rel="nofollow">like this</a> would be ultra cool.Carrickhttps://www.blogger.com/profile/03476050886656768837noreply@blogger.com