Keeping the position of the asset stable and ordered.Īlthough different algorithms put varying degrees of priority on one or the other of these goals, to the extent that they attempt to keep the position of an asset stable, they all (so far as I have seen) focus on positioning assets by size, not by color. Getting the aspect ratio of a particular asset’s rectangle to be as close to square as possible.Ģ. Fundamentally, there seems to be a tradeoff that all tiling algorithms contend with. An asset’s position is hard to control in a treemap, because it is determined by the layout algorithm used to position the assets within the overall containing rectangle of the chart. The main problem, I believe, is that as values in the underlying data change, any particular ‘asset,’ or rectangle, may move to a different position on the map, making it hard for a person to keep track of individual assets as they change over time. However it also has a few weaknesses that, I think, prevent it from being as useful as it might be. The basic treemap concept has many strengths. Treemaps, in any case, are now available as part of many commercial graphing toolsets and javascript charting libraries, further increasing access and popularity. I wish there were a different name for a treemap because a tree seems to suggest to most people something branching, more like:īut that is a tree diagram, not a treemap. Often people we talk to have seen treemaps before, but they just didn’t know that that’s what they are called. When we talk to people who don’t know what a treemap is, we still point to the excellent Map of the Market implementation which was developed originally by Martin Wattenberg in the late 1990s. The treemap has become popular in recent years for many different types of applications, which attests to its versatility and general applicability. Since then, the basic structure of the treemap has not changed much, although several different tiling algorithms have been developed. The treemap was invented in the early 1990s by Ben Shneiderman at University of Maryland College Park. With skillful application of color palettes, coloring and sizing metrics, it can be easy for people to get an overall sense of a situation, and help answer questions like: Is the big picture good or bad? Which items are doing best and which are doing worst? One of the best features of a treemap is that you can see a large number of items at a glance, perhaps several thousand, without any one being obscured by another.
We find tree maps an extremely helpful way for people to see a big picture of a situation made up of many granular parts.
0 kommentar(er)
