I much like the use of the phrase 'visual analytics'. Visualization is analytics, and all analytics should start with visualization of data.
Overall they are basing their changes on the ability to establish 'Flow' (The psychology theory of Mihály Csíkszentmihályi ) when analysing and visualizing the data. Good idea, but this depends on the kind and needs of data and decisions being addressed. I am all for easy of use via Flow.
Quite a few new ways to select data points on maps, which for certain kinds of analyses can be very useful. The selection methods are quite intuitive.
Showed the integration of simple analytic, like regression based forecasting. Easy to overlay on existing data. Well done, but still a minimal amount of analytics. When comparing it to other visualization packages like Spotfire.
I like very much what they have done with being able to derive level of detail (LOD) variables. This is a very powerful concept, an excellent way to do some kinds of analysis, like Cohort methods. But at the same time it takes you away from the purely visual interaction with data. It is nice to know it is there, but it also gets you into the weeds of analysis. The average non analytic user may be confused. They have a white paper coming on its value and use. ( Will place link here ) Good
New kinds of data cleanup, meta data inclusion, ability to manipulate the data during its ingestion. Data that is strangely arranged or poorly pivoted in its original form. All good ideas that should prove useful. Lots of times you just want to get the data in cleanly for ready use with a minimum of effort.
Look forward to getting my hands on this, which should be very soon.
No comments:
Post a Comment