Although many excellent data plotting programs such as Gnuplot exist, as do programs for data fitting and parameter estimation, it is valuable to combine these two capabilities in a single extensible program framework. Cartan makes use of interactive data-loading or data-synthesis, data-fitting and data-presentation ideas and employs Java and Java-Swing software technologies to support many of the x-y (CARTesian-ANalysis ) data manipulation techniques familiar to experimental scientists. A particular feature of the program is its inbuilt and interactive use of first order calculus methods to track the propagation of data uncertainties alongside the independent and dependent variables themselves. The program was originally intended as a teaching aid but can be used as a science productivity tool. A discussion of the code design and architecture is given alongwith a review of ideas on data formats, data parsing and the management and parameter and model estimation issues. Some ideas on how to extend such a framework to handle project-specific data models and formats are also presented.
Keywords: x-y data; model fitting; plotting; error abalysis; data uncertainties; interactive models.
Full Document Text: Not Yet Available.