"Within a few months, Pito had come up with the fundamental idea at the core of Improv: that the raw data in a spreadsheet, the way that the user views the data, and the formulas used to perform calculations can all be separated from each other."

This is exactly what FileMaker Pro does.

You have the underlying data structure that is independent of calculations and presentation. It is sparse by definition, and can store far more data than any spreadsheet in existence. It can include defined relationships between the different kinds of data.

You have the calculations, that are stored with the data structure, but are functionally independent from it. They can transform, combine, and summarize the data in just about any way you like. They can store their results, increasing speed at the cost of storage, or they can be on demand.

You have the presentation, which depends on the data but is defined separately -- you can have as many different presentation forms as you like, hundreds or even thousands.

And you have actual code that can run on the data and validate it, modify it, organize it etc.

I think only momentum and lack of vision has prevented FM from eating spreadsheets' lunch.