Suppose that whatever people, since the invention of drawing symbols andwriting, have entered into the cells of tables consisting of rows and columns have been recorded in a database, it will sure make for a lot of knowledge management potential.
OK, we have spreadsheets available for exactly this purpose. But a lot of tabular data is recorded in word processor documents and database forms.
That spreadsheet confronts the user with thousands of row and hundreds of columns. But you only want to record 20 rows x 5 columns for a specificly identified chunk of data. And then you want to move to a next table of 3 rows x 3 columns.
So suppose datatypists had only one piece of software with one major menu command which is "CREATE TABLE". The button is clicked and the following fields apear:
* Column 1 Heading
* Row 1 Heading
* A blank cell.
The datatypist enters the data for these blank fields. One of the fields remains highlighted. Then Control up, left, down or right arrows are pressed and additional rows and colummns are created. Data is typed into the cells until the table is complete. The datatypist presses the button again and a new blank form appears for the next table.
The software evolves to distinguish between classification tables and alpha-numeric tables. Columns can be added which are based on formulas and colums can be sorted. One can scroll through the whole database and sort it. It even allows searches for cell data. Later versions allow for correlation between tables e.g. the table which displays the GDP of nations is correlated with the table of Ph.Ds in computer science for each nation to give a new table of computer science Ph.Ds per GDP per nation. The structure becomes a de-facto standard for tabular data. Internet users start to query tables for research purposes.
The software also recognises that people people record data in the following ways:
* Numbered an bullet lists. Hence it displays fields for a statement which introduces list items.
* Headings followed by paragraphs.
* Diagrams which indicate relations between variables and components.
Is something like this available as a distinct piece of software or as a front-end to any existing database?