More on Database cells, and other terminology

Now, I'm about to take the other side of the argument, in a slightly different context. I think references, even public ones like Wikipedia, have an obligation to be very precise in their definitions.  I looked up "table (database)" in the wikipedia, and found this article: http://en.wikipedia.org/wiki/Table_%28database%29

I think the use of terminology is too loose, for an encyclopedia article. It could mislead the unwary reader. Here's an example:

"In non-relational / hierarchical systems, the equivalent of a table is a
structured file, containing lines or records (the rows) and fields (the columns)."

I think the use of "equivalent" is too strong here. The concepts, as Joe Celko and others point out, are NOT equivalent. The following wording might be better:

"In non-relational / hierarchical systems, the corresponding thing to a
table is a structured file, containing lines or records (the rows) and fields (the columns)."

"corresponding to" sounds to me more accurate than "equivalent". This may
seem too notpicky to some. But I'd like Wikipedia to be as close to on target as feasable.

Continuing on.

"Unlike a spreadsheet, a database table can not take arbitrary information
in any cell, nor can values be represented as formulae which compute values on the fly. The datatype of each field is strictly defined by the schema describing the table."

Note the use of "cell" for spreadsheets, while the word "field" is used with databases. That suggest to me that the term "database cell", which I advocated, might be less than ideal, and that a better terminology would be to use the word "field" for the intersection of a row and a column. It would still be important to distinguish between a "field" and "column". Received on Sun May 07 2006 - 19:46:50 CEST