I thought I had my naming standards down pat, until another developer pointed out to me that file names in a UNIX server cannot have spaces.


This is a bummer for me, I like spaces. I think users like spaces. And there's nothing uglier to me than an entire TM1 server named like database tables.

Users see the cubes so they should be named articulately and meaningfully, with minimal abbreviations and codes.

Anyway, my current standard for cubes is as follows:


Where MMM is a 3-letter prefix indicating the module or division of the cube or dimension, PPPn is the purpose or description of the data the cube contains, and TTTn is the cube type, such as "Assumption", "Summary" or "Mapping".

For dimensions, I follow a similar standard. Each cube gets a measures dimension with the same name as the cube and a "_Measure" suffix.

All names are proper case, with no plurals, and if there are multiple words in the PPPn part, I separate with additional underscores.

I really believe naming is the fundamental key to usability, browsability, maintainability, trainability and a whole lot of other abilities!

What other naming standards are there out there, and do you think naming is as important as I do?

Flow OLAP -- We think outside the cube!