- From: Richard Cyganiak <richard@cyganiak.de>
- Date: Sat, 6 Apr 2013 18:38:22 +0100
- To: Bill Roberts <bill@swirrl.com>
- Cc: public-gld-comments@w3.org
Hi Bill, Just a small comment: On 5 Apr 2013, at 09:44, Bill Roberts <bill@swirrl.com> wrote: > IC-2. I have generally found little use for data structure definitions, as when using a normalized form of the data cube, the data structure is implicitly but clearly available in each observation. That's not quite true. It's true that each observation tells you what dimensions, measures and attributes exist in the dataset. But it doesn't tell you the range of possible values, in particular for dimensions. In some of our work, we have datasets with an “indicator” dimension. We want to display a list of all indicators when the user first opens the app. In theory we could create that list by enumerating all distinct values of the xx:indicator property from all observations, but this would be rather inefficient. The DSD allows us to easily enumerate the indicators. Enabling efficient discovery of the cube structure was a main motivator for developing the Data Cube Vocabulary. If you don't have that requirement, then its spiritual predecessor SCOVO [1] is sort of okay. Best, Richard [1] http://vocab.deri.ie/scovo
Received on Saturday, 6 April 2013 17:38:50 UTC