Hi,

I'm new to SDMX and have just attended a SDMX workshop. Now that I have an understanding of what SDMX can offer from a theoretical point of view, I am interested in gaining some hands on experience with developing a simple prototype that could be used to demonstrate some of the practical benefits of SDMX. I just have a couple of questions before getting started:

- I can see the natural fit of the SDMX information model with dimensional modelling concepts and hence its applicability in the data warehousing space. What I'm not so sure about is how the information model is applied in the data collection phase. A single data collection could potential have a large number of variables associated with it. So how do you determine the data set definitions that should be defined? ie. A single DSD for the entire collection, or do you break down the collection to be represented by several DSD's, and if so what is the criteria for dividing up the collection?

- I understand that SDMX is primarily suited for aggregate data. But if it is appropriate to apply the SDMX information model to a data set in the collection phase then in the context of a statistical survey doesn't that mean that one of the dimensions for that data set will be the statistical unit (ie. the respondent), and hence you would then be using SDMX for unit record data?

- The SDMX information model much like the dimensional model seems well suited for modelling data sets when the measures are numeric. How do you apply the information model to data sets where there are textual measures?

Thanks,

Keith.