Note: This article has been published at Hitachi Data Interactive, a fantastic review of current topics in XBRL. Check it out, here:
I recently posted about the new Form SD, where I asked a few questions about the structure of the XBRL taxonomy.
Since posting, I analyzed the instance document more closely. I also had a lively chat with the XBRL-Public Discussion group. The conversation was originally about tuples. The “Lines” of data reported (as seen in the picture below), had clear patterns; They need a Payment amount, a Project, a Country, etc. Some members, including myself, thought this was “tuple”-like.
Michelle from the Bank of Italy enlightened me as to why tuple structure is generally avoided in XBRL; the nesting of data allows taxonomy authors to create new meaning, or create the same meaning in a different way, (or hide meaning!); this wouldn’t be standard, so it wouldn’t be understood, and that’s bad.
But the tuple discussion was a tangent; I focused the second part of the conversation on missing data. Continue reading