Do It Yourself (DIY) XBRL - Is There a Better Approach?
I've been working with global XBRL implementations over the past eight years and sometimes get into discussions with experienced software developers in regards to the "buy or build" argument. Obviously this argument isn't unique to XBRL and, being a software developer myself, I totally understand developers questioning the need to use a specialised XBRL package to create "a simple XML document".
Typically I hear comments which can be summarised as - "All we need to do is build a simple XML handler which can generate the XBRL instance documents which SuperStream mandates. They're just XML documents which need to conform to a schema. How difficult can that be?"
While understanding such comments, I don't believe that the requirement to create XBRL instance documents consistently and reliably is quite that simple. The document below is my response to the questions which have been raised with me over the last eight years and, I am happy to admit, reflects my personal opinion. I would also like to highlight that I have more than a personal interest in the topic as the company I work for has a set of commercial XBRL offerings!