Modeling Against the Stream
The two articles, Against the Stream and Meaning is Not Metadata highlight a problem that I have experienced myself and have tried to solve related to digital financial reporting by creating things like my Logical Theory Describing Financial Report.
Communication between business professionals that need software to do something and technical professionals that build that software is hard. Very hard. But it is important to persist to make sure the foundation is right. Why? Because you cannot retrofit a foundation to fix a foundation.
The first article points out a "shared fiction" that currently exists between business professionals and technical professionals.
What is important to understand is the ramifications of that shared fiction.
One ramification that I have experiences is software that is overly flexible in undesirable areas which results in software that is harder to use than it really needs to be. Flexibility makes things more complicated. When you have flexibility that you don't need you get what is referred to as "accidental complexity".
Another ramification is that important things go missing. My blog post, Consequences of Starting at the End of the Chain, points out that the notion of a "business event" is completely missing from every accounting system that exists today. This results in a need to add business event information later in the process of creating financial statements, usually in a spreadsheet. This missing business event information also means that you cannot leverage that information until the information has been added.
This brings up what that second article is pointing out. The notion of a "business event" is core to accounting and financial reporting. Fundamentally, what a financial statement provides is information about the "state" (i.e. balance sheet) and "changes in state" (i.e. income statement, statement of cash flows, statement of changes in equity) of the reporting economic entity. That information is provided using some reporting framework such as US GAAP or IFRS.
One thing a reporting frameworks provides is standard financial line items such as "Cash and Cash Equivalents" and "Trade Receivables" and "Property, Plant, and Equipment" and "Trade Accounts Payable" and "Retained Earnings" that are used on financial statements. Those standard financial statement line items provide comparability with consistency. These standard financial statement line items are the basic level of a taxonomy. Superordinate level taxonomy items such as "Assets", "Liabilities" and "Equity" which are part of the double entry bookkeeping model act as "containers" for the basic level taxonomy items. Other superordinate level items such as "Current Assets" and "Noncurrent Assets" and "Equity Attributable to Parent" provide another operational level. Then subordinate level taxonomy items such as "Cash on Hand" and "Cash in Banks" and "Cash Equivalents" provides important details that might be used in the primary financial statements or to provide the details of subclassifications of the basic level taxonomy items in the disclosure notes.
Another thing a reporting framework provides is the standard financial line items such as "Proceeds from Long-term Borrowing" or "Dividends Paid" which appear in the Statement of Cash Flows and/or Statement of Changes in Equity. Those changes in state of the balance sheet line items are caused by business events.
You can see what I am talking about in a working proof of concept financial statement I created. For example, have a look at the roll forwards in that working proof of concept. Here is the roll forward of cash and cash equivalents:
What is really interesting is that now, with digital financial statements, machines can read and interpret financial reporting frameworks and the financial statements created per that financial reporting framework. Here is one such verification result. If you look at that set of HTML pages which obviously can be significantly improved; you can see how the software can reason with respect to the report.
Something important to recognize is that the rules used to ENCODE a financial statement and the rules used to DECODE a financial statement are the same.
If you look you see several things going on. One example is that you can see that the software reasoning about the report can interpret the information blocks within the report and determine the disclosure the block of information represents. How is this achieved? Well, additional metadata is created to help convey the fundamental meaning provided by the financial statement. Here is the financial reporting framework used by this working proof of concept financial statement. Here is exactly the same information in machine readable form which is global open industry standard XBRL.
Now, all you need is any off-the-shelf software that understands the global open industry standard XBRL. Here is a list of such software which is certified by XBRL International. You can easily download a copy of Arelle, which is fully certified and which is a freely available open source software application which can be downloaded or forked on Github.
So what is my point here? If you feel like a spreadsheet monkey; if you feel things are not as connected as they could be; if you feel like a data janitor; if you intuitively feel that there is a better way. you are right.
When all this is sorted out, new industrial strength processes will be replace what people currently consider normal. A new normal will be created. Technical professionals will build these new business processes. But business professionals will specify what those technical professionals construct.
Don't go against the stream. This new normal needs to minimize the accidental complexity. The new normal that we create should look elegant. That elegance is a clue that you are embracing the ideas of Simple Made Easy. Begin at the beginning. Mistake proof the systems. Make systems fail loudly.
Additional Information:
- Example Financial Statement Holon
- Digital Information Organism
- Come Together
- Complexity
- Pacioli in the Computer Age
- Taxonomies, All the Way Down

Comments
Post a Comment