As I have mentioned in other posts on my blog, a financial statement is a formal semantic structure, in fact a financial statement is a special type of formal semantic structure referred to as a holon. As I mentioned in my blog posts, Kurt Cagle and Chloe Shannon are providing a lot of very good information about holons on their substack.
I have provided a bunch of good examples of financial reporting frameworks and reports In this blog post I want to focus on one specific financial statement and see if I can explain that a financial statement is a "container" of blocks of information and the "information blocks" that the financial statement contains.
Cagle and Shannon point out that a holon is made up of four distinct knowledge graphs which are:
- The Scene Graph or Interior State
- The Boundary Graph
- The Event Graph
- The Projection Graph
What I want to do is explain all of the above as they relate to financial statements and provide examples of each of the above. I am going to run this information by Cagle and Shannon and see if I can get them to help me tune my understanding where I get my explanation and examples wrong.
Also, while I am providing all of this information which was created using the XBRL technical syntax; Cagle and Shannon use the
semantic web stack and more specifically RDF 1.2 (i.e. RDF Star). Personally, I believe that XBRL is a legitimate technical syntax for creating knowledge graphs and expressing holons and even has some advantages over RDF and the rest of the semantic web stack. However, I do acknowledge that RDF and the semantic web stack also has significant advantages over the XBRL technical syntax.
Further, I am very confident that what I have represented using the XBRL technical syntax can be converted to RDF, RDFS, OWL, and SHACL relatively easily. The XBRL could also be converted to Graph Query Language. I say this because the XBRL was converted into PROLOG for processing the XBRL-based information by one of the software applications that I am using.
The Scene Graph or Interior State
The scene graph or interior state is certainly, in part,
the XBRL instance which contains the FACTS provided within the scene graph or interior state and changes in state communicated by the financial statement. What is unclear to me is if the mathematical rules provided by
XBRL formulas as part of the XBRL instance and if the
report model is part of the scene graph/interior state or whether that is more part of the boundary graph. I would speculate that the mathematical rules and the report model which both show the fundamental arrangement of the facts are part of the scene graph or interior state. Could be wrong about this.
The Boundary Graph
The first option is inferior for viewing the information because I have not provided views of the information for all those links at the top of the page because of my limitations as a programmer. The second option does let you see all of the information in that theory file which is 100% XBRL technical syntax and this works using any off-the-shelf XBRL software. For example, I can load that theory file in
Arelle which is freely available open source software. However, the view is incredibly technical oriented and Arelle actually has some flaws in what it shows. For example, I cannot see all the XBRL definition relations that I have defined; Arelle simply does not show them for some reason. Another off-the-shelf XBRL software application, Pesseract, shows all the XBRL definition relations correctly; but it has a different flaw in that it does not show XBRL formulas. But between the two, Arelle and Pesseract, I can get a rather technical view of all the information defined by the boundary graph.
The Event Graph
In the case of a financial statement, the event graph would be the detailed information from the general ledger and subsidiary ledgers that feed the general ledger. This includes information about
business events that have been turned into financial transactions and posted to the accounting system which generated the information that was used to create the financial statement.
Now, this specific example does not provided that detailed transaction information. However, my
closing book example and my
record to report examples do provide that business event information and it shows how the business event information connects to the financial statement information.
The Projection
The projection relates to the separation of the
document "costume" which a financial statement wears and the information contained within a financial statement. The
core pattern relates to the information which is contained within a financial statement. XBRL-based reporting uses a "graph first" paradigm which is also a "model-driven" paradigm. What that means is that the information is provided in the form of a knowledge graph which is machine interpretable. Additional information is also provided to take that graph of information and represent it as a human readable projection. Not a copy of the information, literally there is only one version of the information.
Here are various projections of the scene graph or internal state, the first graph mentioned in this set of graphs that describes the holon that is a financial statement. This first view is provided by off-the-shelf software that supports the XBRL global open standard. Again, this is freely available open source
Arelle which was also mentioned above:
Now, the above projection leaves a lot to be desired. It is not particularly business user friendly. But this rendering as a technical oriented software application, a projection, is only one projection of the XBRL-based information.
Here is another projection which is the same exact information, the XBRL instance and the XBRL based report model, represented as something easier for a business person to relate to:
The above is a desktop software application which was created as a working proof of concept to understand XBRL-based digital financial reporting. It has a lot of interesting and useful capabilities; but is still a desktop application create as a projection of that same XBRL instance and XBRL based report model.
Here is another software application, this time a cloud based software application which you can try for yourself.
Click on this link and you will see the application show in the screen shot below:
And here is yet another projection which uses that same XBRL instance, the same XBRL-based report model; but this time it also adds information from the theory file, also pure XBRL, which is used to apply the boundary graph to the scene graph within the projection to help the user verify that the information is correct: (
click here to view the set of HTML pages that make up this projection)
Now, the quality of the GUI/UX of the set of HTML files in this projection above is not all that fantastic, but to focus on the quality is to miss the main point. The main point here is that one version of the information is used to create multiple different projections. One of those projections is another tool that is used to edit the scene graph or internal state. Meaning, the screen shot below show a projection of a software editing tool used to create the scene graph/internal state:
This
YouTube video playlist shows earlier versions of the above software being used to perform various tasks related to the creation of a financial statement.
Here is one final projection worth mentioning. This projection is automatically generated Inline XBRL. You can use this yourself
simply by clicking on this link.
* * *
And so there you have it; what I showed amounts to a financial statement holon, the container, and the information blocks which exist within that container. The container and the individual information blocks follow those four graphs Cagle and Shannon say that exist for every holon. I am not sure I have gotten all this correct; but I am trying to understand at this point in time. I know I am not using RDF 1.2 as Cable and Shannon suggest; but I think I can say XBRL plus the financial statement structure which leverages the Atomic Design Methodology provided by the Seattle Method does have some advantages.
Additional Information:
Comments
Post a Comment