Interactive Information Viewer (Prototype)

This is a prototype of what a financial statement might look like within an Interactive Information Viewer.  This is a prototype with a very small fragment of a financial statement.  The intension is to expand this prototype to be more like a real financial statement based on a real XBRL taxonomy and XBRL instance.  The list below explains the items on the table of contents.  Click on each item in the contents (on the left) and have a look at it.  Click back on this list to refresh your memory on what each component is intended to show:

1. Overview:  This page.
2. Printed report: This is what a financial statement looks like today.  It is “paper” or electronic paper such as PDF or HTML and static meaning that it cannot be modified by the user(s) of the information in any way because it is static. (It is actually quite easy to turn XBRL into PDF, in fact this PDF was created by applying a style sheet to an XBRL instance.  The problem is, this pixel perfect rendering has to be done for each XBRL instance which is not efficient and the pixel perfect rendering is not interactive, it is one, unchangeable, static view of the financial statement…just like the old days.
3. XBRL instance: This is what an XBRL instance looks like.  It is quite usable by a computer, but not that useable by humans as they cannot really easily comprehend what is in the XBRL instance.  Sometimes users must read business information.  Other times humans will want to get involved to see what is going on, but in general let computers handle the processing.  Both use cases must be met: fully automated and human involvement.
4. Interactive Information Video: Information should be more like a Microsoft Excel pivot table.  This video shows what “interactive” means. The information should be dynamic, interactive; not static and locked into one format chosen by the creator of the information. Should the creator of the information be able to show “the format” that they prefer the information be viewed?  Absolutely.  However, users should be able to pivot the information, slice and dice it, reformate it, reconfigure it, as they see fit.
5. Flow of Report: Business reports have a flow. They are meant to be viewed in a certain order and in many cases the order of the information is critical to getting the right meaning from the information in the report. The flow can be organized into groups (but they are not grouped here).  For example, groups of a financial statement are: primary financial statements, accounting policies, disclosures.  Different users could have different flow.  That is why the information is best interactive, so users can create their own flow.  This flow can be communicated within an XBRL taxonomy using a number of mechanisms such as the order of the networks and an organization of the hypercubes. Hypercubes are fact values which have common dimensions.  Basically the information is of the same shape. By organizing an XBRL taxonomy, rendering the information in an XBRL instance is easier because you leverage the good organization rather than fight the poor organization of the information.  This organization is called an information model which is similar to a data model but significantly more sophisticated. Information models have patterns, they are not random.  These patterns are called metapatterns. There are three very clear metapatterns in financial reports: roll up (or calculation), roll forward (or movement), or hierarchy (basically everything else).  Everything in the US GAAP or IFRS XBRL taxonomies fits into one of these three metapatterns.  If new patterns are discovered, then new metapatterns can be created.  Again, information models are not random.  These patterns can be leveraged in order to render information and make it interactive.  The multidimensional model is a big piece of creating this flexibility or interactivity.  These are examples of different pieces of a financial and examples of different metapatterns and the idea of shape.

a) Sales Analysis

b) Accounting Policies

c) Property, Plant, and Equipment breakdown by component

d) Property, Plant, and Equipment movements between two periods

e) Director Compensation

6. Information Model: This is the XBRL taxonomy which explains the information in the XBRL instance.

There is a lot that goes into making the above work.  Computers can hide the vast majority of the complexity from the business user using things like application profiles.  Consistency and predictability is what makes all this work.  Computers are very, very dumb.  They need to be told, explicitly, everything.  If something is left out software developers have to provide “flexibility” for users to do all the different things they might want to do.  Flexibility, in the right areas, is a very good thing.  Flexibility in the wrong areas causes ambiguity, inconsistency, makes things harder to use for no good reason, etc.

For more information see by blog and specifically the following blog entries on the topic of rendering XBRL information:

· Thoughts on Rendering (First Installment)

· Thoughts on Rendering (Second Installment)

· Thoughts on Rendering (Third Installment)