Digital Financial Reporting

I mentioned in another post a few months back that a good example of how business reporting could be improved is to look at the CIA World Fact Book.  You can go read that post and see why I say the CIA should not make the data available in HTML, but rather in some machine readable XML format; perhaps even XBRL.

Well, guess what.  Now you can SEE why providing this information, and other information, might be a good idea.  Someone wrote a parser or something, took all that CIA World Fact Book data and put the data into a Semantic Web-type database.  You really have to check this out.

There is more.  Check out the elastic lists here.  Or, there is lots of other stuff here.

Now, use your imagination.  Imagine putting a similar type of interface on top of the SEC IDEA system when it gets up and running.  Imagine navigating the database in that manner.  Imagine navigating the information in the fillings in a similar, or even better, manner.

The following is a list of URLs which point the the four XBRL conformance suites currently published by XBRL International.

• XBRL 2.1 Conformance Suite: http://www.xbrl.org/2008/XBRL-CONF-CR4-2008-07-02/xbrl.xml
• XBRL Dimensions 1.0 Conformance Suite: http://www.xbrl.org/2006/XDT-CONF-CR3-2006-06-06/xdt.xml
• FRTA Conformance Suite: http://www.xbrl.org/technical/guidance/FRTA-CONF-RECOMMENDATION-2005-04-25/index.xml
• FRIS Conformance Suite: http://www.xbrl.org/technical/guidance/FRIS-CONF-PWD-2004-11-14/index.xml

Here is a link to an RSS feed which can be read by a software application to grab information about these conformance suite tests: http://www.xbrlsite.com/rss-conformance.xml. 

Here is a link to a web page maintained by one software vendor, Reporting Standards, which publishes their test results against these tests:  XBRL 2.1=http://www.reportingstandard.com/conformance.xhtml; XBRL Dimensions=http://www.reportingstandard.com/xdtresults.xhtml.

Personally, I wish more vendors would publish their test results which explicitly states their compliance to, or non-compliance with the XBRL International specifications, which modules of XBRL they support with and comply with.  Seems like a reasonable, a transparent thing to do.

In addition to being essential for being sure software complies with the XBRL specification or a module or other guidance, these tests are great examples of how to use, or not use XBRL.

Business users purchasing XBRL software should ask software vendors which of these conformance tests the vendor passes.

OK, I will be the first to admit it, I don't totally get what is going on here.  I am trying to understand.  The reality is that I am not sure anyone really understands what WILL happen.  There are probably more who understand what COULD happen.

These seem to be that some of the moving pieces are things like:

• Facebook:  "Facebook?", you may be asking.  Yes, Facebook.  Either the people at Facebook "get it", or maybe they don't, or it may even be that I am seeing this wrong.  But, this is what I see.  One of the biggest problems relating to the Internet is that the Internet does not know who I am, what I like, what I care about, and what I really don't care to have anything to do with.  So, I get everything.  That is overwhelming, information overload.  BUT, does Facebook have the answer?  Profiles.  Yes, your profile.  What if, say, Facebook allowed you to make meta data available about yourself and then other web sites, search engines, robots, and other "stuff" out there on the internet trying to help you could really help you.  I mean, what if as part of Facebook and Facebook's ability to turn information which is made available to categories of users were applied to meta data.  What if you could use Facebook (it really does not need to be Facebook but they do have a lot of market share) to create meta data which would help web sites and other things trying to understand you, to understand you.  Then, in one place (i.e. Facebook) a user could provide a profile in computer readable meta data to help you get customized news feeds, customized magazine articles, etc.
• U.S. Federal Government's Use of Semantic Web:  People are pushing the semantic web, I think I understand at least the basics of why the semantic web is better than the current Web.  It is not the case that we get one or the other, we can have both.  It seems like this financial crisis, the Obama administration's understanding of what technology can do, and some of the things which are already being done (i.e. Recovery.gov, USASPending.gov, the SEC IDEA system, FDIC's use of XBRL, etc).  Now, these things are NOT the semantic web, but it is a good first step...get the data out there.
• Linked data:  Tim Berners-Lee gave a presentation on Linked Data at the TED 2009 conference.  Each agency in the Federal Government independently creating a bunch of data and making it available in XML (rather than HTML) is better than nothing.  But, a coordinated effort would be even better.  I don't see much coordination.  For example, the XBRL made available by financial institutions to the FDIC and the XBRL financial information made available by banks to the SEC are not related in any way, manual relations therefore would need to be created to indicate that the information at the FDIC and the SEC is for the same company.  I don't quite get stuff like this.  Guess it means something.
• XBRL:  How XBRL relates to "the semantic web" is not clear to me.  XBRL is XML, it does express semantic meaning.  It does express meta data.  Most certainly XBRL is data made available on the Web, and it is way better than PDF or HTML financial statements.  But how does XBRL relate to RDF, OWL, and all the other things mentioned above in the link describing the Federal Government's use of the semantic web.  I don't get it.

I do know this.  It seems like the probability of the semantic web becoming something useful increases every day.  That means that the good old ways of doing a lot of things will be changing.  What will change, how to adjust, how to take advantage of these coming changes; that is what I am trying to figure out.  Particularly with regard to XBRL; what exactly is XBRL's role relative to other types of XML, how can this be applied by the accounting profession (I am a CPA), how can a business leverage these technologies, etc.  I am quite sure some very useful things will come out of all this.

I was made aware of a presentation which Tim Berners-Lee gave at Ted about "linked data".  Here is a link to that presentation.  If you are interested in XBRL, you will definitely find this presentation interesting.

Here are two of the links mentioned in the presentation:

DBPedia: http://wiki.dbpedia.org/

Open Street Map: http://www.openstreetmap.org/

Linked data!  Get out there and make it, be sure to demand it!

There is a great book called The Box: How the Shipping Container Made the World Smaller and the World Economy Bigger which analyzes the standardized shipping container and its impact on our world. 

There are a lot of things which XBRL and the business community can learn from the adoption of the standardized shipping container.  I came across a 19 minute interview with author of The Box, Marc Levinson, which can help someone understand why standards are important, how the impact of a standard goes beyond what you might think, and understand parallels between the shipping container and XBRL.

Here is a link to information about the book and the a set of interviews with Marc Levinson.  This breaks the video down into small pieces for downloading.  There are lots of different video formats, etc.

Here is a link to the entire video which i am talking about in QuickTime format.

One of the primary messages which I got out of this was the nature of the uncoupling of the manufacturing process and the transportation process caused by the shipping container.  This allowed for manufacturing to be basically anywhere (i.e. like in China, rather than in the US) and that ports could also be anywhere.