Financial Reporting Using XBRL

The title of this blog entry is a statement made by Christopher Smith of iBanknet in an email exchange he and I had.  Chris pointed out that the SEC has added ATOM feeds to their Edgar system and that the combination of the ATOM feeds and XBRL "...is EXTREMELY powerful in terms of data dissemination."

I totally agree with Chris and I agree that the SEC "gets it".

The SEC started using RSS feeds two or three years ago to push information about XBRL filings received under the SEC's Voluntary Filing Program. See http://www.sec.gov/Archives/edgar/xbrlrss.xml

See the prior entry on this blog for a link to the iBlanket widget which makes use of one of the ATOM feeds.

You can go look at the Edgar system and see the ATOM feed here.  Look for the little orange icon, click on it.  You can read the feed in your browser, but the important thing to understand is that a computer application can read this also.

Basically, according to a IRWebReport blog entry, each filer to the SEC has its own ATOM feed which a user can subscribe to.  Imagine that the ATOM feed making you aware of information in an XBRL filing.  A computer application watches the ATOM feed, becomes aware of a new filing, reads the information from the XBRL filing, and updates your internal analysis models or other applications.  That is extremely powerful.

But the really cool thing that I see is that this same extremely useful technology can be used by any business using off the shelf software.  Imagine that any company (big or small) could create what amounts to something like a "stock ticker" and either read information provided by others or distribute information others would need.

More about this later.  But for now, great work SEC.  And thanks to Chris Smith from iBanknet for creating the demo widget and bringing this to my attention.

This is a new widget from iBanknet. It uses a new feature which was added to the EDGAR system: ATOM feeds (similar to RSS feeds). See the next blog entry for more information.

The IASCF has released an Exposure Draft of the 2008 IFRS Taxonomy.  The taxonomy can be downloaded here.

You may, or may not, be aware that XBRL Formulas is a candidate recommendation.  I have created a number of annotated XBRL Formulas samples which may be useful to business users who want to see how this stuff looks and works by looking at XBRL Formulas at the XML level (the angle brackets). There are five small but comprehendible taxonomies (sales analysis, movement analysis, directors compensation, etc) and one file which puts all the taxonomies together into one taxonomy. I organized the XML so it is easy to read and so it flows in a logical manner, to enable reverse engineering the linkbases. I also carefully named the XLink labels to help someone “walk” through the linkbases. Basically, look for the “*-formulas.xml” file in each subdirectory.

The taxonomies and instance documents validate against three different XBRL processors which support XBRL Formulas (and it is great that there are already three processors!). I did everything I could to get this right, but be warned that this stuff is rather new.

For these samples, see here.  (The XBRL Formulas page on this blog.)

XBRL Formulas is critical.  There are many computations which cannot be validated by XBRL calculations.  XBRL calculations are very limited.  The biggest limitation is that they only work if all the things you want to calculate are in the same context.  For movement analysis (beginning balance, changes, ending balance...each of which is in a different context) and for cross dimensional computations (each dimension in a different context), calculations simply will not work.

Keep watching the XBRL Formulas page, more to come.