Written by Bob Schneider    Posted on December 3, 2009

Auditors must not only be independent but be perceived independent as well. The rules for book reviewers should be no less stringent. Finding critics who have background in the field but no skin in the game, however, is a daunting challenge — even for the biggest publishing institutions.

With these considerations in mind, I wrestled with writing a review of the new XBRL for Dummies by Charlie Hoffman and Liv Watson. That Charlie has always been a good friend of this blog, and that Hitachi XBRL products are mentioned in the book (among many others), could be dealt with through mere disclosure. Because Hitachi created a book with the same name, though, I worried readers would understandably wonder, “where’s he coming from?”, regardless of what I wrote.

That’s why, initially, I thought I’d merely tweet the July post that sought to distinguish the two XBRL for Dummies books and leave it at that. Having now read Hoffman & Watson’s book, however, I think it would be a disservice not to discuss it.

There are so many things I like about this book. Often while reading it, I found myself observing, We should have done a post on that, and I should have suggested to so-and-so to write about that. Hoffman & Watson strike a deft balance between providing readers with a comprehensive-but-not-overly-technical explanation of XBRL and discussing how the standard affects their business lives, and thus why they need to know about it. The book doesn’t overwhelm with an extended technical explanation of XBRL in one barrage. Rather, it gradually introduces knowledge of the standard, so that at each point the reader feels informed to appreciate the authors’ larger points about why XBRL is important to the exchange of business information.

Throughout, the book greatly benefits from all the work Hoffman has done in recent years to make XBRL understandable to business professionals, and a truly massive number of Web links he has assembled for references.

Here are some parts I especially appreciated:

Syntax Is Not Enough (pp. 41-47)

Hoffman & Watson pointedly emphasize that XBRL is information structured for meaning — not merely for presentation. They pithily state, “There’s a big difference between <bold>1000</bold> and <Sales>1000</Sales>” [emphasis added].

Recent articles in the business press often referred to XBRL as “tagged data,” but the discussion reminded me of how inadequate, even misleading, this description can be. Perhaps we can rechristen XBRL as meaningful business information, or MBI? (At the least, it would provide a playful title for a post about the use of XBRL for mortgage-backed securities: “MBS Require MBI — Even an MBA Knows Why”…)

Explaining XBRL at the Water Cooler (49-52)

Hoffman & Watson realize that a primary motivator for buying their book will be, My boss just asked me what XBRL is — what do I tell him? In addition to providing an XBRL-in-a-nutshell explanation, the book offers (extensive) suggestions for having follow-up conversations with anyone who casually asks you to explain the standard.

Creating and Using XBRL (Chapter 15)

I like it when a book about a technical subject gets the reader directly involved with hands-on examples. An entire chapter provides readers with step-by-step exercises to view XBRL taxonomies and instances — small ones that Charlie did himself, as well as large, external ones — and import XBRL data to Excel. (I didn’t have success in downloading the SpiderMonkey software required to do some of the exercises; but that may be because of factors peculiar to me.)

Differentiating XBRL Modules (Chapter 16)

XBRL is a family of specifications with modules for dimensions, formulas, rendering, versioning, and generic linkbase. While this unity may be obvious to practitioners, I don’t think it is fully appreciated by more general observers, who may just see the string of apparently unrelated announcements that have been issued on each specification. Chapter 16 gives the reader a sense of what the whole of XBRL entails.

Looking at XBRL Taxonomies and Instances (308-312, 333-338)

Often, when exposed to new computer areas, the novice asks, “What am I looking at? Is this the same thing I just saw presented differently, or something else entirely?” Hoffman & Watson easily could have dismissed this concern in two sentences, but they opt to present taxonomies and instances as they appear in the physical files, viewers, printouts, etc., to expose the reader to the various ways that content will appear.

A Thought Experiment (330-33)

An extremely interesting exercise discusses some of the difficulties involved in using XBRL data. In the example, the objective is to find a good investment on a worldwide basis. Hoffman & Watson describe all the challenges that await the user, from finding the instances to creating the analysis interface. This exercise well demonstrates the gap between what is technically possible and politically achievable.

Identifying a Good XBRL Taxonomy and Instance (321-323, 343-346)

What struck with me was how many of Hoffman & Watson’s recommendations — no duplication, consistency, documentation, maintenance, adequate testing, and, not least, elegance — are the same as those for developing a great database. That comes as no surprise to the authors: they note the similarity, too.

Specific Uses for XBRL (353-355)

This is the authors’ wish list of what software applications will be capable of accomplishing by making use of XBRL. One potential that Hoffman has emphasized but that still remains under-appreciated is an automated disclosure checklist: basically a set of business rules that can ensure that financial statements are properly created.

Keys to Understanding How XBRL Works (369-376)

A good review of some of the book’s main points, nicely summarized.

Isn’t there anything to criticize? There always is.

On the “what’s missing?” front, there’s little about how XBRL got started. In contrast, XBRL for Interactive Data by Debreceny et al. contains a long introductory letter on XBRL’s beginnings by Hoffman, and the AICPA has published an online history.

I also don’t remember seeing much about XBRL vis-à-vis assurance or auditing, and the absence of these terms in the index would seem to confirm my recollection.

Turning to “what didn’t I like?,” I didn’t find the substantial discussion of networks in Chapter 4 and elsewhere to be helpful to my understanding of XBRL. Networks is one of those MEGO words (like systems) that’s difficult to get your arms around, and I couldn’t do so here.

As well, having made the decision not to write a book primarily for IT people, I could have done without the concluding chapter of technical “odds and ends.” Reference books are rarely read cover to cover, but this chapter of technical tidbits may leave the average reader feeling that he doesn’t really know XBRL after all, despite having just read the book.

These are quibbles, though. Hoffman and Watson’s XBRL for Dummies not only teaches XBRL but gives the reader a strong understanding of the issues involved in business information exchange.

It is a major achievement. Buy the book.
 

Add to Del.icio.us | Digg this

This entry was posted on Thursday, December 3rd, 2009 at 9:57 am and is filed under General, . You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.