It’s a hard road, taking standards forward, but also a rewarding and ultimately positive one. This is a call for your help. The XBRL Formula Working Group has been working hard on a tightly related set of specifications that allow for the definition and creation of facts (numerical or Boolean) using XBRL instance (data) documents as input. The Public Working Draft (PWD) is now out and it needs input from engineers and technical architects with a firm grounding in XML Schema, XPath and XLink. The rewards? Kudos, good karma and (ok, this is a stretch but you never know) possibly undying adulation from a grateful world.
Formula is a pretty obvious add-on module for XBRL: creating analytics and imposing sophisticated validation constraints, with a full set of mathematical functions at your disposal to produce exactly what is needed. There have been proprietary efforts to use XBRL in this way and now, learning from those implementations, the consortium is seeking to define a standardised approach.
Who will use it? Regulators, banks and others collecting large amounts of financial and performance information will use it to impose complex validation rules to both improve the quality of data that arrives and automate the promulgation of those rules so that they can be executed in a distributed manner.
Professional analysts, broker/dealers, hedge and mutual funds as well as a wide range of financial infomediaries will use the formula specification to define proprietary analytics that can be used to add value to raw performance information.
What needs to be done? Well, the specifications themselves need a thorough review. There is a range of functions that need to be written. The largest chunk of work involves creating a conformance suite of tests that can be executed by software that support the specs. Obviously it will work best if the volunteers creating the conformance suite are also creating conforming software. And even better if the volunteers doing that work are not the same people that wrote the specifications in the first place. Practical, sometimes tough work. Need a way to convince management that it’s a worthwhile use of your time? Proven time and time again to be the very best way to learn XBRL at a deeply technical level.
Fall into any of those categories set out above? Pitch in please! Member of the XBRL consortium and either haven’t contributed to a working group yet or tend to do your analysis in-house? Step up to the plate! Software vendor that plans to support this stuff in your products? No excuses: Now’s the time. Join in!