The Great Cornholio

by Rick Jelliffe

Patrick Durusau has fun on his site with a posting satirizing the strategies of some opponents and proponents of OOXML at ISO as Beavis (for the the former) and Butt-head (for the latter.) Wikipedia has a good explanation of the Great Cornholio for people who don't get the reference.

His key passage is probably
I think the Butt-head side seriously abused a process that had been designed by the Beavis side for their own abuse but that hardly qualifies as an objection to OpenXML.

The notional trigger for the commentary is a worthwhile article by IBM's Arnaud Le Hors A Standards Quality Case Study: W3C which asks many good questions. I don't know that he is correct about Candidate Recommendation, however: From what I have seen, OOXML would have passed the Candidate Recommendation requirements from W3C: it clearly has an implementation and the differences between what Office 2007 does and IS29500 says are largely cosmetic; my understanding of the W3C CR regime was that it proved that a technology was implementable, not that every part had been implemented: consider SVG or XSL-FO for example. And I was a little puzzled that Ecma's process should be considered lacking because of its emphasis on timeliness, but later on OpenId was lauded because it is by a group of interested individuals that share a common interest and decide to solve it swiftly in a somewhat informal way using the internet to its full advantage. (emphasis added)

My angle is this: there needs to be a marketplace for standards bodies (plurality), so that stakeholders can choose the one that matches their requirements. And this in turn allows a marketplace for standards (plurality), so that users can choose the one that matches their requirements. And it is only to be expected that when there are competitive standards, vendors will attempt to use the standards process for marketing, to differentiate why their doovalakey is better than their rival's thingamajig. Caveat emptor: competition entails sorting through rival claims.


Rick Jelliffe
2008-04-29 16:48:03
UPDATE: I said more than I wanted to say above...

By "the differences between what Office 2007 does and IS29500 says are largely cosmetic" I mean that the differences as far as the W3C CR process is concerned, I don't mean to say that IS29500 is necessarily a perfect or complete description of Office 2007, though its rationale is to expose Office 200x. If it were perfect, it wouldn't need maintenance, and pro-IS29500 people need not get backed into any corner as if thinking that DIS29500 mkII made the bar for ISO standardization means that it could not benefit from aggressive maintenance: quite the reverse.

(The only area of concern I have heard stories about are spreadsheet formula, where some functions may need further alignment: they are definitely good candidates for a NB to do a review and issue a defect report if there is a discrepency.)