On standards and standards bodies

by David Lane

What does it mean to be open.

My copy of Oxford defines open as: unconcealed circumstances or condition. Way back in the day when the GNU operating system was getting going, they coined the mantra: Free software is a matter of liberty, not price. To understand the concept, you should think of free as in free speech, not as in free beer.

Last month, I talked about transparency and how important it was in software and systems. Just as important are standards, and, more important following those standards. Today, in Computerworld, a different issue has been raised. The value of standards.

Way back, last year, there was a ratification of standard by the International Standards Organization (ISO), the same group of people that brought you the stupid label guy (ISO9000), IS-IS routing (does anyone really use it?) and of course, the OSI stack (Please Do Not Throw Sausage Pizza Away). The standard that was ratified was the Open XML standard. Now, I am not that much of a geek to be able to accurately reflect the arguments for the Microsoft (ratified) version and the non-Microsoft (not ratified) version that came to pass. I won’t lob too many stones at Redmond (that bastion of standardization), but I will highlight one point. There are some countries who are less than happy with the ISO and, in fact, are so dissatisfied that they are questioning not only the Open XML standard, but the value of any of the ISO standards at a national level.

My father used to work for the telephone company, back before Judge Green broke up AT&T. He has since moved on and dabbled in the computer industry and is currently working on smart buildings. One of his constant complaints is the lack of standardization in the computer industry. And this from a man who helped a couple of companies actually make money back when computers were expensive items. In many ways, I have shared in his frustrations. He is management, but technical enough to grasp most of the issues. I am a technician and have had to wrestle with the standardized non-standards in the industry. Even something as simple as a PCI slot is enough to drive you nuts (and if you have been around for a while, we all remember the headaches of EISA, and “where is the disk…”). Standards are important, but for a standard to be accepted, it has to work, and it has to work well. We can all look at the standard wars between Betamax and VHS (or Lightscribe and Labelflash) to see how important, or how mind numbing the different standards can be and how much they can affect the technology that is adopted, and as we have seen, better does not always win.

But when countries start questioning the entire standardization process, or worse, as is the case with the fight over Open XML, start accusing the standards body of being unduly influenced by corporate concerns, we then have a real issue that needs to be looked at deeper. Standards bodies cannot afford to be even thought of being driving by a corporate perspective, despite the fact that many standards start out that way. Standards bodies, to be of any value must be independent, and must be willing to consider, up to a reasonable point, objections to the standard. If not, then the whole issue of a standard is moot.