Bob Sutor, IBM's Vice President of Open Source and Standards talks in his blog entry about the ISO approval process and standards bodies in general, and the need for change.

Mr. Sutor and other ODF proponents are going to be telling you over the next few months about how the standards process has become broken recently and we're in trouble.

I beg to differ.

Some people, it would seem, have very short memories. The standards process for some things has never been right. Indeed, the entire concept that some industries gather vendors together so that users can be stuck with the compromise between their pre-conceived ideas and marketing blather is a little ridiculous. The storage industry springs to mind in this case. But standards bodies in general have suffered for years under problematic relationships that boil down to some company's desire to see a standards body as a way to get a leg up on the competition. The screaming case in point for this goes wayyyyy back to the Tool Interface Specifications (TIS) committee, where allegations of wrong-doing against one organization were rampant.

Is the system broken? Yes, I think so. My problem is that a lot of people who are incensed by Microsoft's ability to push OOXML through ISO are going to present this as proof that the system is now broken. But abuses in every area of computing have not stopped the ball rolling forward. Remember EMCA Script? Yeah, a standards body you had never heard of that had nothing to do with web-based scripting was used as a front for a company to claim they were the standard. The more things change, the more they stay the same.

But look about you. Do your systems inter-operate? Do we have usable (if not optimal) standards for most of the core things that data-centers do? The answers to those two questions are yes. So how much mind share does the problem warrant?

I loathe abuses of the system by companies that can only see their limited world-view. To me standards need to be set for the entire industry in the best interests of end users. But the system, with all of its blemishes, has worked thus far, and we've got 20 solid years of large-scale standards activities in the computer space (longer in the electronics space) under our belt to say this is not a fluke. Could it be better? Yes. Is it broken? Probably.

Is this new? Not a chance.

I prefer ODF to OOXML, but believe the more open choice will win in the end, no matter how many rubber stamps are applied.

So why waste a bunch of cycles talking about it unless you're going to recommend that end users be the only voting members of standards bodies, and vendors that compete in the space have only an advisory role. Since someone has to pay the bills for standards meetings, that's not likely to ever happen.

One thing is for sure. No matter which of these standards wins out, you're going to need more disk space in the future. That's something worth wasting brain power on, and a problem you and I can resolve.

Don.

 

/imbibing: Mt. Dew and Coffee

/reading: Still Service Oriented Modeling by Michael Bell