Innovation, Premature Standardization and Caveat Emptor
Filed under: Internet Protocols
Adam Bosworth's blog had a pointer to an interesting entry in another blog on innovation and standards which itself pointed to another good blog entry on the subject. All three came to the same conclusion – Standards shouldn't innovate. As anyone who has read my rule's of standardization knows, I agree wholeheartedly. And yet, premature standards, standards in areas that are not well understood, keep getting churned out. Why?
The answer is devastatingly simple – money.
Customers love technologies that have the word standard, no matter how disingenuously used, applied to them. Customer reactions to the word 'standard' are so strong that it causes vendors who are dread enemies to get together in back rooms and cook up something they can slap the 'standard' label on. Sure the technologies are typical examples of design by committee, sure they are badly thought through, have no real world experience and more or less fail to work as soon as they are tried in the real world. And of course they are never actually standardized in any useful sense of the term. But hey, the word 'standards' is applied to them, so they must be good!
You'd think that eventually customers would clue in. That they would learn the sort of things I talk about in my buyer's guide to standards and refuse to touch the stuff. But quite the opposite, everyone, everywhere, customers, analysts, journalists, etc. declare these messes to be the greatest thing since sliced bread. Everyone says "well sure, we know it has problems but 'they' will fix it." Personally, I'm still waiting for 'they' to show up.
Customer interest in standards is understandable. A true standard, that is, a specification that is based on long experience that has been through extensive review with clear intellectual property rights (IPR) and solid change processes provide customers with investment protection. It makes sure that no single vendor or even small group of vendors can screw a customer over. As long as the customer sticks to the standard they can mix and match vendors at will. Standards are good stuff. So who can blame customers for being excited when one is available and for preferring standardized technologies to non-standardized ones?
But the cruel reality is that real standards come at the end of a technology's innovation period, not the beginning. By the time a real standard shows up the technology isn't new or inventive. This is inherent to the process. Standards come after a technology is fully understood. Premature standardization results in badly constructed, difficult to use, non interoperable, non portable messes. But so long as the word 'standard' can be said next to these toxic spills, at least with a more or less straight face, customers will lap it up.
In the end premature standards fail. They collapse under their own ill considered, untested, badly understood incoherent designs. But who is hurt? Certainly not the vendors. They will just sell the next snake oil. The ones who get really hurt are the customers who are now stuck with sludge. The ultimate irony is that customers will end up paying the vendors tons of cash to figure out how to get the old snake oil to work with the new.
So who can blame the vendors? If customers won't get educated enough to protect themselves you can't blame the vendors for taking advantage of customer ignorance. Capitalism only works when both the buyer and the seller are vigorous in defending their own interests. If buyers let down their side of the bargain no one should be surprised if the sellers take advantage. Caveat emptor indeed.