Tom Petrocelli's take on technology. Tom is the author of the book "Data Protection and Information Lifecycle Management" and a natural technology curmudgeon. This blog represents only my own views and not those of my employer, Enterprise Strategy Group. Frankly, mine are more amusing.

Thursday, March 30, 2006

Imposed Standards Are The Most Enduring

I hate to admit it but the best standards are the ones rammed down our throats. Technology is chocked full of standards that have been imposed by a company or cabal that have been proven to have staying power. The telephone system, the cell phone system, cable TV (and now Internet access), and WINTEL are just a few of the many examples of imposed standards that we benefit from on a regular basis. In the storage world, many of the Fibre Channel switching standards in current use were imposed by a few companies, though eventually adopted by the T11 committees. Even Mozilla applications and the LAMP stack are controlled by a relatively small group of people, despite the term Open Source. A very small group of people decide on what is to be released and what is not. That's a cabal and that's good.

Constrast that to current situation with SMI, the Storage Management Interface, under development at SNIA for years. It is designed by a committee and as such subject to committee dynamics. So far, the output of the committee has not been widely adopted and is far from complete. Bits and pieces of the model are in use but there is no standard implementation - the original point was to have a standard interface to storage devices - and SMI is barely on the radar when it comes to buying decisions. SMI should have been, by now anyway, a standard feature that every IT manager assumes is in a product and without which no product could be purchased. Heck, it should already be implemetned in freeware and open source software. Too bad that's not the case, since this could be a very useful feature set.

It almost seems counterintuative that a standard that is imposed (is that really a standard then?) by a small group could be better or more quickly adopted then one created from an entire committee of the best minds from many organizations. There are three reasons why it is true.

First, the single-minded vision of a small group is much faster than the pluralistic mindset of a committee. Even if there is debate over direction or approach in the small group, there is usually a single arbitrator (such as an executive) to whom the group can turn to make a decision. That's rarely the case in industry or standards groups. While consensus would seem to make it more likely that a standard would be eventually adopted, trying to acheive consensus all the time slows a project down. Simply waiting for all committee members to go back to their own companies and come back with decisions can be time consuming. In many cases, consensus is nearly impossible because of the competing agendas of member companies. This is the bane of all industry groups - finding common ground in a competative environment. Having endured the old "what's good for the industry is good for your company" speech, I can say with authority that that's usually not the case. This is especially true when you are from a small and innovative company. Where's the incentive to share your best ideas until you have them out in a product?

Which brings us to reason number two - incentive. We live in a capitalistic world that values profit. Sharing your best ideas with a committee rarely results in solid return on investment. That is, unless you have products already that encapsulate those ideas. In that case, you not only control the market through superior products but the technology the market is based on. By opening up and standardizing your technology you remove obstacles to sales (fear of incompatibility and fear of lock-in) and generate side products that enhance the experience of your products. Meanwhile, you keep the good stuff for yourself. Sweet!

Finally, the standards are exposed to and shaped by market forces more quickly. Standards committees can become insular. They spend time contemplating their collective navels and arguing (yes arguing!) about tiny little issues that have no bearing on the market. I can still remember sitting in a standards committee meeting where the group argued for nearly an hour about the state of a single bit. When it was pointed out that the bit had no bearing on products or features that the IT managers cared about in the least, the group hesitated and then continued to argue. I left and had a drink with a potential customer, a much better use of company time. This tendency to lose market focus makes committee generated standards quite abstract and often useless in real world terms.

There are also imposed standards that fail. That's good. Very good. The market can shape products and standards quickly, discarding those that are not feasible or interesting and keeping the ones with real value. Committees cannot do that. The companies that send peopel to these standards groups have money and time invested in their ideas and won't let go of them until their customers make them. Standards committees thus become the zombie world for bad ideas that will not die a natural death.

So, when people whine that Microsoft is shoving a new standard down their throats, or that Apple has gone over to the Intel dark side, or even that we should "wait just a little longer for the committe to finish its work", remember that the imposed standard is usually the better one. Waiting fro the committee is like waiting for Godot. We all know by now that Godot never comes.


No comments: