In a software architecture forum that I follow, a member recently posted a question about the community’s view of a standard named TOGAF. He wanted to know how many of us worked in organizations that were committed to the standard, how many were exploring it, and so forth.
The comment stream was interesting. One early responder asked for a definition, having never encountered the acronym. Others pointed to general descriptions of the standard. Data-rich, experience-based answers to the question seemed noticeably absent.
Why do some standards fail to gain traction, even among those who ought to care about them the most?
The answer is not laziness, or stupidity among implementers, or naiveté among creators.
We all like idealized, tidy answers—yet we deviate from our ideals regularly. Most of us would love to pay cash for the home or condo we live in, but we settle for mortgages instead. Forced to choose between ownership now and a long rent-filled delay, we get pragmatic.
Software organizations and IT pros are no different. Many standards take significant time and organizational discipline to implement. This commitment must be sustained over long periods of time, and be reflected consistently throughout a product line or software stack, in order for a standard to yield dividends. When you consider that standards have greater benefits in industry as the network effect kicks in with mass adoption, built-in incentives to “wait and see” become obvious. Add to this the phenomenon of competing standards, and the fact that standards frequently undergo revision and updates, and serious questions arise about whether adopting a standard is worth the hassle.
The best standards (from a pragmatic—not an ideal—standpoint) accumulate enough momentum to overcome these inhibitors, and even laggards end up signing on. This happened with VHS, Blu-Ray, TCP/IP, HTTP…
I’m beginning to feel a similar tipping point around OpenStack, at least with certain segments of the market. This makes me very excited about the Moab~OpenStack integration that Adaptive Computing has been incubating. It delivers best-of-breed placement, migration, and policy-based optimization on top of datacenter building blocks shaped by standards and broad industry consensus.
We’ve had a proof-of-concept deployment in customer hands since late 2012, and we’ve been steadily absorbing feedback and reacting to OpenStack’s internal maturation ever since. If you’re interested in learning more, leave a comment below or contact an Adaptive Computing representative.