Will Standardizing the Cloud Cause Clarity or Confusion?

The European Commission (EC), the Institute of Electrical & Electronics Engineers (IEEE) and the U.S. National Institute of Standards & Technology (NIST) have taken interest in the cloud for some time. Recent announcements suggest that each wants to make clear its own individual view on the cloud’s potential and how it should be regulated. Each institution also has some interest in standardizing parts of the cloud. But will three sets of priorities and interests underway at the same time leave us with clarity or confusion?

InfoWorld reports this week that the IEEE, which has developed standards for everything from high-speed data transfer to online learning, has chartered two new working groups to look more closely at the cloud. Working Group 2301 addresses existing profiles of cloud-computing standards, enabling “standards-based choices in areas such as application interfaces, portability interfaces, management interfaces, interoperability interfaces, file formats and operation conventions.” The related Working Group 2302 “defines topology, functions and governance for cloud-to-cloud interoperability and federation.” The latter group’s activities have more potential to get bogged down in politics, since its activities aren’t just technological considerations, and address issues like registration, trust and compliance.

The NIST attracted positive attention way back in 2009 with its attempt to define a “working definition of cloud computing.” Now on Version 15 (Microsoft Word document), that definition has been joined by a program of collaborative activities with which interested agencies, organizations and companies can engage. A meeting in Gaithersburg, Md. later this week will review promising deliverables such as a reference architecture (PDF) for cloud computing and a taxonomy (PDF) to clearly describe the roles and relationships involved in providing and consuming cloud services. Such deliverables could be important in enabling everyone involved in the cloud market to unambiguously describe what they want to procure or provide, and why.

In Europe, meanwhile, the EC is moving toward publication of a European cloud-computing strategy. A meeting in Brussels scheduled for May 23 will be the culmination of an online consultation process that is due to begin later this month. Europe continues to emphasize the importance of privacy and data protection, sometimes seen — unfairly — as being at odds with cloud computing. The European Commission’s Vice President, Neelie Kroes, reiterated this week 10 principles underpinning ICT’s role “at the heart of both economic and social progress,” the majority of which could be considered to actively endorse use of a cloud-based approach in enabling that progress.

Individually, each of the above initiatives is taking useful steps to clarify language and practice with respect to the cloud; each seeks to bring some stability to a market where definitions remain fluid. But together, they also face a pair of serious obstacles that could do them — and the cloud — harm:

  • Multiple efforts operating in roughly the same space with different agendas and priorities run the risk of contradicting one another and undermining market confidence in any of their findings. Only constant communication and cross-fertilization between groups can prevent this.
  • The cloud is mature enough for standardized terminology and understanding of key concepts, such as those being prepared by NIST. All three  of the above organizations’ efforts, though, must be wary of standardizing too much, and too soon. Overly prescriptive measures constrain innovation and trap markets into ways of working that may not prove to be what they need or want. Amazon’s recent announcement of the Dedicated Instance, for example, led some to wonder if the company was still delivering a “real” cloud. This market is still evolving, and we can’t afford to be too dogmatic or prescriptive.

A large number of skilled individuals and organizations are investing significant effort in these three parallel initiatives. If we can ensure that realistic objectives are set, and that all three initiatives communicate openly and regularly with one another and with the wider community, we have a real opportunity to gain sound technical and conceptual foundations on which to build the next generation of the cloud. The alternative — interminable committee processes, power struggles between standardization efforts, and over-zealous attempts to regulate and control — is one that we must all strive to ensure does not come to pass.

Question of the week

Can we standardize cloud computing without stifling innovation and new ideas?