Defining Open standards

I have met those who claim defining open standards is next to impossible and hurts innovation. Neither is true. Usually, those who say it is hard to do have something to lose. Like a monopoly or something. Or, could it be that they find it hard but others find it easy?

Critics of open standards do not like the equalizing effect of openness on the market. Open standards are best for business. Open standards are best for governments. Those two facts make open standards a panacea. Recalling for a moment that in Greek mythology, Panacea was the goddess of healing, we can simply say that open standards are pure goodness. Those who defend them are heroes. Superheroes, even. Little iron men and women working silently in small meeting rooms for hours, days, months, years. Oh, how about that definition? Here it the ideal one should strive for, which should not be controversial at all (but, alas, it is):

Open standards are developed in accountable and transparent processes that are open to all stakeholders; open to multiple implementation paradigms including open source; and standards text is available at no cost to all interested parties. Open standards, as a fundamental building block of innovation, become widely implemented because they are supported by all major stakeholders.

Do all standards out there fit this definition? Of course not, but there are a few, and they are important, especially for things such as the Internet and they will be important for cloud computing. Do all standards organizations work towards this ideal? Actually, no, but some do. Certainly, the best of them do. In fact, there are people in all standards organizations who have these ideals. Sometimes they dominate the discussion, sometimes not. But the idea of openness is firmly out there, and that idea is important. I will discuss my view of openness as a policy topic in a subsequent blog entry, stay tuned.

Despite the variety of standards out there, open standards are easy to recognize. Open standards, like HTTP, HTML, TCP/IP, XML and SQL, are evolved collaboratively by software engineers. These engineers typically come from software companies. They meet in standards organizations such as W3C, OASIS, and ISO. They talk. They disagree. They negotiate. Debates over how a standard should look could be lengthy. In the end, there is agreement, of sorts. This is important. Without agreement, no standard, at least no good standard. A good standard is not only open, but it is widely implemented. Without being put to use by the majority of the market, a standard is quite meaningless, and becomes just another weapon in the fruitless fight for final lock-in.

There are many striking aspects of standardization. The fact that competitors come to the table, spend endless hours together discussing and agree is only one. The fact that governments allow such discussions to take place is another. Standardization, when it works well, is good for competition.

Telling when a standard is not open is even easier. Those involved in the struggle to establish an open document format know that very well. After some debate over a company owned document format, the owner decided to submit it for standardization. In the end, it became an ISO-standard, but that does not make it open, or useful, or important. Being developed in a transparent manner does. Listening to input does. Taking on board input counts even more. Meanwhile, another format was also being standardized. After a long fight, only one truly open standard remains, ODF. That standard came late to the table and is currently working on becoming widely available. When it is, and it will, the debate will be over. Innovation will then occur on top of the standard, using the standard as a platform.

Despite the ease with which one may talk of openness and closeness, there is a continuum within open standards, from the completely open to the less open. There are times when the degree of openness of a standard is particularly important, such as in software-to-software interoperability. In fact, if interoperability is the main, agreed purpose of a standard, the best thing for all stakeholders, whether they be governments, multinationals, NGOs, non-profits, or individuals, is that the standard is licensed royalty free. However, these things have to be agreed in each instance. Once everybody agrees to follow certain rules, things flow smoothly. I tell that to my kids daily. Not that it always helps.

Some stakeholders are pragmatic. Whereas they may prefer open or closed standards depending on their business model, if a particular standard potentially is important to them, they participate in standardization regardless where it occurs. That is reasonable. It also explains why defending openness does not need to be a naïve position. Rather, it makes business sense. For most of us.


Post a Comment:
  • HTML Syntax: NOT allowed

Trond Undheim, Ph.D, Director of Standards Strategy and Policy at the Oracle Corporation, speaker, entrepreneur, blogger, and author, is one of the world’s leading experts on technology and society. LinkedIn profile


« April 2014