A Post by Michael B. Spring

(A list of all posts by M.B. Spring)

How do Standards Come About?(February 15, 2010)

With the emergence of new markets in Europe and Asia, there has been a renewed interest in standards. Part of the reason for this renewed interest in standards is that there are new players and standards are being used in new ways. Specifically, Europe and China are playing a bigger role in the development of standards. Simultaneously, standards are being used more frequently as a mechanism that impacts trade. One question that has been raised is how standards come about.

One answer to this question is that they just appear. It is clearly not the case that they magically appear. Perhaps it is better to say that there is no formal standardization effort. As an example of this, consider Microsoft Word and Adobe PDF document formats. Obviously both Microsoft and Adobe would like to dominate the markets within which they operate and they work hard toward that goal. If Microsoft is successful, which in large part they have been with the Office Suite, Word becomes a dominant format for the exchange of documents. It emerges, without concerted standardization effort, as a standard to which everyone adheres.

Under this condition, nobody decides anything.  The standard emerges from market conditions. We just happen to notice one day that people nod as a way of greeting, or people decide to call the process of making a copy of something "Xeroxing". De facto standards vary in their saturation/adoption/penetration and in their longevity.

The second way a standard comes about is by intent. At the most global level, An (A)gent decides that some (G)ood will accrue to some (C)onstituency if some (P)rocess or product is codified. As one might quickly surmise, each of the factors may be further and better defined. Under the intentional condition, (A)gents with the "right" to take action (government, business, professional societies, etc.) will be more or less powerful, more or less licensed, etc. Similarly the decision that something is (G)ood will be more or less significant, more or less shared, e.g, fewer people will die, a business will grow, people will be better served, etc. It is also true that the (C)onstituency may be of different sizes(world, nation, organization, family, etc) and with respect to the Good, more or less homogeneous, e.g. occupational health may be a less widely shared concern globally than prenatal care. Finally, the (P)rocess or (P)roduct may be more or less amenable to standardization, be in a quasi standard form already, etc.

Determinants of standardization can then be derived at various levels of sophistication and complexity dependent upon variations in the values of A, G, C, and P. Consider a couple simple examples:

If the active "(A)gent" is questioned as the legitimate authority, standardization becomes more difficult. Consider just two examples. When does the government have a right to define standards? While we might agree that the Federal Government has the right to define standards for interstate commerce, we might not agree that they have the right to define cryptography standard for banking. This makes it more difficult to impose dejure standards. Similarly, if 5 companies make product X and 3 want to standardize and 2 don't, the authority may be lacking.

As the "(C)onstituency" gets bigger, standards become more difficult. It is more difficult to create a world wide standard than a National standard than a State standard etc. It is pretty easy to define an organizational standard, etc, etc. Even when a global standard is desirable, national interests, commercial or military, may mitigate against it.

The magnitude of “(G)ood” derived from a standard is also a significant force. Few would argue against a standard for automobile safety or fire protection. It is less clear that there is agreement about the good derived from a ban on smoking in public places. The benefits that accrue in a business sense are less clear to the general public, but may be easier to quantify between corporate entities involved.

Finally, the complexity of the “(P)rocess” (or product) also impacts the ease of standardization. It is more difficult to standardize larger and more complex products and processes than it is to standardize smaller systems. Indeed, many larger systems are layered so that multiple standards can more easily be developed.

All of this said, we are still talking in vague generalities about these factors in isolation. There are interactions among the factors and they are all in play at the same time. At a first cut, how might we envision these relationships? Without endeavoring to quantify any of the values, the general nature of the formula for the Difficulty of Standardization, or DOS is along the following lines (where w,x,y,z are all values greater than +1.0):

DOS = (w*C + x*P)/(y*A + z*G)

Thus, when the size of the constituency grows or the complexity of the product or process increases, the difficulty of standardization increases. In contrast, as the magnitude of the social or commercial good increases or the legitimacy of the agent of standardization increases, the difficulty of standardization decreases. Establishing coefficients and units of measurement is no trivial task, but I believe the shape of the equation has some validity.