Bigger Standards Required

It’s not just about the chip or even the device. It’s the world in which they work.


One of the great benefits of standards in the semiconductor world is that they make it easier to move forward and speed up technologies and markets—but not always. They are revered by semiconductor companies, despite the fact that they sometimes outlive their usefulness and need to be folded into other standards, and they are recognized as the best way to get things done across a complex, far-flung and exceedingly complex ecosystem.

Still, there are good standards and bad standards. The good standards are properly timed. They arrive in the market when there is enough visibility to see what’s needed and how new technologies will evolve. And they leave the door open to innovation and change, and possible replacement with a new standard as it is needed.

Bad standards arrive far too early, and often are supplanted with other standards that create widespread confusion. The battle over UPF and CPF is a classic example of this, where two competing standards created havoc before ultimately and painfully being merged into IEEE 1801. And there are some standards that have sat around for years before they have been revisited because no one was quite sure what to do with them.

This, of course, is in no way a reflection on the people who create those standards. The amount of time and dedication required to implement standards is huge, and there should be badges of honor awarded to the people who run standards efforts and those who participate in them. These are the unsung heroes of any industry, and in semiconductors there are many of them who never get any recognition.

Nevertheless, in a connected, global marketplace where companies rise and fall much faster, standards need to be connected on a much broader and bigger scale. What has worked well, and sometimes not so well, in individual groups now has to be expanded. So while it may be important to understand how IP comes together on a chip, it’s also important to understand how that IP interacts with the outside world, how much power can be lost from constantly searching for a signal or a bad connection or using the wrong protocol, and how devices will interact with other devices that may or may not yet exist.

Standards are a basis for getting things done more efficiently and with a fair degree of certainty about the outcome. But increasingly standards will have to work with other standards that may or may not have anything to do with a particular chip, device or even a vertical market. This is a whole new way of looking at standards efforts. Viewed locally, this becomes a much more painful process. Looked at globally, it opens the door to a level of innovation, creativity and connectivity that was never even considered in the past.

The increasingly connected world, whether we call it IoT, IoE, or some other term, means that standards groups also have to be connected on a broader scale—across devices, across markets, and across geopolitical boundaries. Recognizing this need is the easy part. Implementing it will be much, much harder, and it will require a new level of commitment from across ecosystems, not just one or two segments within them.

One can only imagine how the next round of standards meetings will go.