Who remembers the proprietary home network products that came before WiFi?
Who remembers the proprietary wireless mobile phone headsets that were sold before Bluetooth became ubiquitous?

A proprietary platform has a single provider that solely controls its technology, for example, Microsoft Windows, or Google search. Proprietary platforms can be hugely successful, but fail dramatically in typical consumer electronics products.

Before WiFi you could buy a proprietary base station with matching dongles. If you installed this at home it worked fine, but you could not rely on using your computer anywhere else. WiFi changed that. With WiFi you have network access everywhere, and overnight the proprietary products disappeared from the market.

WiFi was a blessing for notebook manufacturers. It did not make sense for them to integrate proprietary network technology. The WiFi standard made integration commercially attractive, and that made WiFi even more popular.

The California Management Review has an excellent paper by Prof. Thomas Eisenmann on the benefits of creating a so-called “shared platform” (California Management Review Vol. 50, No 4, Summer 2008). With a shared platform, such as Visa, DVD, or Linux, multiple firms collaborate in developing the platform's technology then compete in offering users differentiated but compatible versions of the platform. The article examines factors that favor proprietary versus shared models.

Qi is the shared platform for wireless battery charging. More than 100 companies support Qi. Just like WiFi is a shared platform managed by the many members of the WiFi Alliance.

Qi is about compatibility and design freedom.
 

Compatibility means that, when you see the Qi logo, you know "this will work".  A mobile product carrying the Qi logo will be powered by all wireless power transmitters that carry the Qi logo.
 

Design freedom means that manufacturers are not limited in the design of products. Manufacturers must be able to innovate and differentiate their products.
 

Compatibility is not a natural friend of design freedom. It is really easy to design a wireless battery charger that does not charge a Qi-compatible receiver. It is much harder to make sure that a new transmitter design will work flawlessly with all Qi receivers.
 

The Wireless Power Consortium has a policy for dealing with the trade-off between design freedom and compatibility:

1. Compatibility is king. The message of the Qi logo is: "it will work".
2. Design freedom for power receivers has priority. We give receivers priority because Qi adoption in mobile products drives the market for wireless chargers.

The first version of the Qi specification delivers the intended design freedom for receivers. You can see that already in our demonstrations: Qi receivers have been built into batteries, phones, Bluetooth headsets, sleeves, ...
 

There are more requirements for transmitters, but the range of possible transmitters is significant. The first version of the Qi specification allows four types of transmitters. Two multi-coil arrays, a moving coil, and fxed single coil transmitters. And each type allows different shapes and sizes where manufacturers can make their own trade off between cost, user experience, and design. That is already quite a bit of design freedom.
 

We are now increasing the freedom in transmitter design even further. Members of the consortium have proposed new transmitter types. These new types are added to the specification when we are certain that the new transmitter will not break the compatibility rule. The procedure for adding a new transmitter looks straightforward but is quite a lot of work:

1. The new transmitter must pass the certification tests described in part 3 of the specification.
2. The new transmitter must meet the performance requirements described in part 2 of the specification
3. The new transmitter must work flawlessly with all previously certified receivers.
4. The new transmitter must not reduce the design freedom of receivers.

Step 4 of the procedure is the tricky part. How can you be sure that some future receiver does not break compatibility with this new transmitter type? Interoperability with the previously certified receivers and conformity with the certification tests (Steps 1 & 3) are a good indication, but not enough. In Step 4 we challenge the reviewers of the proposed new transmitter type to come up with exotic receiver designs that will work with the current transmitter types but fail with the new transmitter type. When the reviewers fail, the new transmitter passes.
 

When can you expect new transmitter types in future updates of the Qi specification? I will announce that when the update is ready!

Do you want to add a transmitter design of your own? Follow the procedure! Members of the wireless power consortium can find a more detailed description of this procedure on the member's section of the website.

The Qi standard is a document that describes how to design power transmitters and power receivers. It is a rather complicated text, more abstract than your typical product design documentation. It is more abstract because an interoperability standard tries to maximize the design freedom for product developers. Not everything is fixed and that can make it difficult to understand.

Our goal is interoperability. A power receiver must work with all power transmitters. Interoperability follows automatically when the product developer follows the instructions in the standard. Right? Well, ... that is the theory. In practice it does not work like that.

Even a properly reviewed specification can be ambigeous. Product developers can interpret a text in ways the editor never considered. And product developers can make mistakes. The result is a product that might not work with other products.

Publishing the documentation of a standard is not enough. To achieve our goal we also need:

* A set of test procedures that help developers to determine whether the product works correctly.
* A trademarked logo and a trademark license agreement to limit the use of the logo to products that work together.
* Certification by an independent test lab before products are allowed to carry the logo.

The test procedures must cover all aspects of the standard. Everything that could possibly go wrong must be verified. It is interesting to note that, for standards like Qi, the documentation of the test procedures has more pages than the standard itself!

Our customers will see the Qi logo and know: this product works with my other Qi products. That promise holds if the logo is used on properly tested products. No Qi logo on products that don't work! To protect Qi's promise, the logo is trademarked and you need a license to use the logo. That license makes the use of the logo dependent on certification by an independent test lab.

Publishing a standard is not enough. Without test documentation, certification services, and a logo license agreement, interoperability is an illusion.

Wireless power is adopted easily in some applications. Toothbrushes, for example, and in UV lamps for water desinfection. But penetration in offices and living rooms is limited today, even though consumer tests show that users like wireless power very much.

The problem with wireless power in offices and living rooms is that the perceived value increases with the number of different products that can be charged wirelessly. In other words, only few consumers are prepared to pay US$ 50 for a dedicated charger. That perception changes when a wireless charger services many products.

In economics this effect is called a 'network externality'. Products with a strong network externality benefit from co-operation between companies. Consumers see more value when many companies introduce products that work together. That's why companies work together on the standardization of wireless power. They all benefit from increased sales.

Products with strong a network externality are hard to introduce in the market. On the other hand, once successful, the interface tends to stay around. New products benefit from interoperability with the installed base, and the perceived value of alternative solutions remains low.

When network externalities are strong it is, therefore, risky to invest in products that don't work with the established standard. It can be surprisingly expensive to go against the flow.