IT Innovation = (I) ? + (II)

Seventy years ago last December, the transistor was invented by Nobel Prize winners William Shockley, John Bardeen and Walter Brattain.?The original transistor, which was about the size of a small dinner plate, was probably the most influential invention of the 20th century and possibly of the last few hundred years.

The new A11 chip in the latest iPhone has some 4.3 billion (yes billion) transistors packed onto it.

Can anyone imagine trying to make a phone call today while balancing some 4 billion dinner plates?

Something quite magical has happened in the last 70 years to turn the scientists’ dinner plate into a hand-held computer of stunning power, only imagined half a century ago. That magic is often termed ‘innovation’, or the transformation of a novel idea into useful artefact(s).

Truly useful, and widely-used, innovations are not merely bright ideas. It usually takes a (very) long time to popularize an innovation and very often the end-product may be quite different to the initial inventive idea (the dinner plate to the smartphone).?

But, there is nothing smooth or even inevitable about the adoption of technology innovations.?The guru of innovation, Everett Rogers[1], pointed out that many (probably most) new innovations fall by the wayside, because they turn out not to be as useful as first thought, or are sometimes superseded by other innovations, or so-called “technology disruptors [2]”.

An innovation by itself will not succeed unless it first fills a real need for consumers and, importantly, industry processes are adapted to utilize the innovation.?Innovation then is not merely invention, but invention plus implementation[3].?

The British inventor Sir James Dyson developed ?a formula for Innovation which stated that

?Innovation = Creativity + Iterative Development

Dyson’s formula describes the never-ending(?) process of ‘incremental improvement’ during implementation of an invention, constantly inventing new products and, importantly, new ways of doing things with the same basic inventions.

But, what are the factors that drive one IT innovation to be successful as against another equally novel idea that peters out? The following formula identifies some of the key factors that together make a truly successful IT innovation[4]

IT Innovation = (I) ? + (II) =

?(Invention * Integration * Industry standards * Infrastructure * Investment) + (Incremental Improvement)

This formula shows critical factors/prerequisites for successful adoption of an ‘invention’, and as implementing the necessary factors takes time and financial investment, it usually takes years, or even decades, for any IT innovation to fully flower.

Note that all of these factors must be present to create and sustain a successful IT innovation that is widely-used. The role that each of these factors play in making a useful IT invention is described below but for any particular innovation the importance of these factors may vary significantly.

Invention

Invention is the discovery of a novel idea, which may be a physical object (such as a transistor) or a concept (such as Google or Blockchain).?The discovery may result from a long-term research effort or from a single person considering a common problem. Inventions are the ideas that remain to be turned into something genuinely useful.

Examples of inventions that have helped to create very successful IT innovations include: the discovery of giant magnetoresistance [5] (GMR), which has enabled vast amounts of data to be stored onto computer hard disk drives (HDDs); the discovery, in the 1960s, of how ‘fibre optic’ cables can be used for very high-speed data transmission between computers[6]; the invention of the first successful digital imaging technology using a charge-coupled device (CCD) an integrated circuit?which is ubiquitous in modern cameras and smartphones[7];?the invention of the lithium-ion battery[8]; the invention of Wi-Fi[9], used to connect devices to the Internet using wireless signals; the liquid crystal display[10] (LCD), used in TVs and smartphones; the cellular/mobile phone network[11]; and so on and on and on.

These inventions and dozens of others are used in billions of devices around the world everyday – so how did they become so widely-used, many decades after their discovery?

Integration

Raw inventions, such as a lithium battery or a fibre optic cable are rarely useful in themselves but when they are put together, or ‘integrated’, with other inventions they can become very useful indeed.?The inventions described above, GMR, HDDs, CCDs, and Wi-Fi plus, of course, billions of transistors on integrated circuits (ICs) are packed (along with dozens of other inventions) into a modern smartphone.

Innovations become useful when inventions are integrated into our daily lives, usually being incorporated into devices we already use, such as phones.

While Invention is mainly a research/intellectual activity, Integration is an engineering discipline. It is the difficult job of the integrator to fit multiple inventions into a new form and importantly to enable the inventions to work together to perform new tasks and provide new capabilities. And, to be successful, integrators must do that within pretty stringent physical space, power and cost constraints.

For example, the CCD imaging device in a smartphone camera must be able to store the image on internal HDD[12] storage, display it on the phone’s LCD screen and then send a copy of the image to friends/colleagues over the Internet using wireless communications. It is marvel of hardware (and software) integration.

But integration is more than connecting many different components together, it is also about putting more capabilities into smaller and smaller physical spaces. This is best exemplified in what is known as Moore’s Law, a prediction made in 1965 by Gordon Moore, one of the founders of the Intel corporation, that the number of transistors on an integrated circuit (IC) would double roughly every two years. This prediction of exponential growth in effective computing power on an IC ‘chip’ has held for over 50 years and although the rate of growth has slowed somewhat, due to physical constraints, the growth rate remains spectacular if no longer exponential.?Which is why today, we can fit over 4 billion transistors into the latest smartphones.

While inventors may win the Nobel, arguably the real stars of Innovation are the engineers who integrate the inventions and make them useful and then ever more useful year every few years (see Incremental Improvement)

Industry Standards

If an Apple iPhone could not communicate with a Samsung Galaxy smartphone or with an IBM computer over the Internet, it would be pretty useless in practice.?But they can and do communicate with one another and also with Goggle, Amazon, Facebook and with millions of other devices and software programs.

They can do this because, when they need to, they communicate using Industry Standards.?

There are industry standards everywhere.?For example, Wi-Fi communications are based on a family of standards, IEEE 802.11, agreed by the Institute of Electrical and Electronics Engineers (IEEE) while fibre optic communication conform to another of the family IEEE 802.3. Standards build on standards. And, in order that integrated circuits can be integrated into many different types of devices (smartphones, computers, communications devices) there are standards for the dimensions, connectors and power usage of different types of IC. Likewise, there are standards for using the Internet from the ubiquitous uniform resource locator (URL), such as www.dhirubhai.net, to the HTML[13] language used to create web pages.?

?Without industry standards, innovations would not be able to expand beyond very narrow confines.

However, innovations can emerge from industry standards. One spectacular case is that of the universal serial bus (USB) which was developed in the 1990s and was then disseminated royalty free by Intel[14].?USBs are used as mechanisms to connect different types of devices (from keyboards and mice to high-speed video cameras) to a PC or smartphone. It is simplicity itself - just plug and play[15].

Another example of highly developed standards is that of ‘location services’ which are the applications that allow the physical location on earth of a smartphone or computer to be determined, at least to a rough level of accuracy. Standards bodies, such as ISO (standard 19133) and the Open Geospatial Consortium (OGC) have developed standards to allow the location of a device to be determined from sources such as the Global Positioning System (GPS), the mobile telephone network, and local Wi-Fi.

Uber did not invent taxis, nor develop communications technologies but integrated location services standards (for pick-up and destination addresses) with the Internet to create a new type of taxi service.?

Innovation then does not have to begin at invention but can start by integrating industry standards!

But to succeed Uber needed more than just standards, then needed the infrastructure to allow passengers and drivers to communicate and a standardized payments infrastructure to allow safe fare collection.

Infrastructure

In order for an innovation to become truly successfully, it must be ‘diffused’ throughout the economy and be taken up by businesses and people around the world. There must therefore be some form of common infrastructure that allows an innovation to be ‘plugged in’ and then used easily and efficiently without undue costs.

According to the Cambridge Dictionary an infrastructure comprises “the basic systems and services, such as transport and power supplies, that a country or organization uses in order to work effectively”. To country or organization, it is worth adding for the purposes of the discussion here, “or people or technology or legal jurisdiction”. The Internet could, of course, be described as such an infrastructure.

A key example of technology infrastructure is the mobile/cellular telephony network which allows telephone devices to communicate to a nearby ‘mobile base station’ using radio beams, i.e. wirelessly.??For example, there are estimated to be somewhere around 300,000 base stations of varying sizes in the USA (and the number is growing). Such base stations have been built at a rapid rate over the past decade around the world by telephone/network companies and are able to work “effectively” because most are based on, or moving to, what is called the 4G (fourth generation) mobile communication standard, which allows not only voice but high-speed data transmission.

Infrastructure is expensive to build, for both public and private entities, so a fairly well-defined need must be apparent, or at least highly likely in the future. There is a chicken and egg situation with infrastructure – which comes first - the need or the infrastructure??In truth it a bit of both, there must be a need to start to build sections of an infrastructure but in turn once built, new uses and users will come.?The building of infrastructure is evolutionary, usually starting small then expanding outwards as demand increases.

?A classic case of this evolution is the current state of the charging infrastructure for electric vehicles (an innovation based on the Lithium battery). Electric vehicles (EVs) can be charged overnight using standard home power sockets, slightly adapted, but current batteries provide only limited driving range, enough to drive locally but not sufficient for long distance travel. What is needed then is an infrastructure, similar to that of the decades’ old gas station infrastructure. Similar to gas stations there needs to be a national network of charging stations to support long distance travel by EVs.?This is already beginning to happen and in ten years such stations will be (almost) everywhere and new technology innovations will undoubtedly create even more opportunities.

Innovation then can start by building infrastructures!

Investment

Innovation cannot take place without investment, but investment in innovation is far from a simple activity.

Inventions need investment, to undertake research, develop prototypes, test, refine and, only if successful, to package into a useable ‘thing’.?The income (return) from the investment in an invention usually comes in the form of usage fees if patented and /or buyout fees if sold to a larger company. Investment in inventions is a highly speculative with high risks and hopefully/possibly high returns.

Integration similarly needs investment, to undertake research, develop prototypes, test, refine and to package into a saleable product.?In addition, investment is needed to purchase and/or /develop the expensive facilities needed to manufacture the final product(s). The income from investments in integration comes primarily from sales of the final product which has many risks but also, if successful, a ‘long tail’ of income.

Industry standards needs investment, to undertake research, develop proposals and agreeing standards (which are the usable products). Any income from investing in developing industry standards then is highly speculative and comes from future sales of products yet to be developed. Investment in standards then are investments in the (far) future.

Infrastructure needs enormous, multi-year investment in designing and building physical infrastructure. Income from investing in developing infrastructure comes from future fees for using the infrastructure, and at the outset is very uncertain.

Investment in any innovation has a different profile as the innovation grows through various stages, and in each stage, is not only somewhat speculative but also increasing as the shape of the innovation evolves.?Furthermore, it should be remembered that a proportion of innovation will not take hold and these investments will be lost, unless the innovation can be re-purposed.

It is important to note that those who gain most income from investments in innovations are not necessarily those who invested most.?It is possible for companies to leverage the inventions and infrastructures developed by others to piggyback and grab a disproportionate share of income (e.g. Amazon, Facebook, Uber etc.).

Incremental Improvement

Successful innovations cannot sit still, but must be constantly enhanced because the environment, business and technology changes. Innovations do not end when they are successfully adopted, the next chapter must begin immediately.

The latest: iPhones are numbered 8; the Intel Core chip is the 7th generation; Windows operating system is numbered 10; mobile network standard is numbered 5G; the IOS operating system is numbered 11; the latest Android operating system is numbered 8; and so on [note this list will be out of date soon].

The key point is that all of these innovations have gone through multiple iterations of incremental improvements, most successful but some, such as Samsung Galaxy 7 less so (catching fire). In fact, for many of these innovation, improvements are planned in advance with for example major releases of new chips being scheduled for roughly every two years, and minor improvements annually. The next generation of mobile networks, 5G, is scheduled to be operational in the early 2020s and telecommunications industry bodies are already discussing the properties of 6G which, in line with Cooper’s Law, will increase data rates immeasurably in the late 2020s.

Feedback in Innovation

Innovation is not a linear process, with a predetermined path from nothing to something. All of the factors in the innovation equation above have to be present, but their relative importance, and the time and investments needed to get to the next stage are very uncertain.

There is also feedback in the process as shown in the diagram below. For example, the process of integration will naturally feedback ideas for new inventions, concerned mainly with engineering and materials science (e.g. cooling materials for smaller and smaller ICs).?Likewise, industry standards will inform and eventually drive integration and, further on in the cycle, incremental improvements will influence the design of infrastructure and so on down the chain.

As noted in the diagram, innovators can bypass the early phases of innovation and build upon the work and discoveries of others.

An example of feedback in IT Innovation is the credit card, which was developed first in the 1950s by companies such as American Express, before being integrated into the banking systems from the 1960s onward. Credit cards are accepted by merchants around the world, because they are based on global security standards. Payments by credit, and innovations such as the ‘debit card’, are firmly based on the world’s telecommunication and banking infrastructure and are constantly being improved, such as using smartphones as to make credit and debit payments. Outwardly, the credit card remains an oblong piece of plastic, but has today a powerful ‘computer chip’ embedded in it, which is many times more powerful that some computers of the 1960s.

So, what will make an innovation successful?

If the author had the definitive answer to that question, he would be doing it, not writing about it!?Few people[16] take an innovation from idea to incremental improvement and back and forth again.

One of the few such innovators of recent times, is Sir James Dyson, best known as the inventor of the ‘dual cyclone’ bag-less vacuum cleaner. Dyson built the better proverbial mousetrap and integrated it with other mundane components into a usable and popular vacuum cleaner.?The infrastructure for this innovation was already well-established, i.e. home electrical power systems.?But Dyson has not stopped improving, integrating the firm’s inventions into, for example, innovative personal cooling systems (fans), hair dryers and also commercial hand dryers.

Nor is necessary to start at the beginning by inventing something new and, in fact, many very successful innovations enter somewhere in the middle of the innovation equation.?

For example, Amazon did not invent book-selling nor even the book depository.?However, using the existing infrastructures of the Internet, the postal service and banking payments system, Amazon integrated the warehousing and selling of books into an attractive purchasing proposition and has expanded that business model beyond books to many other goods.?This innovation has been copied very successfully by, among others, Alibaba.?

Facebook did not invent gossip, but, like Amazon, used the existing infrastructures of the Internet plus the technology of ‘big data[17]’ to collect and distribute billions of pieces of gossip every day.?And Facebook continually woks on ‘improving’ its offerings, often by buying small start-up innovators, such as Instagram and WhatsApp.

When considering an embryonic innovation (or even a well-developed one) the question ‘when will it be successful and how much investment is needed’ is hard/impossible to answer but it is worth asking questions that will give an idea of the time and investments needed.?

?If an invention can be integrated easily into existing innovations, time and investment will be considerably reduced.?For example, a camera with a lens greater than a few centimetres, no matter how powerful, will not easily integrate directly into a smartphone, so will have to remain as a separate device possibly communicating with the phone using an industry standard such as Bluetooth.

Likewise, an innovation that already uses industry standards can be brought to market faster than one that will require new industry standards to develop, such as for example, quantum computing or Artificial Intelligence[18] (AI). Of course, for some innovations, effective industry standards can be imposed by a large company, such as Intel or Oracle[19].

If an innovation is based on an existing infrastructure, such as the Internet, it will have a major advantage as regards adoption. Lack of an infrastructure, however, such as charging infrastructures for battery or hydrogen cars, will slow down the widespread acceptance of any innovation.

Note it is also possible to innovate in the Incremental Improvement phase, such as by building a better customer experience.?For example, as with PayPal, which provided an easy to use front-end to various credit and debit payments systems, or a hotel booking application, such as Booking.com.

So, it is not surprising that some of the most successful innovations, such as Amazon, Facebook, Booking.com etc. are based on existing infrastructures, allowing innovators to enter late into the IT Innovation equation.

Steve Jobs, one of the great innovators, is quoted as saying “good artists copy; great artists steal”! Maybe all innovation involves a little theft?

?

?

[1] See Rogers E. M., 2003, Diffusion of Innovations 5th Revised edition, Simon & Schuster International, New York

[2] Christensen C. M., 2000, The Innovator’s Dilemma, Harvard Business School Publishing, Boston, MA

[3] Rogers used the term ‘adoption’ for ‘implementation’

[4] Note other non-IT innovations, such as in medicine, may not be driven by this equation

[5] In 2007, Albert Fert and Peter Grünberg jointly received the Nobel Prize for physics for their discovery in 1988 of giant magnetoresistance (GMR), paving the way to what has become known as ‘big data’.

[6] In 2009, Charles K Kao was awarded the Nobel Prize for the discovery of the properties of fibre optic cabling that is used in the vast communications networks that underpin the global Internet.

[7] In 2009, Willard S. Boyle and George E. Smith were awarded the Nobel Prize for their 1969 discovery of CCDs.

[8] Unfortunately, John B. Goodenough has not yet been awarded a Nobel Prize for his invention of the Lithium Ion battery, which is ubiquitous in modern technology.

[UPDATE In 2019, the Nobel Prize in Chemistry was jointly awarded to Dr Goodenough, Stanley Whittingham and Akira Yoshino]

[9] In the 1990s, a team, headed by Australian radio-astronomer Dr John O'Sullivan, while working at the Commonwealth Scientific and Industrial Research Organisation (CSIRO,) developed Wi-Fi, for which CSIRO holds the patent.

[10] The LCD was invented in the 1960s by George H. Heilmeier. Other technologies such as light-emitting diode (LED) are beginning to replace LCD in some applications. Credited with the invention of ‘red’ LED in the 1960s, Nick Holonyak Jr, has controversially not been awarded a Nobel Prize.

[11] The concept of the mobile/cell phone was invented in the 1970s by Martin Cooper who also articulated Cooper’s Law that “the maximum number of voice conversations or equivalent data transactions that can be conducted in all of the useful radio spectrum over a given area doubles every 30 months”.?This law, similar to Moore’s Law, has given rise to the fantastic expansion of mobile telephony around the world.

[12] HDD is gradually being replaced by faster but costlier solid-state drives SSD based on integrated circuits (and transistors),

[13] The Hypertext Markup Language (HTML) is used to format ‘pages’ for viewing using a web-browser, such as Firefox or Internet Explorer. To see what HTML looks like, browsers have a ‘view source’ option somewhere in a toolbar

[14] The concept for USB was developed by a team led by Ajay Bhatt, Intel’s Chief Systems Technologist,

[15] Provided the computer has the software (often called ‘drivers’) to recognize and control a particular device

[16] Few people, except the Moores, Dysons and Coopers of this world, who also build the teams of talented experts needed to make innovation happen, again and again.

[17] Big data is, in turn, built upon the invention of giant magnetoresistance (GMR).

[18] Note standards are beginning to emerge in AI for issues such as ‘ethical standards’ for development.

[19] For example, the Java programming language was developed in the early 1990s by a team led by James Gosling at Sun Microsystems, a company that was subsequently acquired by Oracle, which continues to support, and continually improve, Java as ‘open source’ software. Java, now version 9, is a de-facto standard for some Internet applications.