Red Queen Effect 2

Dave Bayless has responded to Critiques of the Red Queen Model, including my comments on this blog (Red Queen Effect, see also Rates of Evolution). 

Dave chooses to define innovation as "launching new products". Both John Hagel and I believe that there are other kinds of innovation that are important. But I have a more fundamental concern with Dave's definition - if I don't know exactly what counts as a "new product", then I don't know how to count them. If this year's model has a slightly faster chip than last year's model, or a brushed aluminium case, does that count as a "new product"? Let's say the iPod is a new product, but is the iPhone really a new product, or just a fancy redesign of an old product?

Lots of people in product development have a vested interest in labelling everything as "new improved". Pharma companies spend a small fortune looking for variations on existing drugs, so they can get patent protection for the "new" formula. But if you take these descriptions at face value, you get a fundamentally distorted view of the underlying technology change.

This is why I think we need a rigorous model of technology change, which handles some of the complications I raise in my previous blog entries.

 

---

Update March 2022

Following a brief exchange with Dave on Linked-In, I have looked at another of his earlier posts, which refers to the work of Professor Charles Fine on industry clockspeed. 

In a paper written in 1996, Fine compared product technology clockspeed between aerospace (commercial aircraft) and media (infotainment). He noted that Boeing was launching roughly two new products per decade, while Disney was launching one new product per year, and concluding that media therefore has a faster clockspeed than aerospace. 

There are several problems with this comparison. Firstly, it doesn't seem to represent a fair comparison between the true rate of innovation in the two industries. If Disney films are made to a formula, and the basic formula doesn't change, how much innovation does each new film represent? This comes back to the point I raised in my original posts above.

Fine presumed that "Disney's product development teams ... work on a cycle time geared to the time between new product introductions". I don't have any inside information on Disney, and it appears he didn't either but from my experience in comparable media organizations I should expect multiple development teams working in parallel, with the ones doing the most innovative work taking several years to complete something.

Furthermore, Fine also observes that other parties in the Disney ecosystem (notably the distribution channels) have a much faster clockspeed. So he notes that "clockspeed may not be well-defined at the industry level since many industries are composites of others".

For John Hagel and myself, the more interesting innovation in the media industry was not Disney or Pixar churning out blockbuster films, but these firms converting themselves to platforms. But does the platform itself count as a product?

 

---

Dave Bayless, Is the Pace of Business Really Increasing (7 September 2007), Critiques of the Red Queen Model (1 October 2007)

Charles H Fine, Industry Clockspeed and Competency Chain Design: An Introductory Essay (MIT Sloan WP 147-96, March 1996) 

John Hagel, Disney, Pixar and Jobs (3 February 2006)

Richard Veryard, Technology Hype Curve (16 September 2005), Disney, Pixar, Apple and Jobs (6 February 2006), Red Queen Effect (19 September 2005), Rates of Evolution (3 September 2007)

Rates of Evolution

A fascinating paper by Philip D. Gingerich shows how the observed rate of evolutionary change (measured in darwins), varies hugely according to the measurement context. (See table at foot of this post). In other words, the observed rate of biological evolution appears to be proportional to the proximity of scientists. Does a similar phenomenon apply to technological evolution?

(Note: I am not assuming that technological evolution is the same as biological evolution - merely that looking at one domain may prompt some interesting and important questions for the other domain.)

I have always been wary of the common belief that technological change is accelerating. I think this belief derives from a combination of proximity, selectivity and distorted perception. I think we can sometimes be disproportionately impressed by the glamour of recent technology, and misled by the commercially-driven measures of intellectual property (such as volumes of patent activity and product releases). We should also note that some commentators (such as industry analysts) have a vested interest in talking up the pace of technology change (Red Queen Hypothesis, Technology Hype Curve).

But consider these questions:

Did the lightbulb or bicycle change more

• between the years 1880-1900?
• or between the years 1980-2000?

Did the computer change more

• from 1950 to 1970?
• from 1980 to 2000?

It is certainly true that there have been huge numbers of small modifications to devices such as lightbulbs, bicycles and computers since 1980. There has also been a proliferation of variations and mutations. But will any of this innovation be remembered in fifty years time? From a historical perspective, this kind of detailed technological refinement (or even hyperactivity) may seem rather less significant than the initial burst of technical creativity when the device was taking shape in the first place.

---

Notes

Context	Timescale of Observations	Observed Rate of Change
Laboratory	1.5 - 10 years	60,000 darwins
Colonization studies	70 - 300 years	400 darwins
Post-pleistocene mammals	1,000 - 10,000 years	4 darwins
Fossil record	Millions of years	0.1 darwins

Philip D. Gingerich, Rates of Evolution: Effects of Time and Temporal Scaling. Science 14 October 1983: Vol. 222. no. 4620, pp. 159 - 161

Update: Professor Gingerich's book Rates of Evolution was published by Cambridge University Press in 2019.

See also: Delayed Success - Evolution (April 2023)

Bart Simpson Effect

In an earlier post on the Red Queen Effect, I said that the Red Queen has now become an icon for a certain kind of energetic innovation - believing the impossible, accelerating and relentless change.

In his post Bart Simpson and Sun Tzu on Technology Strategy, the Redmonk analyst Stephen O'Grady identifies another paradoxical innovation strategy, which we could call the Bart Simpson effect.

In Episode 802 (production code 3F23) Bart Simpson has been placed in a remedial class for some reason. He spots a paradox. "Let me get this straight. We're behind the rest of our class and we're going to catch up to them by going slower than they are?"

Some of the greatest minds of all time were considered dull at school. Winston Churchill was not clever enough to do Latin, so he had to spend more time polishing his command of English instead. [Wikipedia: Churchill] And Albert Einstein was considered a slow learner. "He later credited his development of the theory of relativity to this slowness, saying that by pondering space and time later than most children, he was able to apply a more developed intellect." [Wikipedia: Einstein]

Perhaps there are some things you can only get when you do things slowly. Part of the problem with the school system is an obsession with speed. [See my previous piece Systems4Success]. Most parents fondly believe their children to be above average in intelligence, based on fluency, literacy, numeracy, for example an ability to follow the rules of school mathematics and solve average problems slightly faster than the other children [Wikipedia: Lake Wobegon Effect]. The dreamer who spends all day thinking deeply about a maths problem probably isn't going to get top marks at school. But mathematical genius is about depth rather than speed, wondering what happens when you change the rules.

Stephen points to the folly of trying to catch the competitors on their own terms. "For all of the bubble era or Web 2.0 talk of 'disruptive' technologies, you'd be surprised at just how many vendors we speak with who anticipate closing marketshare or other gaps by simply outexecuting or outperforming their competitors."

If everyone is trying to be bigger and faster, then why not try smaller and slower. If everyone is trying to be smarter, why not try wiser.

And break the rules.

Red Queen Effect

The Victorian mathematician and fantasist Lewis Carroll used the character of the Red Queen to parody grotesque forms of reasoning and energy: believing impossible things (before breakfast), running to remain stationary.

Believing the impossible, accelerating and relentless change - these are now among the totems of innovation. The Red Queen has now become an icon for a certain kind of energetic innovation. There is a good summary on the Farnham Street blog (October 2012).

There are several other related terms.

• Running Up the Down Escalator. In 1993, the SEI published a video with this title, presented by risk management consultant Bob Charette. (I haven't seen this video, but I've seen other materials derived from Charette's work that portray the staircase as a corkscrew or helix, spiralling downwards as you try to run up.) In one version of the story, the acceleration control is at the top, so those who are most successful at running up the down escalator get the chance to speed up the escalator, thus increasing the gap between themselves and their competitors (Critical Chain, June 2008).
  
• Continuous Bootstrapping / Perpetual Bootstrapping.
  

Dave Bayless (Evergreen Innovation Partners) has recently propounded a model of accelerating product innovation, which he has also named after the Red Queen. He points out that the compound effect of a 10% annual acceleration in product innovation results in a halving of product life cycle duration every seven years See his blog and video on Innovation, Clockspeed and the Red Queen Effect. See also comment by John Hagel: Product Innovation and the Red Queen Effect.

While this model has some intuitive appeal, there are some interesting complications (or asymmetries).

1. The product is not the technology. A product may be composed from a large number of components, each of which may be subject to technical innovation. Product innovation is not a simple linear function of technology innovation; a product lifecycle can be extremely short, but most of the underlying technology may be moving much more slowly. Or vice versa.

2. The adoption is not the innovation (as John Hagel points out). Innovation includes process innovations as well as product innovations. Hagel suggests that "rapid incremental process innovation combined with aggressive leveraging of third party resources may in fact hold the key to diminishing, if not overcoming, the Red Queen effect."

3. The "device" is not the "commodity". Small incremental changes in the product may result in radical changes in user experience and practice, while radical substitutions on the technology side may simply be experienced as slight improvements in service cost and quality. For example, the consumer experience of innovation in automobiles or electronics or mobile telephones is not based on the rapid turnover of model numbers and versions, but on major (and relatively infrequent) step changes in functionality and performance.

A rigorous model of technology change must articulate these different layers clearly.

---

See also further posts on the Red Queen Effect.

 

Footnote added later

Charette's Risk Escalator is described in Tom deMarco and Tim Lister, Walzing with Bears: Managing Risk on Software Projects (Dorset House, 2003) p 12. I also found a video of Bob Charette presenting the model at a conference in 2012 https://vimeo.com/43479018. The original source appears to be R. Charette, Up the Down Escalator: Managing Risk in an Uncertain World (Part 1), (Software Management, Oct. 1993).

Technology Hype Curve

The Gartner Group produces a large range of technology trends and predictions, based on a so-called Hype Cycle model. (The term Hype Cycle implies that things come round again. But the model is not cyclic, so it is more accurate to refer to it as a Hype Curve model.) I have just been looking at a Gartner document that includes curves for 1995 and 2005.

Here are some clues about the degree of rigour and empirical support underpinning Gartner's analysis.

Clue Number One: All technologies appear to have the same eventual outcome.

Clue Number Two: All the points are perfectly on the line. To a scientific mind, this is a strong clue that the coordinates are not based on any real objective measurement, and that the curve itself is not subject to scientific investigation or calibration. The curve itself is based on a standard engineering pattern.

Clue Number Three: The shape of the line has not altered (or accelerated) in ten years. But surely we should expect a shifting (shrinking) curve? For one thing, many technology narratives suggest that the half-life of new technologies is getting shorter. (This is sometimes known as the Red Queen Effect.) Furthermore, we might expect the quantity of attention received by each technology to be affected by the number of technologies competing for attention - and to the extent that this is increasing, the quantity and/or duration of hype might be reduced - in other words the hype curve getting steeper. (Surely technologies used to remain at the top of the hype curve for longer than they do today?)

Rather than just 1995 and 2005, it would be useful to see the whole series - so we can pick out those technologies that have gone faster or slower than Gartner had expected. Interesting that Gartner has chosen not to include this information in the self-congratulatory document I have seen.

The Aye Conference has a good discussion on the Hype Curve (September 2003). And in It's Not A Cycle, Tom Graves lists some other non-cyclic cycles. See also Wikipedia: Hype Cycle.

Related posts: Technology Hype Curve 2 (July 2009), Technology Hype Curve 3 (August 2009), Technology Hype Curve 4 (August 2015). For additional posts on technology hype in general, see my Software Industry Analysis blog.