The difficulties caused by the green frog

Having had many different experiences of explaining the concept of the green frog and Hilbert Solution Space ­ in one to one discussions; in corporate meetings; in technical workshops and in college lectures ­ I knew I'd get a range of different reactions when I sent out chapter 8 to the reviewers in the virtual cafe. I wasn't disappointed.

Many immediately recognise Hilbert Space as a hugely versatile mental modelling environment that could be used as an approach to think about all kinds of problems. Simply by choosing different kinds of dimensions to work in, any kind of bottom up design process can be visualised. Mostly, artists and people working in creative fields found the concept easy to understand. This was exemplified by Blane Savage, a multimedia designer from the UK, who wrote to a cafe group in the virtual cafe explaining why he'd deliberate rub out a successful individual feature of his work in order to go beyond his imagination:

Personally for me this process is more to do with letting go of some small piece of beauty (which when looked back upon is flawed beauty) to risk capturing a greater whole - it is also about pushing your work to destruction (past the instinctive stopping point of a piece) to expand the envelope of your understanding.

How can you develop your self and your art unless you grow beyond your present boundaries. Creation of art is not about methodical stages - it is about starting with nothing and developing in a very 'design up' way. My experience and that of other designers I have spoken to, is that creativity is a difficult process - it does not come easily. Failure is constantly there and success is almost painfully achieved with a leap of inspiration - a single successful feature is a short term resting place to hide in before moving on. If you safely stay and develop around that point you will never achieve what you are trying to find. .....

......do programmers ever find the code that is most eloquent and that they are most proud of is actually stopping the program running once it is complete?

Blane

Others, couldn't seem to connect it to any real life experience. Another multimedia designer from France, provided this pragmatic assessment of the green frog and Hilbert Solution Space paradigms:

A green frog with 12 legs, a blue one with one leg, or a grey one which becomes a grey box... I give up, Hilbert got me....

Chapter 8 make me sleep... maybe, (or for sure ) I'm the kind of person that Elias described as:

"the poor client trying to work out what the hell you are talking about..."

Finding out the best strategy for optimising the use of the Internet for e-commerce and e-business, is discussed since chapter 1 with different concepts and samples to illustrate them. I agree that within a concept must arise an idea, but if this concept drives to intellectual masturbation, it becomes quickly boring and somewhat unproductive. We have to focus on a strategy that is real and feasible for it to be productive.

Fair comment. It does seem to be a completely unnecessary and complicated way to approach e-commerce solutions because it isn't dealing with any reality. But, the reality of e-business and e-commerce is too complex to deal with. There are just too many variables to be taken into consideration to be able to think clearly. We need a technique, a mental tool, that allows us to see through the confusion created by the detail. The point of the green frog and Hilbert space is to be able to have a way of getting at the essence of problems without getting bogged down or side tracked by any irrelevancies.

To be able to think of starting at a random point like the green frog and moving from that point around in space to find a better position, enables us to cut out all the side issues, prejudices and emotive issues that keep cropping up to cloud real life issues. It also allows general strategic policies to be discussed without having to use specific examples that cover only a limited range of applications.

This is particularly important when it comes to dealing with people because human interaction is so beset with emotive overtones that rational strategic thinking is severely handicapped. This is particularly true of Internet communication because human emotions have evolved to deal with real world communication and they conflict quite strongly with communication tactics and strategies appropriate for the environment of the Internet.

Yvan Caron, the canadian systems analyst, had another kind of problem which he explained in an email to a table discussion in the virtual cafe:

" Maybe it's me, trying to distort this idea, but, I don't quite understand it. The observer knows where is the hole, that I can understand, but, what about an observer in the complex world of the Internet; how does he know where the profitability is? From an economic point of view, how many experiments can we afford to exploit?

How are you going to convince an entrepreneur or a corporate entity to start from the metaphor of a green frog? Is not that the reason why, when someone wants to have a web site, that he gives 'carte blanche' to an expert to design it? This is what we do when we want to invest our money. We go out in search of an expert to manage it.

I don't see how I could afford to change my choice when I started with a so-called expert. There would be penalties to pay if I decide to change to another expert. This economical problem is also present in the WWW. So my question is, how are we going to cope with this if we choose to keep the same designer?

Also, to complicate the problem a bit more, what strategy would we use if the small hole in the middle of the large field started to change its position in the process of trying to locate it. To me, this new dynamic property reflects more the situation that we may find in the WWW. Everything is moving, the only discernible pattern is that there is no permanent pattern. This is where I get lost when I try to understand the science of random fluctuation that we find in complex systems.

Yvan

Yvan sees three problems here. Firstly, the connection between the model and the real world. He is mixing up the goal with the solution. The observer, the person using Hilbert Solution Space to find a solution, has to have in mind what problem is to be solved. It is not necessary to know the final answer in advance to act as an observer to guide this process. The marble throwing example was a simplified view, where the answer was a physical location. For this, the observer would have to know where the hole is. But, most times the answer will not be specific: it will be relative.

Top down, structured thinking involves knowing where you want to go before you start out on a journey. Bottom up thinking is where you have no way of knowing where you are going to end up. This doesn't seem to make sense until you realise that in the world of e-commerce the goal posts are constantly moving. In the case of the marble throwers, it's as if the hole is constantly on the move.

Even worse than the hole constantly being on the move, the hole may suddenly make a jump to a different part of the field. There may be many different holes and you have to decide which is the best hole. Just imagine trying to create a route map (the equivalent of a structured top down plan) to reach a hole in a field when there are lots of holes in the field all jumping around like crazy. This is the kind of problem that has to be solved for e-business and e-commerce solutions and is why it is necessary to use an abstract modelling like Hilbert Solution Space to devise a suitable strategy.

Yvan's second problem is with the association with the experts that are needed to help find a solution to an e-business or e-commerce solution.. He accepts that no one knows everything there is to know about e-commerce solutions, so, there is a need to commit to advisors and experts of various kinds. Having committed to any expert, Yvan sees it as creating a relationship that has some degree of permanence. Yvan is suggesting that this imposes either a moral obligation or a financial penalty if that commitment is cancelled. To answer this second problem, we need to solve Yvan's third problem.

Yvan's third problem is, as explained above, he recognises that the best solution for any particular e-business or e-commerce venture is continuously changing with the fluctuations of technology and competitive strategies. A solution in one point in time may be totally inappropriate at a later date. In other words the place (or hole) in Hilbert space that describes the best solution is constantly on the move. The goal posts are moving.

This can be a serious problem if your strategy is to stick with the same group of expert advisors and specialists, because the range of knowledge needed at any new solution hole may not be that of the experts employed in the outdated solution. The solution moves, but, the knowledge of the experts may not necessarily move with it?

In all specialist areas of e-commerce there is too much to know and too much technological advances to continually keep up with. Any particular solution requires a fair amount of expertise, so, to be really an expert, the expert is forced into being confined to a restricted range of technology. This makes them vulnerable to becoming experts in areas of speciality that becomes outdated or superseded by a new technology. In the new hole in Hilbert Solution Space, where the preferable solution is to be found, it may need some brand new thinking that is outside of the original designer's range of expertise or knowledge. This may make it impossible for the original expert to be a dimension of any new solution.

What do you do when your main competitor uses an expert who is a specialist in a newer and superior technological development. Do you pay your current expert for 'on the job' training? Is your project going to be the experimental model for your expert to learn how to change to another area of expertise? Do you wait for your expert to become expert in new areas of technology before you make a competitive move? Maybe the consultant or expert you use will have many other clients, clients who aren't pressured enough to go for a newer technology. Your expert may then not see any great benefit in changing their knowledge base and expertise to suit you?

All these factors make it totally inefficient to carry a group of experts around in Hilbert Solution Space with you. It is far easier to go to a place where the experts you need are already in place. If an e-business or e-commerce solution is visualised as being at a point in a Hilbert Solution Space where the dimensions are people, it is easy to see how moving around in that space is about changing the people involved in the solution. In the bricks and mortar world of the Industrial age this isn't a practical way to think, but, in the Internet environment of the Information Age it is the only way to think.

Yvan mentions the problem of there being too many possible strategies to have to investigate before arriving at the best solution. There isn't. What he is meaning is that there are too many possible solutions. Each hole in Hilbert space is a different solution. The trick is to use a single strategy to find the best hole or point in the Hilbert Solution Space. This will be the place where just the right combination of contacts are to be found: the dimensions of this place being specified by a list of names.

This then, answers Yvan's second problem: a strategy for an efficient e-commerce solution shouldn't involve establishing permanent fixed relationships with experts or advisors. This is one of the big paradigm shifts that e-business and e-commerce will require. This is at odds with all conventional thinking in the Industrial age and, emotionally, does not seem right. But, as we shall see, the logical and emotional objections disappear when we start to apply game theory to Internet communication strategies.

A few days after Yvan sent that post to the table in the virtual cafe, he sent another ­ after having given some further thought to the implications of using the Hilbert Solution Space model:

While reading a magazine lately I was struck by the title of an article: 'Internet access WANTS to be free'. It struck me because it precisely describes what Peter says when he writes 'with a bottom-up approach a system designs itself'. Also, I could connect it to what I have read this summer.

My reading was about the conception of goals or goal-driven systems. 'Goal' has many complicated meanings. A goal-driven system does not seem to react directly to the stimuli or situations it encounters. Instead, it treats the things it finds as objects to exploit, avoid, or ignore, as though it were concerned with something else that doesn't yet exist.

When any disturbance or obstacle diverts a goal-directed system from its course, that system seems to try to remove the interference, go around it, or turn it to some advantage. For example, if you climb a hill and throw a ball down you may have the impression that the ball 'WANTS' to go down trying to go around, exploit or avoid any obstacles on its path.

Looking at the ball we could say that it appears to have some representation of some outcome (a goal: to reach the bottom of the hill) and a mechanism (a WILL) to make it persist until that outcome is achieved. Of course, in this simple case the impression of intention is only in the watcher's mind. A complex system may produce a random fluctuation that may persist long enough and drive with it a representation of some outcome. And, it is where we must be alert because this may drive the next big trend.

Apparently the next big thing will be a tidal wave of free internet access. It is said that the big portals are failing. Originally, there were just a few portals like Yahoo, Netscape and MSN, and these guys are still among the biggest. But now everybody's getting into the act. There are literally thousands of portals out there, with new ones emerging every day. Internet content providers have always dreamed of creating a TV-like captive audience that can be force-fed advertising.

That was the goal of push publishers, and that's the goal of portals, But offering free content and e-mail doesn't attract enough eyeballs anymore because now everybody's doing it. With more Web options to choose from, the big portal companies are losing customers. That's why the rich portal vendors will find a way to blow away the poor ones by partnering with ISPs to offer free or nearly free Internet access.

To me, this is an example of a system that designs itself because this new wave has emerged from the intense competition among the players and we could not have predicted it.

Yvan

Yvan, as a senior system's analyst, is realising that there isn't much you can do to design your way around an environment as complex as the Internet. There is too much emergence of unpredictable events. He has come to the conclusion that the system is being goal driven. Once given a goal, the system will automatically find its way over the hills and valleys to find its own solution: its own best hole.

This gives us another paradox. You control a system by not controlling it. You just set a goal and let events take their course. This is like another of those infuriating, enigmatic conundrums you find in Zen philosophy. It just seems too ridiculous to have any meaning.

Let's invent another little analogy. Imagine you are visiting a strange and exotic country that no one had ever visited before. Your goal is to find and photograph the most beautiful building in the country. It is a very hilly country

You can imagine yourself lost in a hilly landscape and you wander around asking people where you can find the most beautiful building. You don't know what is over the hills but someone you meet might know. As with asking your way around a geographic landscape, you can't expect every stranger you ask to know the exact location you want to find. They'd probably point you in a general direction and tell you to ask someone else when you get there.

Most likely, you'd be sent to a busy village where there would be many people who are familiar with the locality. You'd ask a few people there. The Internet equivalent would be a special Interest group, probably one that was set up to discuss e-commerce. By subscribing to this you can listen in to the discussions and when you started to get a few ideas you'd begin asking a few questions.

In terms of Hilbert space, you'd be at a spot where there are hundreds of dimensions, people dimensions. A very useful hole. Soon you'd get to know people, you'd make friends. You'd find out who you can rely on who you can trust for accurate information. Perhaps someone will act as a guide and introduce you to someone else who has a map and can show you the way.

This is the strategy for exploring Hilbert Solution Space to find the best solution. It's about wandering about in an environment and talking to people who will lead you to others, leading you nearer and nearer to your solution: and the solution is not any specific design it is the people who have the right knowledge and experience to help you go where you want to go. The trick is: how to do that efficiently in the complex environment of the Internet.