20 Rules for a wise company in the information age

1. Save all data possible. Without data: no information, no knowledge and no wisdom resulting in no value. Remember that our processing capability increases exponentially with time and in the near future great value may be derived from seemingly worthless data.
2. When saving data provide data with as much context as possible, what is it? How? and when was it captured ?what is its relationship to other data? and to other phenomena?. That way data is transformed into information. Proper tagging of data is also vital for communication of information and knowledge.
3. Abide to standards which increases the probability that information will remain information over time also to other entities in contrast to just being degraded to data.
4. Make data and information freely available within the company to an as large extent as possible. Remember that there is usually much less motivation to protect data and information compared to knowledge and wisdom. Remove all policy barriers as well as technical barriers. Make all data searchable and connect as many information-carrying systems as possible. Free flowing information represents an enormous value and usually leads to spontaneous creation of knowledge and wisdom. “Destroying information is a sin” and making information inaccessible equals destroying information.
5. Create an infrastructure for knowledge creation and sharing allowing people to exchange their hypothesises of compressed understood information. Hypothesises are embryos of knowledge, and potentialyl of wisdom and value and must be encouraged. We must set up a legal framework to allow for an unlimited discussion of hypothesises at the same time securing against any risk. Also make sure that knowledge from outside sources is readily available as information not just data e.g. textmining.
6. Acknowledge that a computer system is sometimes more creative than a human. A knowledge-extracting algorithm may try one billion ideas per hour, resulting in a creativity far exceeding a human analyst. We must constantly use and connect to data mining and to text mining systems for automatic modelling.
7. Allow for the creation knowledge and hypothesis in the most expressive forms possible. For instance as algorithmic models. Better expressiveness gives better predictive power and results in more wisdom and value. We should allow for multivariate analysis, algorithmic analysis, Bayesian representations etc on as many levels as possible. The most expressive representation is the Turing complete computer program. Use automatic modelling techniques that can model in these highly expressive languages.
8. The highest form of knowledge refinement is reasoning we must support formal and automatic reasoning. Reasoning could be what separates man from animals. We can reason about the future based on symbolic information. This allows for a much fast acquisition of knowledge and production of plans. A wise company has an infrastructure for reasoning, also in an automatic form and allows for its knowledge to be stored in a way suitable for automatic reasoning..
9. Expand the use of simulation. The low hanging fruit of science which lends it self to analytic analysis may be disappearing. Instead, we need to simulate everything to evaluate future scenarios and communicate relationships. Simulators are becoming the base for most of science and we need to lead this trend on all of our areas.
10. Use automatic planning. Planning is too complex to be carried out manually and need a large component of trial and error. Optimal planning requires constant replanning (agility) and the use of adaptive representation (e.g. adaptive study plans) and decision points (algorithmic plans).
11. Use models and representations which allows for representation of uncertainty. Science is about prediction and prediction is about uncertainty and many future scenarios. Few models are of value without some representation of uncertainty.
12. Support different persons’ needs for knowledge creation. Some prefer visual tools while other prefer symbolic tools. Sometimes “Push” is needed not “Pull” since people don’t always know what information they need.
13. Acknowledge that people come in different flavours. Some are more prone to generating ideas, hypothesises and new knowledge while others are more prone to evaluating and criticising them. Both are needed in the right balance and sometimes bridge-competences are needed for communication between the two.
14. Define value of all knowledge. That is a prequisite for wisdom
15. Acknowledge the value of both deep and shallow models. Much success is achieved by doing what works without explicitly knowing why it works, but sometimes a deep model gives superior performance.
16. Make sure that ALL plans, algorithms, actions and events are tied to a measurable quantifiable value, without tying knowledge and information to value there is no wisdom and no learning organisation.
17. Make sure that history of everything that happens, large and small is recorded. This is the basis for acquiring long-term wisdom in the organisation. All decisions and plans must be documented.
18. Monitor the compliance to prescribed procedures. Clever methods not being used are worthless.
19. Make everyone aware of the tools and methods available.
20. Measure value in a single currency. Different plans must be able to be evaluated on the same terms regardless of completely different approaches and ideas stemming from different disciplines.