|
p.8 Knowledge is a protean [JLJ - capable of assuming many forms] concept. Although there have been many attempts to define and classify it, whenever a specific case of any complexity is examined the definitions and classifications do not quite apply. This is a problem that plagues any discussion of the general concept of knowledge. [JLJ - Practically, knowledge is just a revision of earlier knowledge. Or, knowledge is a required intermediary in the pursuit of a scheme. Knowledge is a packaged cleverism, that can be placed on a shelf and stored, waiting for a strategic or appropriate time for use. When one uses knowledge appropriately and in a timely manner, one appears to be clever and one often succeeds at what one is doing.]
p.9 In many cases technique can be derived from the general body of understanding... A technique that can be derived in fine detail from more basic understanding can be called perfectly understood. Few techniques are perfectly understood in this sense... The fact that technique can often be largely or partially derived from the more general body of human knowledge crucially affects the way many new techniques get discovered or invented... In theory any technique can be described by and communicated as a set of instructions - as a cookbook recipe or a computer program - whether or not the technology is understood in terms of more fundamental knowledge.
p.9 [footnote] While good strategies often can be derived from logical manipulation of the general problem formulation... good strategy may sometimes be discovered by trial and error.
p.11 The techniques relevant to most jobs have a large common set of elemental building blocks. It generally simplifies the teaching and learning of the techniques if these more primitive elements are already learned... General education imparts these general purpose skills, relations, data, and language.
p.12 Skillful decomposition of activities, and an education and occupational training system well tuned to the pattern of decomposition, permit a great reduction in the amount of knowledge that must be specific to any particular activity.
p.15 The principal source of growth is investment
p.15 [footnote] Potential output has been defined as the possibilities open over a period of time, not at a specific moment... Another way of defining the potential output frontier is in terms of the alternative output combinations the economy can be producing at some specific time in the future. In either case, the production possibility concept inherently involves future adjustments and reallocations over a period of time
p.20 Final products are, after all, merely ways of satisfying consumer wants. New final goods are often simply more efficient processes - less costly ways of meeting needs that were met before.
p.23 In addition to the stock of presently operational techniques, at any time there is a considerable store of ideas reasonably well worked out, but still short of operational, and an almost infinite stock of partial or embryonic ideas which are not even close to operational. Technology advances as these ideas are developed into operational form. The conscious application of resources to this objective may be defined as applied research and development.
p.26-27 Most R&D projects aim for rather marginal improvements in a product or item of equipment, and usually the effort is directed toward improving a few dimensions of performance. As a result, the effort generally can focus on several components and take the rest as given... the number of diverse elements is significant mainly through their interdependency... The problem with trying to achieve major advances in large and complex systems - products with a large number of tightly interdependent components - is that to change any one item causes reverberations throughout the system. Even small changes in performance in just one or two dimensions may require a vast amount of redesign. For smaller or less complex systems involving fewer or less interdependent components, one aspect of design can be changed with relatively little overall redesign.
p.34 Whether or not people try to solve a problem depends on whether they think it can be solved, and at what cost, as well as on the gross returns from solving it. [JLJ - Not the view of a Synthesist thinker, such as I am. A Synthesist thinker would play with the possibilities just because he enjoys doing so. If anything useful emerged, it would be pure luck.]
p.42 the distinguishing aspect of basic research is emphasis on certain key areas where improved understanding is judged particularly likely to yield solutions to practical problems and open up promising areas for development.
p.43 while he recognized that major advances come more easily in the science-based industries, Schumpeter failed to see that the reason is... the large and continuing effort toward basic and exploratory research.
p.56-57 There are, of course, independent inventors outside the formal R&D organizations. A diverse lot, their ranks include full-time free lancers... trying to invent in off-hours... The relative and perhaps absolute importance of this group has been declining for years... Nonetheless, the independent inventor is far from extinct... Perhaps the largest part of the work of the independents appears to be the "follow on inventions" ...those design improvements suggested by practical experience.
p.86 As pointed out in Chapter 2 [JLJ - p.22-43], the total costs and risks of development efforts aimed directly at creating new products can be considerably reduced by undertaking basic research and experimental development to probe and open up interesting possibilities before commitment to full-scale design and development.
p.90 The stock of embryonic ideas for new or improved products and processes is very large. At any time many ideas are being worked on in a limited way with the objective of making their concepts more explicit, testing their technical feasibility, and assessing their value and cost. This kind of exploratory work often can proceed with a moderate investment.
p.92 Rising costs of continuing an R&D effort require a more hard-headed appraisal of costs and benefits. While the early work often can be carried on by an enthusiast despite the scepticism of the majority of experts... the middle and later stages require someone or some organization to commit large quantities of resources... As rising costs force a more calculating appraisal, this is often made possible by the knowledge created by earlier work... In most far-reaching efforts the initial uncertainties may be extremely great, and the probing for a workable design concept may be the only practicable key to their resolution.
p.93 Marshalling major resources to a project usually awaits the articulation of a design concept which appears technically workable and its careful economic evaluation.
p.94 As development proceeds, assessments become sharper. Sometimes the early uncertainties are resolved easily and development proceeds on the "best bet" course. In other cases, problems prove much harder to solve than had been hoped and expected... Thus, as work on a project proceeds the cost of continuing increases, perception of cost and value becomes sharper, and projects are screened out.
p.102 Some innovations, representing only marginal modifications from existing equipment, may be relatively easy to demonstrate and evaluate. For major innovations it may be considerably more difficult for sellers to demonstrate that a potential user could effectively and beneficially adopt the new technique or product, or for potential users confidently to make their own evaluation... Various studies show that, where possible, early users of a new product proceeded experimentally.
p.108-109 Many writers have commented on the apparent waste in the... change process: a large number of projects never result in anything useful... Yet sequential decision making in situations of great uncertainty which can be reduced, either by small-scale efforts or by waiting, is rational both from the point of view of individuals and the economy as a whole. If waste is involved in duplicate efforts, it is also involved in commitment too early to an inferior approach. If there are economic costs of delay in accepting superior technology, there are also costs in accepting new technology that turns out to be uneconomic.
p.144 As knowledge progresses, it results in routinized and mechanized processes capable of being easily operated.
p.173 While identification of current market failure problems can provide guidance for direction, it cannot provide guidance as to the size or kind of program which would be most appropriate. The analysis of market failure together with some imaginative thinking regarding the broad nature of a program that might deal with the problem can provide a policymaker with an idea of how to start, but somehow he must acquire additional information to determine exactly what direction and distance is most desirable.
p.173 Under such circumstances the most fruitful way to proceed is sequentially and experimentally; neither doing nothing because knowledge is less than perfect nor leaping farther than necessary in a prejudged direction.
p.174 Conditions of great uncertainty call for imaginative and flexible probings, not vacillation between inaction and commitment. Indeed, the way uncertainties are dealt with in the R&D [JLJ - research and development] process would appear an applicable model. As described in Chapter 5 [JLJ - p.89-109] the typical R&D strategy of the business firm is to avoid major financial commitments to untried ideas; rather, it seeks to obtain knowledge and thus to reduce the uncertainty surrounding the idea by investing relatively small sums in additional research. At each stage in the process, the company spends money to generate the knowledge necessary for deciding whether to proceed or retrench. As the idea proceeds from design concept to laboratory experimentation to prototype construction to production of limited batches, the investment becomes larger, and is undertaken only if the evidence increasingly points to the probability of profitable production.
The policy proposals discussed here are modeled on this strategy.
p.175 judgment can be greatly facilitated by careful prior consideration of [JLJ - numbering added for readability]
- the major uncertainties surrounding the program,
- the most fruitful ways to acquire knowledge to reduce or resolve these uncertainties,
- the nature of the decisions that should be made at some later point on the basis of the knowledge acquired in the course of the program, and
- the decision making machinery suited to make these decisions effectively and without bias.
p.175 Policies and programs differ intrinsically in the extent to which they are amenable to a sequential decision approach. Specific research and development projects... are well suited to this strategy. The early decision can be in terms of a small scale commitment to undertake research on the key uncertainties, and to undertake experimental development of critical but uncertain components. Decisions regarding larger scale financial commitments and the details of the design, if development is to be pressed, can await better estimates of costs and product attributes.
p.176 [footnote] More to the point, it is agreed that limitations on the information available to decision makers generally means that the preferred policy strategy is an incremental one. Exploratory, serial, and reconstructive policy action is the heart of disjointed incrementalism and also of sequential experimentalism.
p.177 For each of the areas it appears possible to design an experimental program which, without massive commitment of resources, can probe the problem and feed back information that will help improve judgment regarding whether any kind of program at all is likely to have net payoff, and if so the kind and scope of the program. While formal evaluation procedures cannot substitute for judgment, careful planning of data collection and analysis can help to make judgment somewhat less capricious.
p.177 there is little to be lost and much to be gained from an experimental probe of the possibilities.
p.209 Without an early orientation to experimentation, data collection, and analysis... the proposed evaluation... may not be able to provide a firm recommendation supported by empirical quantitative evidence. Indeed, it might be able to do no more than recommend further experimentation.
Predictability of the Costs, Time, and Success of Development, A. W. Marshall, W. H. Meckling, in: The Rate and Direction of Inventive Activity, 1959, 1962
p.461-462 In this paper we present the results of some recent research into the extent and nature of the uncertainty in new developments. Our general conclusions are:
- "Early" estimates of important parameters are usually quite inaccurate...
- The accuracy of estimates is a function of the stage of development, i.e., estimates improve as development of the item progresses.
p.463 Ultimately, the measure of success of any development project is the difference between costs and the value of output.
|
