Artful Questions

Artful Questions

Pynthantics

The term “pynthantics” is taken from the Greek pynthan (πυνθαν), meaning “to pose a question, and perhaps get an answer”. It is the science and art of posing questions. More generally, it the art and science of extracting information from a system in a usable form at an affordable cost. As a formal science, it is the science of scientific method. An important tool for use with it is questimpet, a method of a group asking as many questions as they can, then selecting the best questions for their purposes.

Operations

There are several key elements: (1) The extractional operation. Applying it, the result is an observation. The simplest extraction distinguishes an object, the phenomenon, from its background. (2) The comparative operation. If two different comparative operations on the same phenomenon produce the same result under all circumstances, we can assume the two phenomena are equivalent. (3) The model operation. This operation presents a model that may explain, predict, or control a phenomenon. (4) The fitness operation. This extracts information that a model fits the observations, and how well, to distinguish that some models are better fits than others. (5) The utility operation. To find how well it enables explanation, prediction, or control of a phenomenon. (6) The cost operation. This yields the cost of use of a model.

Questimpet (Questorming)

Questorming is a variant of brainstorming, the technique developed at MIT in the 1950s for getting a group of participants to come up with more creative solutions to problems. In brainstorming, a moderator first presents a problem to a group, who then propose as many different kinds of solutions as they can, but without evaluating any of them until the moderator halts further proposals and begins the evaluation phase. During the proposal phase they are encouraged to be as imaginative as they can, and not to restrict themselves to what they or others might consider “good” solutions, but to propose anything that might be remotely relevant. During the evaluation phase, the proposals are consolidated, perhaps reformulated, evaluated, and by process of elimination, reduced finally to one by votes of the group. The technique emphasizes the importance of suspending criticism, both of one’s own ideas and of the ideas of the others, to foster creativity and original solutions. It recognizes that “bad” ideas may often be more productive of “good” ideas than “good” ideas are. The method also involves the application of various standard constraints on the solution adopted, such as that it can be carried out by the members of the group without requiring any resources not available to them.

Questorming takes a somewhat different approach. Its aim is not so much to get a group to come up with “solutions” to a “problem” as to come up with well-stated and well-selected questions or problem formulations. In one sense it addresses the process leading up to what is done in more conventional brainstorming: formulating the problem to be solved by the group. In another sense it is brainstorming in which the problem for the group is to find the answer to the metaquestion, “What are the best questions we need to ask right now?”. Questorming is based on the recognition that if people can ask the right questions, the answers are often easy. It also does not allow the moderator to control the outcome by the way he or she initially formulates the problem for the group.

This does not mean that the moderator does not need to impose some constraints on the subject matter of the discussion. This may be done in various ways. One way is to briefly describe the situation as he or she sees it, including the members of the group and their concerns, and invite them to then ask and try to answer the metaquestion with reference to that situation. The answers they then propose may actually involve a rejection of the situation as described by the moderator, but it serves as a point of departure to keep the discussion focused.

As with brainstorming, criticism of proposed questions is suspended for a period of time until a sufficient number and variety of questions has been achieved, after which the evaluation phase is begun. The objective in questorming is not necessarily to come up with one best question, but a list of questions ordered from best to worst, or perhaps a tree structure in which related questions and subquestions are organized and ordered by quality.

The technique also involves guiding the discussion with a standard list of generic questions which are considered givens that do not need to be proposed themselves. One example of such a standard list is provided at the end of this document. The moderator should direct variants of such standard questions to the group or its members to keep the discussion moving and focused.

The technique does not forbid all discussion of answers or solutions to the questions proposed. Mention may be made of the kinds of answers or solutions a question might have as a way to provoke ideas for more questions or to indicate ways to evaluate the questions, but the moderator is supposed to discourage the group from getting sidetracked by a discussion of answers or solutions that are not themselves questions or problem formulations. This is most easily done if it is agreed that such answers will be addressed in a later session devoted to more conventional brainstorming.

Part of what is sought in questorming is a determination of whether the concerns and perceptions of the members of the group can support concerted action toward a common goal which meets the perhaps divergent goals of each of the members of the group. This involves identifying a point of agreement among the members and building upon it until a basis for group action is established. If no such consensus can be achieved, then the membership of the group may need to be altered or the group dissolved.

Although some groups, because of their backgrounds, may be able to assume certain common concerns and perceptions, it is often useful to go back to basics and re-establish the principles which unite them. This effort can often uncover and resolve divergencies that will later interfere with concerted action by the group. On the other hand, it may also sometimes be best not to examine first principles too closely, lest disclosure of divergencies result in dissolution of the group or impair its ability to function. That is a judgment call by the moderator.

One of the jobs of the moderator is to make sure all proposed questions are kept before the members of the group, and direct their attention to them as required. This may involve posting them on a board, or providing each member with a scrolling display or with a display large enough to allow all questions to be kept in view at once. He must also keep the questions down to a manageable number. If the number becomes too large, it may be necessary to break down the task into smaller tasks, each with its own set of questions, to be dealt with during separate sessions.

The moderator may also need to introduce some levity in case one or more of the members begins to become too emotional about the subject matter or the process. Members should be instructed that even though they may have strong feelings, there is likely to be a better outcome of the session if everyone treats it as an intellectual game.

For more information on this technique, contact Jon Roland, jdr@pynthan.com .

Standard Questions

  • What are the best questions we need to ask right now?

  • Is there a better way to state that question?

  • What are the standards by which we decide that one question is better than another?

  • What are the best evaluation methods by which we decide that one solution is better than another, and how can they be ordered?

  • How can we best organize the questions in descending order of quality?

  • Who or what should ask the question?

  • To whom or what should the question be directed?

  • How should the question be asked?

  • When or under what conditions should the question be asked?

  • Where should the question be asked?

  • Why should the question be asked?

  • Should the question be asked?

  • What are likely to be the consequences of asking the question?

  • What purposes might the question serve other than getting an answer?

  • What might be the purposes of someone who asked that question in a given situation?

  • How could the question be misunderstood?

  • Is the question meaningful?

  • Is the question answerable in principle?

  • What is the operation by which answers might be produced and selected?

  • Is the question-answering operation reproducible by others?

  • How can we tell whether the question has been answered?

  • By when are we going to need to have an answer?

  • By when are we going to need to have the question?

  • What are the concerns and goals of the members of the group?

  • What are the perceptions and abilities of the members of the group?

  • What are the kinds of resources associated with this question and the group? Space? Time? Energy? Materials? Information? Agents? Skills? Initial state? Operating conditions?

  • What are the tools and resources available to the group? How can they be obtained?

  • What kinds of costs are associated with asking this question and answering it?

  • What are likely to be the costs of asking the question?

  • What are likely to be the costs of trying to answer the question?

  • Is the question answerable by the group given its resources? If not, what do we do?

  • Who or what can answer the question in a timely manner and at a reasonable cost?

  • Should two or more of the questions be combined into one? If so, how?

  • Should the question be split into two or more questions? If so, how?

  • Have we come up with more questions than we can handle during this session?

  • Which questions may have to be answered before others can be answered or even asked?

  • What are the questions or kinds of questions we may need to ask later?

  • How well is this questorming session working? Are we criticizing too soon? Too late? Are we getting sidetracked into a discussion of answers?

  • Do we need to change the composition of the group? Add members? Expel someone?

  • Are we lacking information critical to proceeding further? If so, how can we get it?

  • Have we covered all the major possibilities? Are we overlooking anything?


Fetura

Another term refers to a tool for this process, fetura, or breeding, meaning a process of alternately applying a random selection or reconfiguration operation, and testing the fitness of it, saving the cases that best fit. It is the algorithm of evolution. It is applied to the above modeling process, saving the best fits for further use.