Dear Ken and list, [warning, long post] Thanks Ken for an excellent post. And for the close reading of Frayling as an entry into the complexities of the issue in question. I fully agree with your analysis and believe this is the kind of reasoning that can and will develop the notions around design and research. Inspired by your post, I do want to make a couple of comments. As I read your post, the overall message is what you summarize at the end as "I merely point to the fact that explicit and articulate statements are the basis of all theoretical activities, all theorizing, and all theory construction." [from Ken's post] and I basically agree with this. It is however possible to develop the argument a bit (which you also do in your post). The basic argument for the "explicit and articulate" is that theoretical knowledge is all about communication, so, knowledge has to be explicit and articulated to be possible to be communicated, analyzed, tested, criticized, etc., and of course, to be something "objective" that others can build upon. And again, the reason for this is the idea that nothing is knowledge until we (or as many as possible) find the knowledge proposal to be trustworthy and believable. This is where the highly refined methods of scientific research (in all areas) fill their purpose and have been extremely successful. If we focus on this overall quality of trustworthiness, it means that it is possible to imagine new and other forms of knowledge (that don't fall into the category of being 'explicit and articulated' in the traditional fashion) that could have the same qualities (that is, possible to communicate, analyze, critique, test, etc). BUT, these new forms of knowledge must have to pass the most crucial quality test: they must be able to evoke trustworthiness, that is, people will actually have to believe in the results. And this is the hardest test of them all. People or researchers only believe in others research if it has certain qualities, and we all know how hard we are as judges of others research :-) So, even though it is possible to argue for and develop a rationale for non-explicit and non-articulated knowledge as a 'true' form of knowledge, the test is if research results in that form will convince people about its truthfulness. The power of traditional scientific research is a result of centuries of developments and refinements of the process, the methodology, and of a philosophical foundation around what makes scientific results believable. When it comes to design in research, we have nothing of that in place yet. To summarize this point: the idea that there might exist other forms of knowledge, that don't comply with the rule of being "explicit and articulated", is not necessarily a dead idea, it might be a possibility, but it is extremely difficult and can not be solved by argumentation only. It is only when people actually do believe in results presented in these new forms that it can be said to be valid form of knowledge. So, the resistance to new forms of knowledge production is not (only) a consequence of not good enough arguments, it is in most cases a consequence of the fact that the results in themselves are not convincing, trustworthy and believable. My other comment is about the notion of "research by design". I agree with Ken's statement: "While the phrase research by design has been widely used by many people, it has not been defined. I suspect, in fact, that those who use the phrase have not bothered to read either Fraylings (1993) paper or Reads (1944, 1974) book. Instead, they adopt a misunderstood term for its sound bite quality, linking it to an ill-defined series of notions that equate tacit knowledge with design knowledge, proposing tacit knowledge and design practice as a new form of theorizing. " I do believe that "research by design" is a possibility. And I relate this to my first comment above. Any kind of research can and will be accepted and valued if the results are trustworthy. But, it is extremely important to carefully respect the intention of the activity in question. The purpose of design is to create new designs. The purpose of research is to produce new knowledge. This means that research can be done "by design" (or "by" any other process for that matter), but, it will always be judged and evaluated as research, not as design. That is, the final evaluation is if the process has produced new and agreed upon knowledge, and as such is has to live up to the expectations that I discussed earlier in this post. This means that research can "use" design within or under the umbrella of research. But it is not possible to substitute research with design, since it also means that the overall 'measure of success' is substituted! In my own field, Human Computer Interaction or Interaction design, there is at the moment several attempts where researchers (usually with a design background) are trying to carefully develop frameworks, methods, and principles for how research in the field can "use" and "exploit" the strengths of design (not making research into design). This means that these strengths have to be incorporated into a larger framework of knowledge production, with all its requirements and limitations, while carefully keeping the specific values of designerly inquiry and action. One interesting reflection from these attempts is that this is a delicate task, it takes a lot of "homework", a serious understanding of the scientific tradition AND of what the core qualities of a designerly approach are, combined with real concrete attempts where researchers actually show how this can be done. If done in a careful and reflective way, there might be exciting opportunities with this approach. Ok, that is it for now. Erik Erik Stolterman Professor of Informatics Director of HCI/design School of Informatics Indiana University web: http://hcid.informatics.indiana.edu/eriksite/ blog: http://transground.blogspot.com/