What is the important literature regarding knowledge creation via formalization/structuring?

August 18, 2004 2:30 PM

EpistemologyFilter What is the important literature regarding knowledge creation via formalization/structuring? [mi]
posted by badstone to (8 comments total)

We have this design system which basically consists of some templates and metadata that users fill out. Using the system, they take a rough internal sketch of an idea and translate it into our formal structure. (The design system is for the creation of educational assessments, but I don't think that is necessarily relevant to my question.) I need to make an argument to non-techies that expressing their ideas within our ystem is not just a matter of jumping through a lot of arbitrary hoops to translate their idea to a technological format, but is in fact an act of knowledge creation that not only communicates their idea, but actually enriches it.

I've poked around the educational design literature on this point and am unsatisfied. I need to branch out to philosophy and design science in general but am having trouble articulating my search as this is not really my field. I have an intuitive grasp of the importance of this idea, but I need to back it with serious references. Where should I start? What are the seminal papers?
posted by badstone at 2:30 PM on August 18, 2004

personally I would start with gang of four, but quite possibly that is one of the things you were unsatisfied with.

so, I dnno, umberto eco is always good; and so is jaako hintikka.
posted by dorian at 5:13 PM on August 18, 2004

Classical epistemology concerns itself with stuff like the justification of belief, and is well and thoroughly out of favor. (Cf. Quine 'Epistemology Naturalized, or just talk to anyone who has been near any kind of graduate seminar.) Also, the term 'epistemology' is freely used by literary critics and cultural studies people to denote a wide variety of theoretical fauna. So, it's hard to know what you mean by 'epistemology', especially in the context of a literature search.

That said, I think the academic fields you should be looking at are cognitive science, cognitive psychology, and AI, where modelling knowledge systems has been an active area of research in recent years.
posted by crunchburger at 5:49 PM on August 18, 2004

Actually what might be interesting on the subject of knowledge and frameworks is Gombrich's Art and Illusion. There's also a really interesting (though it overgeneralizes) essage by Galison and ... Daston? on the way encyclopedias presented and justified their claims to knowledge and authority. I'd give an actual reference but it's in a box somewhere.

Neither of them is likely to be of much direct use to you but they're interesting.
posted by kenko at 6:13 PM on August 18, 2004

I was lucky enough to have lunch with quine once, at a ramen place in boston. brilliant man, but I'm still not quite suire why he put salt and pepper into his tea.

very sad at his passing -- when I was growing up, so many of the axioms in my textbooks were '*proof by quine'.
posted by dorian at 6:14 PM on August 18, 2004

epistemology out of favor? Crap. So much for my minor.
posted by Grod at 12:37 AM on August 19, 2004

My take: what does a formal structure system really get you? There are, in fact, a lot of useless formalization processes I've gone through (most of which, truth to be told, were probably designed by me), so it's not automatically true that it buys you anything. But... if you're lucky, you can move your ideas into the context of a formalism that people have already dedicated a lot of time to understanding the pattern and structure of, and it gives you a set of possibilities for either new, derived insights, or for making something new -- and this stuff comes for free, because of all the time people have put into understanding general things about the formalism.

One example: raster image formats and image manipulation. An image drawn on paper is something you can do some limited image manipulation on. Scan it into a computer, though, and put it in a format Photoshop can read, and suddenly, you have the opportunity to analyze or transform the image in thousands of ways because of decades of research that've been done. This turns out to be so useful that it's very popular. And of course it's not just images, it's audio, video, text documents -- it's almost self-evident that moving things into digital domain gives us considerably greater power over them. And what is the digital domain except a set of elaborate formalisms?

Not that digitizing things is the best example available -- but I think it's one of the most encountered in society, and people might relate to it somewhat.
posted by weston at 2:39 AM on August 19, 2004

I need to make an argument to non-techies that expressing their ideas within our ystem is not just a matter of jumping through a lot of arbitrary hoops to translate their idea to a technological format, but is in fact an act of knowledge creation that not only communicates their idea, but actually enriches it.

one of the big deals in 20th cent. philosophy may be of great use to you in fashioning a justification for your purpose: the linguistic turn. after not much happening in the world of logic for hundreds of years, there was a series of breakthroughs by frege, russell, wittgenstein & others around the turn of the century. this, and the threat of psychology which was encroaching on philosophy's turf, led to a concerted effort to analyze the relationship between thought and language. it began as a quest to map thoughts as they are fluidly expressed in human languages into rigid logical systems which mimic mathematics, and in the process improve how we communicate. Although it broke down as some examples of strange language began to be perceived as intractable or at least beyond agreement on a solution, there was much excitement as several interesting concepts emerged from this process.

the basic concept should be obviously valuable to you: russell discovered that certain types of confusing language were much easier to understand by translating them directly into linguistic symbols, and seeing whether they are self-contradictory, or tautological, or incomplete, etc. his was a sort of obvious restating of a concept that everyone knew, but nobody had really paid attention to, that characterizes the best spirit of philosophy... in a nice way it was an elaboration of the socratic method of simply examining an idea [eg "what is justice?"] but the point is, can we do what socrates did but in a more systematic way? in a way which, as in science, produces the same results for whoever replicates our thought experiment?

moreover, the very fact that there are so many problems which resist linguistic mapping onto a strictly logical plane suggests for many philosophers and scientists [of cognition and AI] that either the brain is fundamentally different [in some way we don't yet understand] from a logical processor, or that at a certain level logical processors have emergent complexity with surprising non-logical characteristics.
posted by mitchel at 1:26 PM on August 19, 2004