When two things are like each other

In the October 22nd issue of The Economist there’s an article about urban pulses (“Listen to the music of the traffic in the city,” p. 70). It reports on research (Miranda et al. 2016) that measures activities as diverse as Flickr posts and traffic volume, which together generate an impression about ebbs and flows of activity in a place over time, as well as identifying other similarities. The hook for the article is the notion that Alcatraz and Rockefeller Center turn out to have the same pulse.

It’s just one more example of the kind of situation that I wish we in knowledge organization (KO) were more concerned with. This is the notion that when two things are like each other it might be meaningful, whether the relationship between them is semantic or not. I think in KO we are too much oriented to semantic similarity systems to the exclusion of almost everything else. A good place to start might be to look for more research like this and subject it to meta-analytical analysis from the KO domain-analytical point of view. What sort of domain is urban pulse, or social-pulse taking (which apparently is a broader term, see the end of the article)? I don’t mean, who are its authors and what are its keywords, although that would be interesting too; I mean, what are the heuristics that lead to classes and how are the classes ordered?

I have been very interested in this approach to KO for a long time. It is one of the reasons I am so enthusiastic about the CIDOC-Conceptual Reference Model (CRM), a meta-level ontology for cultural heritage information sharing (http://new.cidoc-crm.org/). Apart from all of the other virtues of the CRM, it is obvious to me that metadata conformed to it can have a footprint made up of the particular combinations of entities, properties, and relationships expressed in the ontology. This was the subject of research undertaken in my last years at LIU (“Mining Maps of Information Objects” and “Classifying Information Objects” 2008). It also is the theoretical basis for my work on classification interaction (Smiraglia 2013; 2014a; 2014b), of my work with knowledge maps (Scharnhorst et al. forthcoming) and my work with Korean open government data (Park and Smiraglia 2014; Smiraglia and Park 2016).

The point is to use empirical research to discover instances when things that don’t seem to be the same actually are like each other, to generate classifications from those observations, and then to create pathways for navigating similarity discovery.

 

References

“Classifying Information Objects: An Exploratory Ontological Excursion.” Sergey Zherebchevsky, Nicolette Ceo, Michiko Tanaka, David Jank, Richard Smiraglia and Stephen Stead. Poster at 10th International ISKO Conference, Montréal, 5-8 August 2008.

Miranda, Fabio, Harish Doraiswamy, Marcos Lage, Kai Zhao, Bruno Goncalves, Luc Wilson, Mondrian Hsieh and Claudio Silva. 2016. “Urban Pulse: Capturing the Rhythm of Cities.” IEEE Transactions on Visualization and Computer Graphics PP, 99:1-1. doi: 10.1109/TVCG.2016.2598585

“Mining Maps of Information Objects: An Exploratory Ontological Excursion.” Sergey Zherebchevsky, Nicolette Ceo, Michiko Tanaka, David Jank, Richard Smiraglia and Stephen Stead. Poster at American Society for Information Science and Technology Annual Meeting, Columbus Ohio, October 24, 2008.

Park, Hyoungjoo and Richard P.  Smiraglia. 2014. “Enhancing Data Curation of Cultural Heritage for Information Sharing: A Case Study using Open Government Data.” In Metadata and Semantics Research: 8th Research Conference, MTSR 2014, Karlsruhe, Germany, November 27‐29, 2014. Proceedings, ed. Sissi Closs, Rudi Studer, Emmanouel Garoufallou and Miguel-Angel Sicilia. Communications in Computer and Information Science 478: 95‐106.

Scharnhorst, Andrea, Richard P. Smiraglia, Alkim Almila Akdag Salah and Christophe Guéret. 2016. “Knowledge Maps of the UDC: Uses and Use Cases.” Knowledge Organization 43 forthcoming.

Smiraglia, Richard P. 2014a. “Classification Interaction Demonstrated Empirically.” In Knowledge organization in the 21st century: Between Historical Patterns and Future Prospects, Proceedings of the 13th International ISKO Conference, Krakow, Poland, May 19‐22, 2014, ed. Wiesław Babik. Advances in Knowledge Organization v. 14. Würzburg: Ergon‐Verlag, pp. 176‐83.

Smiraglia, Richard P. 2014b. “Extending the Visualization of Classification Interaction with Semantic Associations.” In Proceedings of the ASIST SIG/CR Classification Workshop, Seattle 1 November 2014.

Smiraglia, Richard P. 2013. “Big Classification: Using the Empirical Power of Classification Interaction.” In Proceedings of the ASIST SIG/CR Classification Workshop, Montréal, 2 November 2013, ed. D. Grant Campbell, p. 21‐29. doi: 10.7152/acro.v24i1.14673

Smiraglia, Richard P. and Hyoungjoo Park. 2016. “Using Korean Open Government Data for Data Curation and Data Integration.” DCMI 2016 OCS447 http://dcevents.dublincore.org/IntConf/index/pages/view/abstracts16#Smiraglia

On an Epistemic Center

A SIG/CR panel at this year’s ASIST annual meeting in Copenhagen was devoted to the concepts of global and local knowledge organization. It was a continuation, of sorts, of the 2015 conference held under the auspices of the Royal School of Library and Information Science at the University of Copenhagen,  orchestrated by Jens-Erik Mai, to consider whether approaching knowledge organization from the dual poles of global and local might shift the domain in new and useful directions from its standard dichotomy of universality versus domain-specificity. A report of the conference appeared in Knowledge Organization (Martinez-Avila 2015). The ASIST panel spent only a small portion of the allotted time in presentations, choosing instead to pose a series of questions in order to promote discussion.

One of the more intriguing ideas arose from John Budd, who spoke from the floor for a bit about phenomenological approaches to knowledge organization. I also have embraced phenomenological KO from time to time because of my conviction that a core problem in any form of KO is perception. A basic problem always exists because perceptions are not fixed, even in any particular individual, let alone in a community or among strangers. If KO is based ultimately on the atomic concept, how can concurrence on ontological distinctions ever be reached if every concept is subject to individual perception? There is no good answer to the question. There only are utilitarian explanations about what, in fact, is done in individual knowledge organization systems or applications to force perceptual concurrence. The central problem remains.

In his comments, which I by no means remember in full, Budd made reference to work by Paul Ricoeur in which the notion of “just institutions” plays an important role. Institutions that are just, in a phenomenological sense, must become so by arriving at, negotiating, navigating or even simply hovering over an epistemic center. Budd asked the group to consider whether there is an epistemic center in KO. If there were, presumably it would be located between and overlap the interests of both global and local considerations.

It’s an interesting question, by which I mean it is a critical question for KO, to consider how or whether an epistemic center, or epistemic centers, exist or function in the domain. If so they would provide a kind of unity, or maybe community is a better term, of perceptual conceptual loci. There are various approaches to phenomenology, of course, but most assert the role of the individual as a lens for perceiving reality. In Husserlian phenomenology (see Smiraglia 2014b, 28-29), noesis is the action by which the individual perceiving any entity sweeps his own ego for experienced perceptions and settles on, however briefly, a synthesis of experience that becomes in that moment the perception of that entity. When the entity is a concept, it means the noetic act involves sweeping and synthesizing cognizance based on experiential evidence. The problem for KO, of course, is that, if every individual is perceiving every concept individually, there is a high probability of misunderstanding from one individual to another or from individual to community. How can there be a knowledge organization based on concepts, if all concepts are subject to perception, and perception is a function of individual lived experience?

The answer has to lie in this notion of an epistemic core, a central gathering space as it were of overlapping perceptions that arrive at overlapping noetic synthesis and thus an agreed (conscious or not) ontology of any particular concept or set of concepts.

Such an epistemic center (epistemic because it relies on knowledge) has to be what we often refer to as culture. In Cultural Synergy in Information Institutions I wrote (Smiraglia 2014a, 1):

Cultural forces govern the synergistic relationship among information institutions and thus [shape] their roles collectively and individually. Synergy is that combination of forces whose power is greater than the individual power of its constituent elements. Culture is that base of knowledge that is common to any particular group of people, such that it shapes their perception as well as their behavior as a group and as members of that group. Cultural synergy, then, is the combination of perception- and behavior-shaping knowledge within, between, and among groups that contributes to the now realized virtual reality of a common information-sharing interface among information institutions.

Culture then, such as it is, must be the ethos of an epistemic center. If so, it must necessarily be a dynamic space, much like Peirceian semiotic space (see Smiraglia 2014b, 23-26), in which a constant process of synthesis on the part of individuals and groups sweeps experience for perceptual understanding arriving momentarily simultaneously on overlapping ontologies of concepts. (By ontology of a concept I mean its definitive boundaries, and the factors that determine what is or is not an exemplar.) Such a dynamism is the combination of semiosis with noesis.

The epistemic center must be the space in the universe of knowledge where perception takes place, leading to signs and concepts, the combination of which constitute works, which in turn constitute taxons, which constitute canons that represent cultures (see Smiraglia and van den Heuvel 2013, 374). The epistemic center (or any epistemic center) is the working place from which knowledge organization arises; it is the foundry where Paul Otlet’s grinder chugs away rearranging perceived knowledge into newly discoverable clusters (378). Culture, in all of its meanings, defines the boundaries of epistemic center. Ontological boundaries must therefore be constantly shifting (this we know already from common sense and more recently, empirically from ontogenetic studies of KO).

 

References

Martínez-Ávila, Daniel. 2015. “Global and Local Knowledge Organization, Copenhagen, August 12, 2015. Knowledge Organization 42: 470-3.

Smiraglia, Richard P. 2014a. Cultural Synergy in Information Institutions. New York: Springer.

Smiraglia, Richard P. 2014b. The Elements of Knowledge Organization.  Cham: Springer.

Smiraglia, Richard P. and Charles van den Heuvel. 2013. “Classifications and Concepts: Towards an Elementary Theory of Knowledge Interaction.” Journal of Documentation 69: 360-83.

Posted October 23, 2016 by lazykoblog in epistemology, phenomenology

Tagged with , ,

Teaching epistemology

For years students and colleagues rolled their eyes any time I said anything about epistemology. Once, famously, a paper submitted for a major international conference was accepted with revisions, one of which was to explain what that was because the readers (all holders of PhDs mind you) didn’t know what epistemology was, and (apparently) were unable to look it up or otherwise figure it out. Sigh.

But then epistemology became an important part of the science of knowledge organization, originating in papers by diverse scholars, leading to panels at international ISKO conferences and eventually even two anthologies (Smiraglia and Lee 2012; Ibekwe-SanJuan and Dousa 2014). Epistemology, how we know what we know, is one of the two major poles in a domain-centric knowledge organization.

So the question arises, how can we teach it without getting everybody dizzy from eye-rolling? I took a stab at it recently and I think it worked out nicely. In my introductory course in KO I often begin class discussions by asking the students to post an observation about some thing that has some order. This time around I got everything from a grandmother’s farmer’s market vegetable table to how big-box hardware stores hide microfiber cloths. I enjoy responding to each post, pointing out the nature of the organization–hierarchy for the vegetables, for example. It helps to point out to them that everything everywhere is organized in some manner, not just libraries.

This time I followed up with a secret epistemology discussion by asking them to post something they absolutely knew for sure. I gave them a couple of examples: “I know George Washington was president because it’s history I was taught; I never met the man. I know the speed limit on Lincoln Memorial Drive is 30 but I’d better drive 70 or people will plow into me or drive me into the lake. I know this because it happens to me daily. Of course, we have here examples of historicism and empiricism (okay, facetious empiricism, but if you live in Milwaukee you’ll get it). They did a great job. Here’s a table of a few of the things they came up with:

fastest routes to campus empirical
swimming must blow air out your nose empirical
Rush has 3 members … historical
death and taxes historical
difficult to buy a car with manual transmission empirical
get out of the pool if there’s thunder and lightning pragmatic
yellow mustard eases pain of small burns empirical
pick from the back when shopping pragmatic

 

I think it turned out well, and although we didn’t go further into epistemology in this introductory course, it allowed me to reference this discussion at the end when they were exploring the heirarchies in DDC numbers assigned to specific resources, which subtly makes the point about the role of epistemic stances.

References

Ibekwe-SanJaun, Fidelia and Thomas M. Dousa, eds. 2014. Theories of Information, Communication and Knowledge: A Multidisciplinary Approach. Dordrecht: Springer-Netherlands.

Smiraglia, Richard P. and Hur-li Lee, eds. 2012. Cultural Frames of Knowledge. Würzburg: Ergon-Verlag.

The order of all things*

p. 83-84: Descartes made one of the first attempts to explain the concept of the affections in his last book, The Passions of the Soul of 1649 …. Descarte’s book was followed a year later by the Musurgia universalis of the “last Renaissance man,,” Athanasius Kircher …. Kircher was very much a believer in music not only as a mirror but as a fundamental element of Creation itself …; music was not so much a reflection or approximation of God’s perfect design, but an emanation of the divine Itself. Kircher’s motto was: “Music is nothing other than the knowledge or the order of all things.”

James R. Gaines, Evening in the Palace of Reason: Bach Meets Frederick the Great in the Age of Enlightenment. New York: Harper Perennial, 2006.

 

Sinecure and credentialism

I’ve been reading a book by Randall Collins. His book The Sociology of Philosophies (1998) made a major impact on me. In a direct way, and it was for this reason a dutch colleague referred me to the book, his demonstration of splits and concretizings of intellectual circles gave us an hypothesis to use in domain analytical research. It boiled down to the notion that no school of thought can tolerate more than six theoretical paradigms (or ideas if you prefer) at once without either splitting apart (schism, as it were) or ejecting competing points of view until a tolerable level is restored. I first went looking for this in the evolving domain of music information retrieval and sure enough, as a new research interest it opened its arms widely and expanded rapidly, but once query-by-humming had been achieved the society (it had been first a symposium, but at this point it became a society with bylaws, officers, and a scope statement) quickly reigned in acceptable research. Others have found similar evidence (Hoeffner in Social informatics, for example, LIU diss. 2012).

It’s a marvelous book all by itself of course, if very long. The critical realization I had while reading it was that I had spent almost two decades working with doctoral students who not only were not forming a school of thought around my work, but likely would never conduct any research after their dissertations, and certainly would never contribute to growth of my theory of instantiation. (Although, to be fair, near the end of my time at LIU several students did take up domain analysis and contribute to the cumulative effect of domain analytic research in knowledge organization, although to the best of my knowledge none of them has conducted or published any follow-up work.) And with that I determined to move elsewhere while I still had energy to take on new students.

Somewhere last fall I read about Collins’ much shorter 1979 work The Credential Society, and I’ve just finished reading it after proudly hauling it to Copenhagen, Heraklion, Amsterdam and back. It also is a remarkable work, hence this post, and I think it will have something to contribute to domain analysis, although at present I’m not quite sure how.

The book is about the myth of technocracy, that as society evolves and technology becomes more complex and we become ever more highly educated so as to deal with technology and (here’s the myth) therefore society gets better, people get richer, everything becomes more egalitarian, etc., etc. You’ll recognize the myth. In a short 220some pages Collins shatters this myth, demonstrating that no amount of education has made any difference and neither has technology. In fact, the only evidence about career and social status that makes sense is the age-old truth that (male) children follow in their father’s footsteps in both career and social status. I know, you want to protest, and so do I. My parents weren’t professors (but, my biological father was a musician and my biological maternal great-grandfather was an ordained pastor … hmmm). As usual, I’m not doing the book justice, you’ll have to read it.

Two things stood out for me. First the notion that we have accomplished a sinecure society. Sinecure, he writes, means literally “without care” and refers to a job in which there is actually little work. Collins points out that society has succeeded at installing a sinecure society by making most work into what once would have been leisure. Most of us read, write and think as work these days. Once upon a time that would have been a life reserved to only those who did not need to work.

The other idea, and here probably is the connection to domain analysis, is that professions secure their concretization and hence their survival with rather medieval approaches to credentialism. The easiest example from the book is the practice of medicine, which has the highest status and salary potential in our society, and for which the education (which is lengthy and expensive) has almost nothing to do with the practice except to confer the credential. As Collins shows, most medical practitioners learn on the job from other practitioners. But the mostly upperclass male medical doctors have succeeded with their credentialing in shutting out all other actors in healthcare from orderlies to nurses to social workers, most of whom labor for little in feminized professions. I know, this isn’t pretty. Collins takes on engineering and law too, but I won’t go into that here.

Along the way his narrative about the evolution of higher education in the US through the 18th, 19th and 20th centuries is about the clearest explanation I’ve seen, although it is consistent with the same trajectory painted by Louis Menand in The Metaphysical Club (erm, can you paint a trajectory? oh well, a topic for another time). All of us who make our livings and livelihoods as professors ought to have a better sense of how things got this way and I recommend both books for that reason.

I’ll keep pondering this of course. I think somehow we might be able to discover that a concretized domain is also somehow credentialed. Evidence of that might be useful for determining who the relevant actors are in the evolution of a domain ontology at any given moment. As it said, it needs some pondering.

 

Cataloging rules, Michael Gorman, memories ….

I’ve been working quite a lot lately with RDA, and that also means working with AACR2. (A paper called “Bibliocentrism Revisited” was presented at the 3rd Milwaukee Conference on Ethics in Knowledge Organization; it will appear in print in Knowledge Organization later this year. I also have just taught music cataloging entirely with RDA, which was an experience, all the while rewriting Describing Music Materials, which will appear in a 4th edition conformed to RDA soon, I hope.)

Somewhere along the line one of those helpful boxes on the right of a search screen showed me an autobiographical work by Michael Gorman, Broken Pieces: A Library Life, 1941-1978. I was looking for something to read on a short trip so I ordered a copy (ALA Editions, 2011). I was surprised I hadn’t heard about the book before.

I worked for Michael Gorman at the University of Illinois at Urbana-Champaign Libraries in the late 1970s and early 1980s. I was then head of music cataloging, which at various times was a department of the music library or a division of technical services (long story for my autobiography!). I remember early in our acquaintance, we met at a social event in town and chatted briefly about cataloging rules, because he was then finishing AACR2 and I was then becoming more active in CC:DA, the ALA committee that in those days was responsible for the ALA position on catalog code revision. I remember he told me he had a copy of the draft I could read through, and I remember that I did read through it and was startled at all of the uproar about basically nothing.

I also remember that when the Library of Congress decided not to implement AACR2 for awhile (3 years eventually), he decided we would go ahead at UIUC. This was remarkably problematic for music cataloging, because rules for the construction of music uniform titles had changed, and rules for name headings had changed. It meant we had to recreate authority records for almost everything for awhile, because unlike book libraries where the imposition of old name headings helped ease the way, we had a smaller set of composer’s names for whom every record had to change and every work needed a new uniform title. Nobody else in the world was interpreting AACR2, and that meant we had to go it alone for rather awhile. And that meant our productivity plummeted. I do remember the pivotal meeting where this was discussed–I’ll save that story for my autobiography too.

At any rate Michael’s book is interesting. There is rather a lot of childhood and young adult narrative before he gets to the “library life” part. His stories about moving from England to Champaign-Urbana are fascinatingly funny. I was living there at t he time, of course, and remember it well. But then there is quite a lot of fascinating detail about his work in public libraries as a young man, and about the technology then in use in the 1960s and 1970s. His narrative about the development of AACR2 helped me fill in some gaps in my own memories. I had hoped for rather more about his time at Illinois, but then, this narrative ends in 1978, so perhaps that will be forthcoming.

We don’t often include descriptive cataloging when we talk about knowledge organization. On the face of it, descriptive cataloging is not obviously a part of conceptual ordering. But the resource description part of it is a form of knowledge representation, and the indexing function of ordering works by name headings is form of alphabetical classification. Names of “creators” form classes, within which the titles of “works” form divisions, and the representations of specific pubilcations grouped variously form subdivisions. Insofar as works are conceptual ideations, their ordering is a form of conceptual ordering.

It is in this phase that we can understand the importance of the argument that took place around the writing of AACR2 about what was called “main entry,” because it really was about how to name these classes. There is a long history of attribution of creativity to iconic names of creators (i.e.,”authors”). How to preserve that function without the mechanical superimposition of a name above a title transcription wasn’t clear at the time. Lubetzky was both right and wrong on this; the creator-title citation was critical to keep as the name of a class of works, but the Animal Farm-ish approach to which access point was more equal than the others was not. In this RDA has made a small step forward, by separating resource description from ordering of works.

Vanishing point

Recently our research group hosted the Third Milwaukee Conference on Ethics in Knowledge Organization. It was chock full of good things, and I’ll try to write about that later.

It was the occasion of the publication of our group’s second book Ontology for Knowledge Organization.

As part of the celebration we had a small “exhibit” of our two books (the other one was the 2012 Cultural Frames of Knowledge) and some example issues of Knowledge Organization together with some posters provided by Ergon Verlag.

Curiously, all of the example issues of KO and some of the “for-sale” copies of our new book vanished into thin air.

Posted June 16, 2015 by lazykoblog in Uncategorized