Miskatonic University Press

Putting Facets on the Web: An Annotated Bibliography

Denton, William. “Putting Facets on the Web: An Annotated Bibliography” Oct. 2003. https://www.miskatonic.org/library/facet-biblio.html.

I wrote this for Prof. Clare Beghtol of the Faculty of Information Studies at the University of Toronto, who led me in a reading course named "Applying Faceted Classification in an Online World." After this, I wrote How to Make a Faceted Classification and Put It On the Web.

Introduction

This is a classified, annotated bibliography about how to design faceted classification systems and make them usable on the World Wide Web. It is the first of three works I will be doing. The second, based on the material here and elsewhere, will discuss how to actually make the faceted system and put it online. The third will be a report of how I did just that, what worked, what didn't, and what I learned.

Almost every article or book listed here begins with an explanation of what a faceted classification system is, so I won't (but see Steckel in Background below if you don't already know). They all agree that faceted systems are very appropriate for the web. Even pre-web articles (such as Duncan's in Background, below) assert that hypertext and facets will go together well. Combined, it is possible to take a set of documents and classify them or apply subject headings to describe what they are about, then build a navigational structure so that any user, no matter how he or she approaches the material, no matter what his or her goals, can move and search in a way that makes sense to them, but still get to the same useful results as someone else following a different path to the same goal. There is no one way that everyone will always use when looking for information. The more flexible the organization of the information, the more accommodating it is. Facets are more flexible for hypertext browsing than any enumerative or hierarchical system.

Consider movie listings in newspapers. Most Canadian newspapers list movie showtimes in two large blocks, for the two major theatre chains. The listings are ordered by region (in large cities), then theatre, then movie, and finally by showtime. Anyone wondering where and when a particular movie is playing must scan the complete listings. Determining what movies are playing in the next half hour is very difficult. When movie listings went onto the web, most sites used a simple faceted organization, always with movie name and theatre, and perhaps with region or neighbourhood (thankfully, theatre chains were left out). They make it easy to pick a theatre and see what movies are playing there, or to pick a movie and see what theatres are showing it. To complete the system, the sites should allow users to browse by neighbourhood and showtime, and to order the results in any way they desired. Thus could people easily find answers to such questions as, "Where is the new James Bond movie playing?" "What's showing at the Roxy tonight?" "I'm going to be out in in Little Finland this afternoon with three hours to kill starting at 2 ... is anything interesting playing?" A hypertext, faceted classification system makes more useful information more easily available to the user.

Reading the books and articles below in chronological order will show a certain progression: suggestions that faceting and hypertext might work well, confidence that facets would work well if only someone would make such a system, and finally the beginning of serious work on actually designing, building, and testing faceted web sites. There is a solid basis of how to make faceted classifications (see Vickery in Recommended), but their application online is just starting. Work on XFML (see Van Dijck's work in Recommended) the Exchangeable Faceted Metadata Language, will make this easier. If it follows previous patterns, parts of the Internet community will embrace the idea and make open source software available for others to reuse. It will be particularly beneficial if professionals in both information studies and computer science can work together to build working systems, standards, and code. Each can benefit from the other's expertise in what can be a very complicated and technical area. One particularly nice thing about this area of research is that people interested in combining facets and the web often have web sites where they post their writings.

This bibliography is not meant to be exhaustive, but unfortunately it is not as complete as I wanted. Some books and articles are not be included, but they may be used in my future work. (These include two books and one article by B.C. Vickery: Faceted Classification Schemes (New Brunswick, NJ: Rutgers, 1966), Classification and Indexing in Science, 3rd ed. (London: Butterworths, 1975), and "Knowledge Representation: A Brief Review" (Journal of Documentation 42 no. 3 (September 1986): 145-159; and A.C. Foskett's "The Future of Faceted Classification" in The Future of Classification, edited by Rita Marcella and Arthur Maltby (Aldershot, England: Gower, 2000): 69-80). Nevertheless, I hope this bibliography will be useful for those both new to or familiar with faceted hypertext systems. Some very basic resources are listed, as well as some very advanced ones. Some example web sites are mentioned, but there is no detailed technical discussion of any software. The user interface to any web site is extremely important, and this is briefly mentioned in two or three places (for example the discussion of lawforwa.org (see Example Web Sites)). The larger question of how to display information graphically and with hypertext is outside the scope of this bibliography.

There are five sections: Recommended, Background, Not Relevant, Example Web Sites, and Mailing Lists. Background material is either introductory, advanced, or of peripheral interest, and can be read after the Recommended resources if the reader wants to know more. The Not Relevant category contains articles that may appear in bibliographies but are not relevant for my purposes.

Recommended

Broughton, Vanda. "Faceted Classification as a Basis for Knowledge Organization in a Digital Environment; the Bliss Bibliographic Classification as a Model for Vocabulary Management and the Creation of Multi-Dimensional Knowledge Structures." The New Review of Hypermedia and Multimedia 7 (2001): 67-102.

Broughton is one of the key people working on the second edition of the Bliss Bibliographic Classification (BC2). Her article has a brief, informative history of facets, then discusses semantic vs. syntactic relationships, standard facets used by Ranganathan and the Classification Research Group, facet analysis and citation order, and how to build subject indexes out of faceted classifications, all with occasional reference to digital environments and hypertext, but never with any specifics. It concludes by saying of faceted classification that the "capacity which it has to create highly sophisticated structures for the accommodation of complex objects suggests that it is worth investigation as an organizational tool for digital materials, and that the results of such investigation would be knowledge structures of unparalleled utility and elegance." How to build them is left to the reader, but this article provides an excellent starting point. It includes an example that shows how general concepts can be applied to a small set of documents and subjects, and how terms can be adapted to suit the material and users.

Doyle, Bob. "The Classification & Evaluation of Content Management Systems." The Gilbane Report 11 no. 2 (March 2003): 2-13. http://www.gilbane.com/artpdf/GR11.2.pdf.

This is a report on how Doyle and others made a faceted classification scheme for content management systems and made it browsable on the web (see CMS Review in Example Web Sites, below). They discuss why they did it, how, their use of OPML and XFML, how they did research to find terms and categories, and they also include their taxonomy. It is interesting to see facets used in a business environment.

Fast, Karl, Fred Leise, Mike Steckel. "0. All About Facets and Controlled Vocabularies" Boxes and Arrows, 9 December 2002. http://www.boxesandarrows.com/archives/all_about_facets_controlled_vocabularies.php.

——— "1. What Is a Controlled Vocabulary?" Boxes and Arrows, 16 December 2002. http://www.boxesandarrows.com/archives/what_is_a_controlled_vocabulary.php.

——— "2. Creating a Controlled Vocabulary" Boxes and Arrows, 7 April 2003. http://www.boxesandarrows.com/archives/creating_a_controlled_vocabulary.php.

——— "3. Synonym Rings and Authority Files" Boxes and Arrows, 26 August 2003. http://www.boxesandarrows.com/archives/synonym_rings_and_authority_files.php.

——— "4. Controlled Vocabularies: A Glosso-Thesaurus" Boxes and Arrows, 27 October 2003. http://www.boxesandarrows.com/archives/controlled_vocabularies_a_glossothesaurus.php.

——— "Facets and Controlled Vocabularies: An Annotated Bibliography" Boxes and Arrows, n.d. http://www.boxesandarrows.com/archives/Facets_CV/Bibliography.htm.

An online series of articles explaining controlled vocabularies and, in particular, faceted classification. It is not yet finished, but what they have covered is very well done, practical and informative, with useful advice and a full treatment. It is worth reading now, and when they actually get to performing facet analysis and making a faceted system, it will make a very useful reference.

Kwasnick, Barbara H. "The Role of Classification in Knowledge Representation and Discovery." Library Trends 48 no. 1 (Summer 1999): 22-47.

A fascinating, broad-ranging article about classification, knowledge, and how they relate. Hierarchies, trees, paradigms (a two-dimensional classification that can look something like a spreadsheet), and facets are covered, with descriptions of how they work and how they can be used for knowledge discovery and creation. Kwasnick outlines how to make a faceted classification: choose facets, develop facets, analyze entities using the facets, and make a citation order. Facets are useful for many reasons: they do not require complete knowledge of the entire body of material; they are hospitable, flexible, and expressive; they do not require a rigid background theory; they can mix theoretical structures and models; and they allow users to view things from many perspectives. Facets do have faults: it can be hard to pick the right ones; it is hard to show relations between them; and it is difficult to visualize them. The coverage of the other methods is equally thorough and there is much to consider for anyone putting a classification on the web.

Louie, Aaron J., Eric L. Maddox, and William Washington. "Using Faceted Classification to Provide Structure for Information Architecture." Paper presented at the ASIS&T 2003 Information Architecture Summit, Portland, OR, 21-23 March 2003. http://depts.washington.edu/pettt/presentations/conf_2003/IASummit.pdf.

This is a short, but very thorough and very interesting, report on how the writers built a faceted classification for some legal information and used it to structure a web site with navigation and searching. There is a good summary of why facets work well and how they fit into bibliographic control in general. The last section is about their implementation of a web site for the Washington State Bar Association's Council for Legal Public Education. Their classification uses three facets: Purpose (the general aim of the document, e.g. Resources for K-12 Teachers), Topic (the subject of the document), and Type (the legal format of the document). See Example Web Sites, below, for a discussion of the site and a problem with its design.

A very large PDF of the six-foot-wide illustrated poster from their poster session is available at http://depts.washington.edu/pettt/presentations/conf_2003/IASummit-Poster-Louie.pdf.

Priss, Uta, and Elin Jacob. "Utilizing Faceted Structures for Information Systems Design," in ASIS '99: Proceedings of the 62nd ASIS Annual Meeting, Washington, D.C., October 31 - November 4, 1999: Knowledge: Creation, Organization, and Use, edited by Larry Woods, 203-212. (Medford, NJ: ASIS, [1999?]). http://www.upriss.org.uk/papers/asis99.pdf.

The writers show that a faceted navigation structure makes web sites easier to use. They begin by analyzing the web sites of three library and information science faculties, and seeing if the sites easily provide the answers to five specific questions, e.g., how the school ranks in national evaluations. (It is worth noting that the web site of the Faculty of Information Studies and the University of Toronto, where this bibliography is being written, would fail on four of the five questions.) Using examples from LIS web site content, they show how facets can be related and constructed, and use concept diagrams for illustration. They briefly discuss constraints necessary when joining facets: for example, enrolled students can be full- or part-time, but prospective and alumni students cannot. It should not be possible to construct terms such as "part-time alumni" (see Yannis Tzitzikas et al, below in Background). They conclude that a faceted approach is best for web site navigation, because it can clearly show where the user is in the site, what the related pages are, and how to get to them. There is a short discussion of user interfaces, and the diagrams in the paper will be of interest to anyone making a facet-based web site. This paper is clearly written, informative, and thought-provoking. Uta Priss's web site lists her other publications, many of which are related and some of which are online: http://www.upriss.org.uk/top/research.html.

Ranganathan, S.R. Elements of Library Classification. 3rd ed. New York: Asia Publishing House, 1962.

A medium-length book, based on lectures, aimed at classificationists, not cataloguers or classifiers. Ranganathan begins with his Five Laws and a definition of classification and its purposes. He gives a list of 108 subjects in "unhelpful alphabetical sequence" and shows how they can be grouped into subjects, and then how each each subject's terms can be organized in a helpful and useful way, thereby demonstrating and building up his basic canons, postulates, and principles of classification. All of that, roughly the first half of the book, will be of interest to anyone starting to make classification systems. It has all of what makes Ranganathan's work so delightful to read: his unending concern for the user, his deep thought, and his warmth, humour, and spirituality. The second half of the book, however, has what can make his work difficult: his unyielding belief that the Colon Classification is the only system worth using. The reader will not be very interested in repeatedly classifying books under various systems and then reversing the process to see how closely the call number matches the subject. However, the reader can take the ideas so clearly presented in the first half of the book and then veer off to build his or her own system, while remembering that if his classification laws are likened to Euclid's laws of geometry, there is no room for a Riemann or Lobachevksy in Ranganathan's strict world.

Spiteri, Louise. "A Simplified Model for Facet Analysis: Ranganathan 101." Canadian Journal of Information and Library Science 23 no. 1/2 (April-July 1998): 1-30. http://aifia.org/pg/a_simplified_model_for_facet_analysis.php.

Ranganathan's canons, principles, and postulates can easily confuse readers, especially because he revised and added to them in various editions of his many books. The Classification Research Group, who drew on Ranganathan's work as their basis for classification theory but developed it in their own way, has never clearly organized all their equivalent canons and principles. In this article Spiteri gathers the fundamental rules from both systems and compares and contrasts them. She makes her own clearer set of principles for constructing facets, stating the subject of a document, and designing notation. Spiteri's "simplified model" is clear and understandable, but certainly not simplistic. The model does not include methods for making a faceted system, but will serve as a very useful guide in how to turn initial work into a rigorous classification. Highly recommended.

Van Dijck, Peter. "Introduction to XFML." xml.com, 22 January 2003. http://www.xml.com/lpt/a/2003/01/22/xfml.html.

Van Dijck builds up an example of actual XFML by showing how to organize tourist information about what restaurants in what cities feature which kind of music: <facet id="city">City</facet> and <topic id="ny" facetid="city"><name>New York</name></topic> combine to mean that New York is the name of a city internally represented as "ny". It is written in the usual clear and practical style of articles on xml.com. Highly recommended as an introduction for anyone interested in XFML.

——&mdash "XFML Core - eXchangeable Faceted Metadata Language." 2003. http://www.xfml.org/spec/1.0.html.

The specification for XFML, a markup language designed to handle faceted classifications. Browsing the site (http://www.xfml.org/) will reveal news about XFML and links to related software and web sites. XFML is not an officially recognized Internet standard, but is the de facto standard.

Vickery, B.C. Faceted Classification: A Guide to Construction and Use of Special Schemes. London: Aslib, 1960.

A perfect little book, with just 63 pages of text. From chapter A, Introduction, to U, Mechanization, it covers everything about making a faceted classification: what they are, why they are needed, how to do facet analysis, examples from existing faceted schemes, orderings, common subdivisions, the contents of each facet, notation, filing order, how to perform classification with the created system, and indexing. Each chapter is brief but has full coverage of the subject. "The technique of constructing a special faceted classification is not a settled, automatic, codified procedure. Nothing so complex as the field of knowledge could be analysed and organized by rule-of-thumb. We can therefore offer no more than a guide, describing tested procedures and discussing some difficulties." Vickery was a member of the Classification Research Group and one of the foremost classificationists.

Background

Broughton, Vanda, and Heather Lane. "Classification Schemes Revisited: Applications to Web Indexing and Searching." Journal of Internet Cataloguing 2 no. 3/4 (2000): 143-155.

A short discussion of using classification systems to organize the web, one of many such. The authors are both involved with BC2 and naturally think it is the best system for organizing information online. They list reasons why faceted classifications are best (e.g. no theoretical limits to specificity or exhaustivity; easier to handle complex subjects; flexible enough to accommodate different user needs) and take a brief look at how BC2 works. They conclude with a discussion of how and why it should be applied to online resources, and a plea for recognition of the importance of classification and subject analysis skills, even when full-text searching is available and databases respond instantly.

Chan, Lois Mai, Eric Childress, Rebecca Dean, Edward T. O'Neill, and Diane Vizine-Goetz. "A Faceted Approach to Subject Data in the Dublin Core Metadata Record." Journal of Internet Cataloging 4 no. 1/2 (2001): 35-47.

This article describes FAST, the Faceted Application of Subject Terminology, a project at OCLC to make Library of Congress Subject Headings easier to use in Dublin Core metadata by breaking out facets of space, time, and form. Work on FAST can be watched at its web site, http://www.oclc.org/research/projects/fast/, which has recent presentations and reports. It is interesting to see facets and Dublin Core combined, though both LCSH and FAST subject headings are beyond what most people making a small faceted classification would want or need.

Duncan, Elizabeth B. "Structuring Knowledge Bases for Designers of Learning Materials." Hypermedia 1 no. 1 (Spring 1989): 20-33.

——— "A Faceted Approach to Hypertext." In Hypertext: Theory Into Practice, ed. Ray McAleese (Oxford: Intellect, 1989): 157-163.

——— "A Concept-Map Thesaurus as a Knowledge-Based Hypertext Interface to a Bibliographic Database." In Informatics 10: Prospects for Intelligent Retrieval, ed. Kevin P. Jones (London: Aslib, 1990): 43-52.

Three pre-web articles about using hypertext for knowledge representation. Duncan discusses how to use graphical, hypertext displays (she used Xerox PARC's NoteCards on a Xerox 1186 workstation) along with concept maps and facet analysis, a combination that would now be done with topic maps. The screen shots of her graphical displays are quite interesting. Her interest in facets is in how to use them to show things to different people in different ways, for example, so that experts can enter knowledge into a system in one way while novices can see it in another. Duncan found that facet labels (e.g. Process and Product) prompted the expert to think of related concepts when inputting data, and made navigation easier for users. Facets can be joined together, e.g. "Agents (causing) Process," leading to a "reasoning system." She is especially interested in how to show relstionships between two things: e.g., A causes B, A uses B, A occurs in B. This is an important question in facet theory, but probably not worth worrying about in a small online classification where the relations are fixed and obvious.

These articles may be difficult to find, in which case the reader can find a nice sumary in the next article, by Ellis and Vasconcelos (2000). Anyone interested in tracing the history of facets and hypertext will, however, want to see the originals.

Ellis, David, and Ana Vasconcelos. "The Relevance of Facet Analysis for World Wide Web Subject Organization and Searching." Journal of Internet Cataloging 2 no. 3/4 (2000): 97-114.

This is a revised version of the earlier article by Ellis and Vasconcelos (1999) (see Not Relevant, below), though that is not indicated, and much of it is identical, word for word. There is a new section covering the work of Elizabeth Duncan, which is useful and informative, but the reader is better advised to go to the originals if available.

Foskett, D.J. "Ranganathan and 'User-Friendliness.'" Libri 42 no. 3 (1992): 227-234.

Ranganathan always showed a great concern for the user and all the ways he or she would use a library. His work is no less applicable in the computer age, and principles such as the Five Laws of Library Science are valid no matter how information is sought. Foskett discusses user friendliness and the usefulness of facet analysis in online systems, which he says it will work very well for information storage, retrieval, and searching. Time will undoubtedly prove him correct.

The Knowledge Management Connection. "Faceted Classification of Information." http://www.kmconnection.com/DOC100100.htm.

An explanation of faceted classification meant for people working in knowledge management. An example given for a high-technology company has the fundamental categories Products, Applications, Organizations, People, Domain objects ("technologies applied in the marketplace in which the organization participates"), Events (i.e. time), and Publications.

Steckel, Mike. "Ranganathan for IAs." Boxes and Arrows, 7 October 2002. http://www.boxesandarrows.com/archives/ranganathan_for_ias.php.

A short but useful introduction to S.R. Ranganathan, his laws of library science, and how they apply to the web, and the Colon Classification. There is nothing new here to anyone with a more than passing acquaintance with the man and his work, but that is a small subset of the world's population. For the rest, this is a good, short place to start. It would be useful for anyone wanting to explain to non-information studies people why they want to start work on a faceted classification.

Tzitzikas, Yannis, Anastasia Analyti, Nicolas Spyratos, and Panos Constantopolous. "An algebraic approach for specifying compound terms in faceted taxonomies," in Proceedings of the 13th European-Japanese Conference on Information Modelling and Knowledge Bases, EJC’2003 (Kitakyushu, Japan: 2003). http://www.csi.forth.gr/~tzitzik/XFML+CAMEL/papers/CoTeCAlgebra_paper.pdf.

Tzitzikas, Yannis, Nicolas Spyratos, Panos Constantopolous, and Anastasia Analyti. "Extended Faceted Taxonomies for Web Catalogs." Paper presented at the Third International Conference on Web Information Systems Engineering, WISE 2002, Singapore, December, 2002. Gzipped PostScript version available at http://www.csi.forth.gr/~tzitzik/publications/Tzitzikas_WISE_2002.ps.gz.

These two papers are very mathematical and use lots of formulas, but that should not frighten anyone interested in current advanced work on facets and their use online. The articles cover much the same matter, but the second would be the best one to read first, since it has fewer formal proofs and more background on facets and web navigation. The writers have designed algorithms that will make it possible to easily define which facet combinations are impossible in a given classification system. Because of the way facets work, as compared to lengthy enumerative systems, a few restrictions can have wide-ranging effect on the entire scheme. This information can guide interfaces and prevent the user from ever being able to ask for something impossible. Their examples involve Greek resorts and sports. A person planning a trip to Greece might want to get a list of what sports are possible on Crete. Boating is an option, but skiing is not. However, skiing is possible on Mt. Olympus. The system described in these papers would make it possible for the software behind a tourism web site's interface to stop such impossible options from ever being shown to the user. (There are other ways to do these restrictions, such as manually entering every possible combination in every facet ordering, but they would be very laborious.) If these algorithms are turned into working code and made available in easy to use software packages for popular programmings languages, online faceted classification would benefit immensely.

Tzitzikas's online list of publications at http://www.csi.forth.gr/~tzitzik/publications.htm has many more related papers available.

Not Relevant

Devadason, Francis J., Neelawat Intaraksa, Ponprapa Patamawongjariya, Kavita Desai. "Faceted Indexing Based System for Organizing and Accessing Internet Resources." Knowledge Organization 29 no. 2 (2002): 65-77.

An interesting but somewhat confusing article telling how the writers described web pages with Dublin Core metadata, including a faceted classification, and built a system that lets users browse the collection through the facets. They seem to want to cover too much in a short article, and unnecessary space is given over to screen shots showing how Dublin Core metadata was entered. The screen shots of the resulting browsable system are, unfortunately, not as enlightening as one would hope, and there is no discussion of how the system was actually written or the technology behind it. Still, it could be worth reading as an example of such a system and how it is treated in journals.

Ellis, David, and Ana Vasconcelos. "Ranganathan and the Net: Using Facet Analysis to Search and Organise the World Wide Web." Aslib Proceedings 51 no. 1 (Jan 1999): 3-10.

This article gives a cheerfully brief and undetailed account of how to make a faceted classification system, then describes information retrieval and searching on the web. It concludes by saying that facets would be excellent in helping users search and browse the web, but offers no real clues as to how this can be done.

Gödert, Winfried. "Facet classification in online retrieval." International Classification 18 no. 2 (1991): 98-109.

"Online retrieval" conjures up a very different mental image now than in 1991, the year this article was written, and the year Tim Berners-Lee first revealed the new hypertext system he called the World Wide Web. Gödert shows that truncation and Boolean logic, combined with notation from a faceted classification system, will be a powerful way of searching for information. It undoubtedly is, but no system built now would require a user searching for material on "nervous systems of bone fish" to enter "Fdd$ and Leaa$". This is worth reading for someone interested in seeing how searching and facets can go together, but the web has made this article quite out of date.

Example Web Sites

CMS Review (http://www.cmsreview.com/Directory.html)

As discussed in Doyle (2003) above (see Recommended). CMS Review made their own faceted classification system for content management systems (software that makes it easy to manage web sites, especially large ones). The user can start to browse it and choose one facet (perhaps operating system), then another (perhaps price), then another, in any order, and the available systems fitting those restrictions will be listed. Content management systems are complicated, and there are many of them. This site makes it easy to find systems that match user needs and make comparisons between any two. It would be nice if it was easier to rearrange facet ordering, but that is not simple to do in a web page. The work here is of extra interest because content management system vendors may not be interested in making it easy to compare everything about their product with another vendor's. Users will benefit if it is easy for them to find out what others have already unearthed, and to share their own work.

Dive Into Accessibility

In his weblog entry "This is XFML" (http://diveintomark.org/archives/2002/12/03/this_is_xfml, 3 December 2002) Mark Pilgrim says, "XFML is a new format for providing hierarchical faceted metadata.... Frankly, I couldn't make heads or tails of it until a kind soul (Albert de Klein) mocked up an XFML representation of Dive Into Accessbility, my tutorial on web accessibility techniques. Then it all became clear." With XFML he could make a version of his site (http://www.diveintoaccessibility.org/) that looked like a portal (http://facetmap.com/demo/browse.jsp?map=diveintoaccessibility&v=1&s=000000000000) or that was searchable and browsable by facet (http://facetmap.com/demo/browse.jsp?map=diveintoaccessibility&v=2&s=000000000000). Dive Into Accessibility is the sort of web site that is easy to organize into facets and benefits greatly from doing so. Compare the original version of the site with the facet-based interfaces created with XFML to see the improvements.

Epicurious (http://www.epicurious.com/)

Epicurious has thousands of recipes and has organized them into a faceted classification, by Main Ingredient, Cuisine (ethnic origin), Special Considerations (e.g. low fat), Preparation Method, Season/Occasion, and Course/Dish. The user can start browsing the collection through the main headings in each facet (http://eat.epicurious.com/recipes/browse_home/index.ssf?/recipes/browse_home/index.html) and narrow the focus as desired, or use the advanced search (http://eat.epicurious.com/recipes/enhanced_search/index.ssf/?/recipes/enhanced_search/index.html) which lets the user build up, in a straighforward way, a complicated search by picking and choosing which headings in which facets should be included.

FacetMap's wine demonstration (http://facetmap.com/browse.jsp) and Wine.com (http://www.wine.com/)

FacetMap has a simple facet demonstration on their site . One can browse a wine classification by varietal (red, white, bubbly, etc.), region, and price. Clicking on "bubbly" narrows down the options, limiting the remaining choices in all three facets. Choosing "sparkling" from the varietal facet narrows it still more, and choosing bottles priced under $20 lists six bottles that could still be divided by region. The user can find an appropriate bottle of wine in many different ways depending on how he or she thinks about wine and what his or her needs are (a cheap bottle, something white to go with fish, something local to support nearby vintners).

Wine.com's system is similar. From the home page, the user may choose to browse by type of wine (for example, white), then by varietal (Semillon) and see a listing that can be sorted by price, rating, winery, or type (which is redundant, because the type has already been defined—this option should not be visible in this case). The advanced search option off the home page lets the user form a query by defining price range, type of wine, and region.

There are other ways that wine could be classified, of course, but these are very accessible examples and good ones for showing the nature of facets. Those unfamiliar with facets (but familiar with wine!) could think about how they would classify wine themselves, based on how they choose and drink wine, and make their own faceted scheme in an easy and fun thought experiment.

lawforwa.org (http://lawforwa.org/)

As discussed in Louie, Maddox, and Washington (2003) above (see Recommended). "Your gateway to Washington state law and government." Choose the Advanced Search option to see the faceted classification they used and how it guides the search tool. It is a very interesting interface and when the site has been finished and tested with users, their experiences will be worth reading. It has one problem that is addressed in part by the work of Yannis Tzitzikas et al (see above, in Background): it is possible for the user to construct a query that is already known by the system to have no answers or to be impossible. Their advanced search screen would be better (though this is not simple to program) if, as the user selected a subject in a facet, the options displayed under other facets changed to reflect the known (or possible) combinations. The user would be saved the frustration of carefully choosing from three listings, one of them very long, only to find there is nothing available. A faceted navigation system like this could be made to tailor itself to the user while the user is actually constructing a search—surely a fine example of how to save the time of the reader (one of Ranganathan's Five Laws of Library Science).

Mailing Lists

</p>

The Faceted Classification Discussion (FCD) Mailing List. http://poorbuthappy.com/fcd/.

178 people are on the list, as of 14 October 2003, including some writers mentioned above. The list was founded in December 2002, and the archives are open to all. Worth joining for anyone working with, or just interested in, faceted classifications.

xfml (a mailing list about XFML). http://groups.yahoo.com/group/xfml/.

A slightly older list, created in May 2002, but with fewer members (79 as of 14 October 2003). It concentrates on XFML, and so is more focused and technical. It is the only place to go to ask questions about XFML.

Alphabetical List of Resources

This is an alphabetical list of everything above, without annotations.