1!"#$ ?

!"#$%&'( )"*+&,-*,."-!$%"&"%" "'& ($)*'&

$""$ Transactions on Architectural Education no. +,

Matteo Zambelli, Anna Helena Janowiak and Herman Neuckermans

!"#$ Consortium

Collaboratorio$""$ European Association for Architectural Education

Fraunhofer -.( Verlag

European Commission, eContentplus Program

edited by

produced by

powered by

published by

supported by

MACEBOOK_01.indb 1 25-08-2008 13:07:50

2 (.*/0-'1 ".#2-%$#%3.$. !$%"&"%" "'& ($)*'&

Bibliographic information published by the Deutsche NationalbibliothekThe Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;detailed bibliographic data are available on the Internet at http://dnb.d-nb.de.-0(' 456-7-6895-555,-,printed by Fraunhofer -.( Media Services

© by Fraunhofer -.( Verlag, :,,6. All rights reserved Fraunhofer-Informationszentrum Raum und Bau -.(Postbox 6, ,+ 94, 5,;,+ Stuttgart, GermanyNobelstrasse 8:, 5,;94 Stuttgart, Germanyphone ++4 (,) 588 45, :;,,fax ++4 (,) 588 45, :;,5e-mail: irb@irb.fraunhofer.dewww.irb.fraunhofer.de

graphic designstudiovisuale.it

cover illustrationMoritz Stefaner

printingFraunhofer IRB Mediendienstleistungen, Stuttgart

typographyCsuni (Carattere Senza Un Nome Importante/Typeface With No Important Name)by Molotro (Luciano Perondi, Federico Zerbinati)

papercover Schneidersöhne Luxo Satin :;, g/m:

pages 8-7:, printed with 8 colouron Schneidersöhne Luxo Satin 87; g/m:

pages 7:8-796 printed in +-colourson Schneidersöhne Luxo Satin 87; g/m:

MACEBOOK_01.indb 2 25-08-2008 13:07:50

introduction

!"#$%&#'Herman Neuckermans

(#"%)*+, &#-.*/$-/0#$. 1$/&'&/& &+' ($2"+': The conceptFurio Barzon, Anna Janowiak, Matteo Zambelli

mace

1&-$: Connecting and Enriching Repositories for Architectural LearningMoritz Stefaner, Vittorio Spigai, Elisa Dalla Vecchia, Massimiliano Condotta,Stefaan Ternier, Martin Wolpers, Stefan Apelt, Marcus Specht, Till Nagel, Erik Duval

1"'$3) "! '$)*,+ &-/*4*/*$): Towards E5ective Design Sca5oldingMario De Grassi, Alberto Giretti, Roberta Ansuini

1&-$: Enabling Legacy RepositoriesStefan Apelt, Christian Prause, Mathias Casaer, Ann Heylighen

$4$#24*33$ *+)/&33&/*"+: 1&-$ at La Biennale of ArchitectureInteraction Design Lab and Interface Design, University of Applied Sciences Potsdam (!.6)

e-learning

/$7/*'&//*-&: Saggio Teaches Information Technology in LatinAntonino Saggio

*+'$7$' &+' (#"%)$': A New Didactic Approach Towards the Orders of ColumnsSusanne Schumacher

."0)*+,@89.$0: Integrating Learning Spaces and Architectural RepositoriesLeandro Madrazo, Paul Riddy, Luca Botturi

($2"+' '*,*/&3 #$6")*/"#*$): Architectural Information PresentationVerdy Kwee

/

0

1

2

3

4

5

6

7

table of contents

5

4

87

:8

::

;,

9:

5:

68

6:

4+

8,+

886

MACEBOOK_01.indb 3 25-08-2008 13:07:51

/"%&#') 0)*+, '*,*/&3 1"'$3*+, )2)/$1): The Context of E-learningNada El-Khoury, Giovanni De Paoli

@,': Keeping a Record of Learning Paths on Digital Graphics Representation for ArchitectureAdriane Borda, Neusa Felix, Luisa Dalla Vecchia, Janice Pires

4*#/0&3 )/0'*": A Digital Repository in Architectural EducationAndy Earl, Carl O’Coill, Joss Winn

,"*+, '*,*/&3 *+ /.$ -3&))#""1: Lessons from the Integration of Wikis in an Architectural History CourseFrancesca Torello, Marie Norman

archives (0*3'*+, 06 '*,*/&3 -"33$-/*"+): From Policy to ImplementationPatricia Alkhoven

/.$ "+-3*+$ -&/&3",0$ "! /.$ ,*&+-&#3" '$ -&#3" &#-.*4$): An Attempt to Integrate Paper-Based and Digital-Born DocumentsRiccardo Domenichini

$3$-/#"+*- '"-01$+/) *+ &+ &#-.*/$-/0#&3 6#&-/*-$: Graphics Creation, Maintenance, Selection and PreservationElena Triunveri

*1&,$ &#-.*4$: A Hybrid Structure for the Enhancement of Architecture VideosPaola Ricco

/.$ !&-$/&, $+,*+$:A Semantic Collaborative Tagging ToolEmanuele Quintarelli, Andrea Resmini, Luca Rosati

-"160/$# *+/$,#&/$' (0*3'*+, )2)/$1: Preliminary ThoughtsHong Zhang

)&4$ /.$ (0((3$: The Architectural Archive in the Digital AgeBernhard Franken, Berthold Scharrer, Suemri Nina Michaela Vogel

7

/8

//

/0

/1

/2

/3

/4

/5

/6

/7

8:9

879

8++

8;9

895

896

856

869

849

:,+

:86

::+

MACEBOOK_01.indb 4 25-08-2008 13:07:51

08

0/

00

01

02

03

04

& -3&))*!*-&/*"+ )2)/$1 !"# -"+)/#0-/*"+ 0+*/) &+' 6#"'0-/): Methodological Hypothesis and Application Testing on Pre-Modern BuildingSara Scapicchio

/.$ 6#"(&'" !#&1$%"#:: A Repository for Architectural ;'-modelsIna Blümel, Harald Krottmaier, Raoul Wessel

websites

1*1"&: An Interactive Architecture GuideMieke Vullings, Naomi Schiphorst

+$7/#""1: The European Hub for Contemporary ArchitectureJuerg Meister, Helga Kusolitsch, Stephan F. Haupt

/.$ 3&!*) 6#"<$-/: Managing Publicly Induced Data into a Scholar EnvironmentFrancisco Agostinho

& '&/&(&)$ "! &#-.*/$-/0#&3 #$6")*/"#*$): Criteria for Selection and EvaluationStefan Boeykens, Herman Neuckermans

-03/0#&3 .$#*/&,$ #$6")*/"#*$): Digital Archives for Conservation and ManagementAlonzo Addison, Mario Santana Quintero

images

collection of essays’ images

index of terms – folksonomy

colophon

table of contents

:76

:;,

:98

:9:

:6,

:4,

:46

7,9

784

7:,

79+

795

MACEBOOK_01.indb 5 25-08-2008 13:07:51

204

!"

› classification› facetagging› faceted classification› folksonomies

› information architecture› social classification› tag› tagging

FaceTag is a working prototype of a semantic collaborative tagging tool conceived for improved bookmarking, searching and re-finding of resources in specialized domains.FaceTag’s goal is to show how the vast, homogeneous, flat keywords’ space created by users while tagging can be e!ectively improved by using a richer faceted classification superstructure to supplement the “information scent” and “berry-picking” capabilities of the system. The additional layer is aggregated both implicitly, observing user behaviour, and explicitly, by introducing a compelling user experience that facilitates end-user creation of relationships between tags.

#$% &'(%#') %*)+*%:A Semantic CollaborativeTagging Tool

MACEBOOK_01.indb 204 25-08-2008 13:08:19

205

%,'*-%.% /-+*#'0%..+FaceTagRoma, Italywww.facetag.org

Emanuele Quintarelli is an Enterprise ".# specialist and evangelist. He has spoken at several international conferences and wrote the Enterprise ".# chapter of the book Web ".# by Il Sole "$ Ore. He is the organizer of Web".Oltre (the Italian conference about Web ".#) of the Italian Information Architecture Summit and of the International Forum on Enterprise ".#.His blog is http://socialenterprise.it.

'*10%' 0%2,+*+FaceTagBologna, Italywww.facetag.org

Andrea Resmini is an information architect. An ICT professional since %&'&, Andrea holds an MA in Architecture and Industrial Design and is currently pursuing a PhD in Legal Informatics specializing in IA and UX issues concerning historical and juridical databases. Andrea chaired the "nd Italian IA Summit and leads the Higher Education in IA WG for the Information Architecture Institute. His website is http://resmini.net.

.-(' 032'#+FaceTagPerugia, Italywww.facetag.org

Luca Rosati is a freelance information architect and professor of Human Computer Interaction at the University for Foreigners of Perugia (IT). He has spoken at several international conferences and published the books Organizing Knowledge: From Libraries to Information Architecture for the Web; and Information Architecture: From Everyday Environments to the Web. His website is http://lucarosati.it.

MACEBOOK_01.indb 205 25-08-2008 13:08:20

206

MACEBOOK_01.indb 206 25-08-2008 13:08:20

207!"# $%&#!%' #(')(# !"

+*#031-(#+3*Collaborative tagging systems have been largely adopted by end-users as useful and powerful tools to organize, browse and publicly share personal collections of resources on the World Wide Web through the introduction of simple metadata.

The aggregation of these metadata is often referred to as a folksonomy, a user-generated classification, emerging through bottom-up consensuswhile users assign free form keywords to online resources for personal or social benefit.

Del.icio.us (http://del.icio.us), Flickr (www.flickr.com), *+things (www.*+things.com), Furl (www.furl.net) and Technorati (www.technorati.com) are all web-based collaborative systems for building shared databases of items, enriched by a flat metadata vocabulary that can be used to perform metadata-driven queries, to monitor change in areas of interest or to discover emergences or trends, such as the hottest / most popular topics in the system [Quintarelli ,--.].Folksonomies have often been seen as orthogonal to taxonomies and controlled vocabularies: flat, inclusive and emerging information spaces created by bottom-up user input and consensus versus rigid, hierarchical and coherent systems profession-ally hand-crafted a priori [Quintarelli ,--.]. In a flat tagging system eachdocument can be retrieved through a simple set of keywords, collaboratively introduced by users to describe and categorize that document, very much like in a keyword-based search process in which descriptive terms can be used to get a set of applicable items.Despite their low cognitive cost, their capability of matching the users’ real needs and language and their great value in a serendipity research task, folksonomies imply however a lack of precision, a very low findability quotient (especially in a known-item approach) and a limited scalability for the intrinsic variability of lan-guage [Quintarelli ,--.].

Tags also help users identify articles and posts of interest on websites, provid-ing a complimentary freelinking super-structure. This marks a major shift in that tagging becomes a tool to maximize findability and browsability, requir-ing new, faster and more powerful ways to access content, far beyond what was allowed by earlier tag cloud approaches [Feinstein & Smadja ,--/].

Tag clouds, where the most popular tags are usually displayed through an alphabeti-cally ordered list with the font size increasing with the tag’s relevance, are widely used visual interfaces for information retrieval that provide a global contextual view of tags assigned to resources in a system [Hassan, Montero, Herrero, Solana ,--/].As they are currently implemented, tag clouds cannot clearly provide a rich multidi-mensional browsing experience over large tagging spaces. There are several reasons for this:0. Choosing tags by frequency of use inevitably causes a high semantic density with very few well-known and stable topics dominating the scene.,. This in turn might lead to a self-sustaining cycle, where well know and stable top-ics become ever predominant because they are predominant.+. Providing only an alphabetical criterion to sort tags heavily limits navigation, scan-ning and extraction, and in the end prevents the user from building a coherent global map of tag-space.

folksonomykeywordstag cloudstagging

browsing

4 56 75 77 74

metadata

5 4 8 59

5: 56 75

MACEBOOK_01.indb 207 25-08-2008 13:08:20

208 12345)(' %2&")!#&!62#. 7#!%8%!% %(8 1#93(8

*. A flat tag cloud cannot visually support semantic relationships between tags. We suggest that these relationships are needed to improve the user experience and general usefulness of the system... Current tag clouds are simple lists of links and fail to provide any complex seman-tical operation, besides allowing to click on any given tag one at a time.

As a result of the inherently inconsistent, evolving and much variable process of associating words and meanings, tagging systems are implicitly plagued by a number of linguistic issues which include polysemy, homonymy, plurals, synonymy and basic level variation, which do not appear easy to solve [Golder & Huberman ,--.]. Any of these problems can dramatically reduce the e:ectiveness of the application and the benefits brought by the use of tagging systems.

#'))+*) '*1 &3.;23*3,+%2 +* +' .+#%0'#-0%The widespread adoption of tagging systems by end-users has greatly stimulated discussion about their long-term implications inside the information studies com-munity. In Ambient Findability, Morville [Morville ,--.] states that tagging has its own proper place inside information architecture theory and practice, suggesting that these systems can be productively considered a complementary fast moving layer over slower layers represented by more traditional architectural information practices. Karl Fast reinforced this position showing how pace layering theory can be combined with the complexity and resilience theories to provide a working model of interaction between folksonomies and conventional information architecture [Camp-bell & Fast ,--/]. Golder and Huberman [Golder & Huberman ,--.] analysed the structure and dynamics of collaborative tagging systems trying to extract stable pat-terns and recurrent tag types in del.icio.us. With Mefeedia [http://mefeedia.com], Dutch information architect Peter Van Dijck was among the first to propose a mixing of facets and tags, even if in Mefeedia tags are statically assigned to a fixed number of editorially designed facets. Discussions of faceted classification models in current websites, and the relationship between facets and tagging, have been presented by Travis Wilson at the IA Summit ,--/ [Wilson ,--/].

Marti Hearst and The Flamenco project have been investigating for more than fourteen years how faceted interfaces can help users flexibly navigate and search through large information spaces [Hearst, The Flamenco Search Interface Project].

Hierarchical relationships can be implicitly found in tagging systems as showed by Sam H. Kome [Kome ,--/], while Heymann and Garcia-Molina presented a simple algorithm to automatically convert tags associated to objects into a hierarchical taxonomy [Heymann and Garcia Molina ,--/].The pioneer social bookmarking site RawSugar, now closed, implemented a similar hierarchical approach complemented by clustering techniques.

Context, language, communities of practice and specialized domains all have a large impact in the way users tag, as reported by Resmini [Resmini et al. ,--;].

folksonomytaxonomy

facet

5 4 79

tagging

4 8 59

MACEBOOK_01.indb 208 25-08-2008 13:08:20

209!"# $%&#!%' #(')(# !"

Finally, Bar-Ilan [Bar-Ilan et al. ,--/] compares unstructured (freely assigned tags) and structured tagging (tags assigned to predefined metadata elements), suggesting that structured tagging may be able to produce stronger user guidance, hence pos-sibly resulting in higher quality descriptions.

2%,'*#+( 2#0-(#-0%2 +* #'))+*) 2<2#%,2Usability studies show how information seekers in large domains of objects prefer meaningful groupings of related items, in order to quickly understand relationships and so decide how to proceed [Hearst ,--/a]. In other words, it seems quite clear that without any means to explore and make sense of large quantities of similar items, users feel lost and fail. Clustering and faceted classification have both been proposed in the past as useful techniques to address this issue.

Clusters

Document clustering refers to the act of grouping items according to some measure of similarity, such as for example, identifiable repetitive patterns of words and phrases.

Some advanced tagging systems like Rawsugar and Flickr are already using clusters to address the issues that plagued the first generation of folksonomy-based applica-tions: clusters help reduce the semantic density and improve the visual consistency of tag clouds. Moreover, clustering is automatable, can be used to refine vague queries and to disambiguate search keywords. Nonetheless, clustering techniques and clustering algorithms are not perfect and often generate unclean, hard-to-predict groups. These groups tend to conflate many di:erent dimensions as well, and easily become hard to label in meaningful ways. Finally, clustering does not generally allow refinements and follow-up queries, thus resulting in limits to the exploratory capabilities of the system.Usability results show that users prefer clear hierarchies with categories at uniform levels of granularity, to the unpredictable and unlabeled groupings typical of cluster-ing techniques [Hearst ,--/a].

Hierarchical facets

At the other end of the classificatiweb, traditional hierarchical categories are coher-ent and complete systems of meaningful labels, which systematically organize a domain. The main drawback to this approach is that a single monolithic hierarchi-cal organization rarely has the capability to match the varied ways of thinking and organizing that the world of di:erent users in a live social environment has.

In this respect, hierarchical faceted metadata has emerged as a promising middle ground, which can satisfy the needs of a wide range of users withdi:erent mental models and vocabularies [Yee et al. ,--+].

Facets are orthogonal descriptors (i.e. categories) within a metadata system. Each facet has a name and addresses and a di:erent conceptual dimension or feature type relevant to the collection. As facets can be flat or hierarchical and they can be

clusteringtag clouds

facet

5 4 79

MACEBOOK_01.indb 209 25-08-2008 13:08:20

210 12345)(' %2&")!#&!62#. 7#!%8%!% %(8 1#93(8

assigned single or multiple values, a faceted search interface requires each object in the collection to be classified using labels from di:erent facets. In a hierarchical faceted navigation tool, choosing a label from one of the facets is equivalent to performing a disjunction over all the labels beneath the selected one, while choosing labels from di:erent hierarchies results in a conjunction of disjunctions over the selected labels and their sub labels.

In this kind of interface, users can navigate multiple faceted hierarchies at the same time [English et al. ,--,b]. Usability studies show how this approach is preferred over single hierarchies because users feel in control without getting lost [English et al. ,--,b; Yee et al. ,--+]. Additional features exposed by faceted-based interfaces are the possibility to suggest logical alternatives at each step and to avoid dead ends, which produce no result whatsoever.

For these reasons, faceted metadata can be used to support navigation along several dimensions simultaneously, allowing seamless integration between browsing and free searching, and an easy way to refine (zoom in) and broaden (zoom out) any query while retaining a feeling of control and understanding [English et al. ,--,b]. In short, this approach favours recognition over recall and provides better support for exploration, discovery and iterative query refinement [Hearst ,--/a]. Again, usability studies attest how hierarchical faceted interfaces are preferred over simpler keyword based search interfaces and how they can be easily understood by the average user [Yee et al. ,--+] if iteratively designed and tested to address usability issues [English et al. ,--,a].

3=%0=+%> 3& &'(%#')FaceTag introduces an innovative, multidimensional semantic paradigm for organizing, navigating and searching large information spaces through the use of tags. We believe this approach can limit the impact of linguistic com-plications, as FaceTag introduces a number of correctives to explicitly address these issues:

0. It allows tag hierarchies. Users have the opportunity to organize their resources by means of parent-child relationships.,. By means of the underlying, domain-related faceted classification structure, tag hierarchies are semantically assigned to established facets. These are used to section and navigate the resource domain flexibly.+. Tagging and searching can be mixed to maximize findability, browsability and us-er-discovery. Similarly, one of the main limitations of hierarchical faceted categories lies in the di<culty to automate both the creation of categories and the association of single items to a label hierarchy under each facet [Hearst ,--/a]. FaceTag avoids this particular issue entirely by reversing the paradigm, as it uses no algorithmic round-ups: FaceTag is built around the notion that users provide the legwork, and one of the side-goals of the project is to investigate how a hierarchical and faceted metadata structure can be added to user generated content making use of tags provided by end users in collaborative systems, limiting the amount of e:ort and toil required through a careful user interface design [Mascaro ,--=].

MACEBOOK_01.indb 210 25-08-2008 13:08:20

211!"# $%&#!%' #(')(# !"

Although facet and faceted have become very common terms in the information architecture field, their application falls often far from its original meaning. The at-tribute faceted, indeed, is often loosely referred to the availability of means to search by di:erent keys [La Barre ,--*].

The full theory of faceted classification, as developed by S. R. Ranganathan and the Classification Research Group (&2') and which includes rules for citation order and notation, has definitely enjoyed a less widespread use as far as website organization is concerned. Notable exceptions to this rule are usually o:ered by projects sta<ng librarians, such as $%!>5 [Slavic ,--,], and FaceTag.

Reference projects were Flamenco (http://flamenco.berkeley.edu), Facetious (http://demo.siderean.com/facetious/facetious.jsp), and Etsy (www.etsy.com). It must be noted that both Facetious and Etsy propose a mixture of properly designed facets and simpler, less-compliant metadata.

(0%'#+*) &'(%#2 &30 #'))+*)The choice of facets is based on the CRG theory [Vickery 0?/-]. Indeed, an aspect of-ten underestimated on the World Wide Web is that both Ranganathan and the CRG described a generic schema for faceted classification, which every actual schema can refer to. Thus, when working on a faceted classification project one does not have to rebuild a schema from scratch every time, but may follow guidelines. CRG postulates 00-0+ general categories [Broughton ,--0]: since for the test case implementation, the chosen reference domain was information architecture (IA), in Tab. 0 is shown the matching between CRG standard categories and information architecture-relat-ed categories that were used to define the facets. Tab. , purports the list of facets and a number of sample foci derived from the initial data pool, a preliminary corpus of IA resources gathered from the Information Architecture Institute Library [http://iainstitute.org/library].

CRG FaceTag

Thing [Documents, resources] Type Resource Types (e.g. online report, case study...) Part - Property Language Material [Format] Process - Operation Activities/Subjects (e.g. competitive analysis, faceted classification...) Product [Deliverables] Byproduct - Patient Usage (e.g. Industry, Health...) Agent People Space [Country]Time Date

Tab. !: FaceTag facets definition by CRG standard categories.

clustering

facet

5 4 79

MACEBOOK_01.indb 211 25-08-2008 13:08:20

212 12345)(' %2&")!#&!62#. 7#!%8%!% %(8 1#93(8

Facet Examples

Resource Types white paper, case study, blog>enterprise web Language Predefined values (based on ISO Standard ISO "#$-%) Activities / Subjects Competitive analysis, classification > facets, web ,.- > folksono mies, information Design > navigation design > breadcrumbsUsage industry, public administration, health, software > companies > Google, education > conferences > www,--/ People Morville, E. Reiss, WeinbergerDate Automatically added

Tab. %: FaceTag facets and examples of foci.

It must be noted that these facets are being constantly revised upon users’ feedback, as they are guidelines and not a restrictive framework. User needs are paramount and should be employed to determine the most valuable facets for every system and community of practice. Moreover, in the actual implementation foci, the values each facet can assume in faceted classification theory, will be user-generated – as in this model tags are the foci. The only exceptions are the language facet, which uses a predefined list of languages in the ISO /+?-, notation, and the date facet, which will receive a software-generated time-stamp upon resource creation. Tab. + lists the definitive list of facets and labels used by the FaceTag engine.

Facet Examples

Resource Types white paper, case study, blog > enterprise webLanguage Predefined values (based on ISO Standard ISO /+?-,) Themes competitive analysis, classification > facets, web ,.- > folksono mies, information Design > navigation design > breadcrumbsPurpose industry, public administration, health, software > companies > Google, education > conferencs > www,--/ People Morville, E. Reiss, WeinbergerDate Automatically added

Tab. #: Final list of facets.

?%00<-@+(;+*), +*&30,'#+3* 2(%*# '*1 #$% #>3 'A+2 3& +*&30,'#+3* '0($+#%(#-0%Facets are particularly suitable to classify a homogeneous collection of items – i.e. a set of resources belonging to a specific disciplinary area, and because of their very nature, facets constitute an adaptive classification system capable to easily repre-sent both knowledge in movement, such as that observable in a social, collaborative context; and several mental models at the same time, such as those playing their role in this context. As such, the blend of tags and facets helps empower and amplify the “information scent” [Chi et al. ,--0] and the “berry-picking” capabilities of the system.

The term “berry-picking” refers to a model in which searching is described not as a single query-response process, but as an evolving temporally laid out set of sometimes conflicting search terms along a berry-picking-like path [Bates 0?;?].

berry-picking

5B

MACEBOOK_01.indb 212 25-08-2008 13:08:21

213!"# $%&#!%' #(')(# !"

Every information architecture deals with two di:erent information axes [Rosati ,--=: =/-?0]: a vertical (or paradigmatic) axis, consisting of the hierarchical relation-ship that each item in a system engages with the others; and a horizontal (or syn-tagmatic) axis, consisting of the semantic relationships of contiguity that each item engages with the others. We believe that the combination of tags and facets allows for better use and management of these:@ paradigmatically, when users associate a keyword to a facet (to tag a resource), FaceTag suggests similar tags or a hierarchy of tags pertaining to the same facet;@ syntagmatically, FaceTag allows users to see all the other tags belonging to the same facet(s).

Fig. !: The two axes of Information Architecture (p. #&&).

-2+*) &'(%#')FaceTag works di:erently from the average collaborative bookmarking ap-plication. It o:ers both a browsing/searching mode and an administrative/editing mode as separate instances, as these are two di:erent activities to which the user interface adapts providing di:erent aiding tools (navigation, resource management) and di:erent behaviours (zooming, tag suggestions).

At the heart of FaceTag is a zooming engine, which maintains a history of all the tags and searches that a user is requesting (this is called “engaging a tag”). When users engage a tag, the interface adjusts and zooms in providing a filtered view, a subset of all available resources based on this active selection. By selecting more tags from dif-ferent meaningful facets (as they are editorially chosen for the current domain), users can e:ectively shrink the results they obtain until they find either what they are look-ing for, or the last possible result set arising from the intersection of all active tags. This feature set enables considerably enhanced, berry-picking-like search strategies: since tags can be disengaged freely and in whatever order suits the users, opening up new unpredicted result sets, users might decide to pursue another search and en-ter new di:erent subsets following some other information scent (that is, a theme or term that appears promising), or start anew by clearing up their history completely, never finding themselves facing dead-ends, which produce no results at all.

2',@.% 2%22+3*When a user accesses the application, first FaceTag replies in browsing mode and the user is presented a page, which lists the most recent additions to the system in the main body. The other relevant parts of the user interface are the facet containers and the search box at the top.

Fig. %: FaceTag main page, with the most recent resources and the facet containers (p. #&').

Inside FaceTag, a user can decide to look for content a) by entering tags manually or b) by clicking on first-level tags in a specific facet container.

If the user enters a tag manually, FaceTag returns the paginated results set of all the resources, which either contain that tag in their tag pool or have the same term

MACEBOOK_01.indb 213 25-08-2008 13:08:21

214 12345)(' %2&")!#&!62#. 7#!%8%!% %(8 1#93(8

in their title, description or notes. The facet display is adjusted accordingly to show only those facets and pertaining first-level tags that are related to the results set. Any active tag is called an “engaged” tag.

Fig. #: Searching for resources. The user chose “information architecture” and facet containers and tag lists adjust. Stickers for the engaged tag appear in the status bar top, grey (p. #&').

In case the keyword happens to be an umpteenth-level tag, the corresponding facet will show all umpteenth+0 tags and add any broader tag in the hierarchy up to the umpteenth-0 tag to the facet title as clickable items, which allow zooming out. If there is no umpteenth+0 tag, the facet is not displayed. If the user clicks on a tag from any of the facet containers, FaceTag returns the paginated results set of all the resources that have been tagged with that tag. The currently engaged tag (or tags) is displayed in a status bar to show it’s active. The facet containers are adjusted con-sequently. The active facet container shows all broader tags from the hierarchy, the selected tag may be part of and alongside the facet title and all pertaining narrower tags. Inactive facets show first-level tags, which relate to the resources pertaining to the results set. Upon subsequent zooming in and refining the query, when there are no narrower tags, the breadcrumb display is maintained to allow zooming out or what we call “disengaging”, resetting the search, while the facet container is greyed out and becomes inactive.

Fig. &: Empty facet containers and greyed out pertaining terms at the end of a search session (p. #&').

Obviously, a user may start searching for a keyword and then adjust her results set using facets, combining the two approaches in any way she prefers until she reaches a satisfactory answer, or proceed vice versa and zoom in and out by using tags. Simi-larly, tags pertaining to di:erent facets can be used together during a single search to narrow down a results set quickly and e<ciently. If there is no disengagement, all subsequent operations are performed on the intermediate results set.

If a user logs in, access to the administrative interface is granted and adding, editing and deleting resources and tags become possible. Upon entering new resources, a user is provided with a simple form with entry fields for every facet.

These tag fields are optional, and can be left empty at will: there is no mandatory facet. However, if a user start to enter a tag, the completion tool suggests similar tags from the pertaining facet only. Moreover, since users can optionally identify two or more tags as a hierarchy through a simple syntax (using the “>” character), the completion tool can suggest, again facet per facet, not just similar tags, but similar tags as parts of a hierarchy of tags, hence e:ectively suggesting an entire hierar-chy. Gradually, with use, these hierarchies acquire complexity and become globally significant in the system.

Fig. ': The administrative interface for adding resources to FaceTag (p. #&&).

Editing or modifying can be done seamlessly from the browsing interface, by clicking icons, which appear next to one’s own resources. Noticeably, the same happens if a user tries to add a resource she already added (based on URI identification): FaceTag simply supplies the editing interface and preloads the original data.

MACEBOOK_01.indb 214 25-08-2008 13:08:21

215!"# $%&#!%' #(')(# !"

(3*(.-2+3*2By providing the user with facets to which hierarchical sets of tags relate and pertain and a usable interface, which adapts to the ongoing query, FaceTag may solve, through contextualization and user-added semantic value, most of the basic issues connected with polysemy, homonymy and base level variations.

While further testing and usability studies are needed to verify to what extent users are motivated to use our prototype and to introduce structure in addition to flat tags, preliminary user evaluations show how the addition of hierarchies and facets can improve and disambiguate the meaning of tags by means of contextualization and organization.

This approach also augments the overall architectural scalability of the system and directly addresses one of the largest issues faced by the most appreciated social applications. Iterative card sorting tests with di:erent user groups is being consid-ered for all labelling, with the purpose to map some of the mental models by which users represent the IA knowledge domain. These results will provide further data to tune the facet architecture. Similarly, the user interface is being designed through documented heuristics and patterns and verified at each iterative step by small us-ability tests. More extensive user research will involve the use of think-aloud protocol sessions with more than five testers for each session. Scenarios will include storing bookmarks and retrieving them. Looking at preliminary results, a critical task ad-dressed by the application is the assignment of new bookmarks and the association of tags to relevant facets. The current interface is rather simple but other options, leveraging advanced tag suggestion and tag/facet association, are beginning to be evaluated.

FaceTag began from the assumption that bottom-up and top-down classifi-cation methods are complementary: tagging is inclusive, simple and matches the user’s real needs and use of language, whilst formal methods provide coherent, consistent systems for organizing data, but are expensive and do not have the advantage of corresponding to the user’s use of language.

The system that was built as a result combines both approaches, to mutual benefit. As testing and evaluation is now underway it is hoped that results will become avail-able later this year.

0%&%0%*(%2 !

MACEBOOK_01.indb 215 25-08-2008 13:08:21

216 12345)(' %2&")!#&!62#. 7#!%8%!% %(8 1#93(8

0%&%0%*(%2Bar-Ilan J., S. Shoham, A. Idan, Y. Miller, and A. Shachak. ABBC. Structured vs. Unstructured Tagging – A Case Study. WWWABBC, Edinburgh. http://www.rawsugar.com/wwwABBC/DA.pdf.

Bates, M. DEFE. The Design of Browsing and Berrypicking Techniques for the Online Search Inter-face. Online Review DG: HBI-HAH. http://www.gseis.ucla.edu/faculty/bates/berrypicking.html.

Broughton, V. ABBD Klasifikacija za AD. stoljece: nacela i struktura Blissove bibliografske klasifikacije [= A classification for the ADst century: principles and structure of the Bliss biblio-graphic classification]. Vjesnik bibliotekara Hrvatske HH, no. D-H, GF-JD [Italian translation: Una classificazione per il AD’ secolo: principî e struttura della Classificazione bibliografica Bliss, AIB-WEB. Contributi. http://www.aib.it/aib/contr/broughtonD.htm].

Campbell, G.D., and K. V. Fast. ABBC. From Pace Layering to Resilience Theory: The Complex Implications of Tagging from Information Architecture. Proceedings of IA Summit ABBC, ASIS&T (Vancouver, March AG-AI, ABBC). http://www.iasummit.org/ABBC/files/DCH_Presentation_Desc.pdf.

Chi, E.H., P. Pirolli, K. Chen, and J. Pitkow. ABBD. Using Information Scent to Model User Informa-tion Needs and Actions on the Web. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (Seattle, Washington, ABBD). New York: ACM Press. http://wwwA.parc.com/istl/projects/uir/publications/items/UIR-ABBD-BI-Chi-CHIABBD-InfoScentModel.pdf.

English, J., M. Hearst, R. Sinha, K. Swearingen, and P. Yee. ABBAa. Hierarchical Faceted Metadata in Site Search Interfaces. CHI ABBA Conference Companion. http://flamenco.berkeley.edu/papers/chiBA_short_paper.pdf.

English, J., M. Hearst, R. Sinha, K. Swearingen, and P. Yee. ABBAb. Flexible Search and Browsing Using Faceted Metadata (Unpublished). http://flamenco.berkeley.edu/papers/flamencoBA.pdf.

Feinstein, D., and F. Smadja. ABBC. Hierarchical Tags and Faceted Search. The RawSugar Approach. In Proceedings of SIGIR (Seattle, Washington, C-DD August, ABBC).

Flamenco Group. ABBA. How to Build a Flamenco Instance.http://bailando.sims.berkeley.edu/flamenco/howtobuild/howtobuild.html.

Gnoli, C., V. Marino, and L. Rosati. ABBC. Organizzare la conoscenza. Dalle biblioteche all’architettura dell’informazione per il Web [= Organizing Knowledge. From Libraries to Information Architecture for the Web]. Milan: Tecniche Nuove.

Golder, A.S., and B.A. Huberman. ABBJ. The Structure of Collaborative Tagging Systems, Information Dynamics Lab. http://arxiv.org/pdf/cs.DL/BJBFBFA.

Hassan-Montero, Y., and V. Herrero-Solana. ABBC. Improving Tag-Clouds as Visual Information Retrieval Interfaces. International Conference on Multidisciplinary Information Sciences and Technologies, InSciTABBC. http://www.nosolousabilidad.com/hassan/improving_tagclouds.pdf.

Hearst, M.A. ABBCa. Clustering Versus Faceted Categories for Information Exploration. Communication of the ACM HE, no.H (April). http://flamenco.berkeley.edu/papers/cacmBC.pdf.

MACEBOOK_01.indb 216 25-08-2008 13:08:21

217!"# $%&#!%' #(')(# !"

Hearst, M.A. ABBCb. Design Recommendations for Hierarchical Faceted Search Interfaces. ACM SIGIR Workshop on Faceted Search. http://flamenco.berkeley.edu/papers/faceted-workshopBC.pdf.

Hearst, M.A. The Flamenco Search Interface Project. http://flamenco.berkeley.edu/pubs.html.

Heymann, P., and H. Garcia-Molina. ABBC. Collaborative Creation of Communal Hierarchical Taxonomies in Social Tagging Systems. Technical Report InfoLab. http://dbpubs.stanford.edu/pub/ABBC-DB.

Kome, S. H. ABBC. Hierarchical Subject Relationships in Folksonomies. School of Information and Library Science, University of North Carolina at Chapel Hill. http://hdl.handle.net/DEBD/AGF.

La Barre, K. ABBH. Adventures in Faceted Classification: A brave New World or a World of Confusion? In Knowledge organization and the global information society. Proceedings of the (th international conference of the International Society for Knowledge Organization, ed. C. McIlwaine (University College London, DG-DC July, ABBH). Advances in Knowledge Organization E: IE-FH

La Barre, K. ABBC. The Use of Faceted Analytico-Synthetic Theory as Revealed in the Practice of Website Construction and Design. http://leep.lis.uiuc.edu/publish/klabarre/facetstudy.html.

Mascaro, L. ABBI. User Interface per applicazioni Web A.B – FaceTag [= User Interface for Web A.B Applications – FaceTag]. In Proceedings of the Second Italian Information Architecture Summit )**+ (Trento, Italy, DC-DI November, ABBI). http://www.iasummit.it/download/BJ-siiasABBI.pdf.

Morville, P. ABBJ. Ambient Findability. Sebastopol, California: O’Reilly.

Quintarelli, E.. ABBJ. Folksonomies: Power to the People. In Proceedings of ,th ISKO Italy-UniMIB meeting (Milan, Italy, AH June, ABBJ). http://www.iskoi.org/doc/folksonomies.htm.

Resmini, A., E. Tonkin, E. M. Corrado, H. L. Moulaison, M. E. I. Kipp, H. D. Pfei:er, and Qiping Zhang. ABBF. Collaborative and Social Tagging Networks. Ariadne JH. http://www.ariadne.ac.uk/issueJH/tonkin-et-al.

Rosati, L. ABBI. Architettura dell’informazione: Trovabilità dagli oggetti quotidiani al web [= Information Architecture: From Everyday Things to the Web]. Milan: Apogeo.

Slavic, A. ABBA. FATKS: Facet Analytical Theory in Managing Knowledge Structures for Humanities. http://www.ucl.ac.uk/fatks.

Scott, J. ABBB. Social Network Analysis. London: Sage Publications.

Travis, W. ABBC. The Strict Faceted Classification Model. In Proceedings of IA Summit )**-, ASIS&T (Vancouver, AG-AI March, ABBC). http://facetmap.com/pub/strict_faceted_classification.pdf.

Vickery, B. C. DECB. Faceted Classification: A guide to Construction and Use of Special Schemes. N. p.: Aslib.

Yee, K.P., K. Swearingen, K. Li, and M. Hearst. ABBG. Faceted Metadata for Image Searching and Browsing. In Proceeding of CHI )**.. http://flamenco.berkeley.edu/papers/flamenco-chiBG.pdf.

MACEBOOK_01.indb 217 25-08-2008 13:08:21

This book has been published on occasion of the International ConferenceOn-line repositories in architectureVenice, Italy, !"-!# September, !""$

Curator of the International ConferenceOn-line repositories in architectureFurio Barzon (Collaboratorio, Italy)

Scientific CommitteeHerman Neuckermans, coordinator (%&&%, Belgium)Markus Eisenhauer (Fraunhofer '(), Germany)Alberto Giretti (Università Politecnica delle Marche, Italy)Erik Duval (Katholieke Universiteit Leuven, Belgium)Markus Specht (Open University Netherlands, Netherlands)Martin Wolpers (Fraunhofer fit, Germany)Mario Santana Quintero (%&&%, Belgium)Ezio Arlati (Politecnico di Milano, Italy)

edited by Matteo Zambelli Anna Helena Janowiak (Collaboratorio, Italy)Herman Neuckermans(%&&%, Belgium)

produced by !"#$ Consortium*+,- Collaboratorio snc, Italy.'/( Deutsches Forschungszentrum für Künstliche Intelligenz, Germany%&&% European Association of Architectural Education, Belgium'01 Fachhochschule Potsdam University of Applied Sciences, Germany'() Fraunhofer Institut für Angewandte Informationstechnik, Germany(2- Fraunhofer Informationszentrum Raum und Bau, Germany/3, Katholieke Universiteit Leuven, Belgium034 Humance &5, Germany6&3 Nautes srl, Italy+36, Open University Netherlands, Netherlands1+,(4( Politecnico di Milano, Italy36(714 Università Politecnica delle Marche, Italy(3&7 University IUAV of Venice, Italy31* Technical University of Catalonia, Spain

powered byCollaboratorio, Italy%&&% European Association of Architectural Education, Belgium

published and distributed byFraunhofer (2- Verlag, Germany

supported byEuropean Commission, eContentplus Program

thanks toAlessandro Miniussi, Italy (typesetting)Nick Simcock, England (English check)Hans-Martin Barth and Klaus Kaiser, Germany (publishing)

This book is part of the series%&&% Transactions on Architectural Education no. 8"

MACEBOOK_01.indb 367 25-08-2008 13:09:35View publication statsView publication stats