IntroductionExisting ClassificationsJohansen's Time-Space MatrixOther ClassificationsProposed SolutionCSCW CharacteristicsTime-Space CharacteristicsResulting Classification ArrayClassification ExamplesAssessing The Proposed ClassificationAnalysis ResultsConclusionsReferencesA Classification Method for CSCW SystemsV.M.R. Penicheta,1I. Marinb,2J.A. Galluda,3M.D. Lozanoa,4R. Tesorieroa,5aI3A, University of Castilla-La Mancha, Albacete, SpainbEPSA, University of Castilla-La Mancha, Albacete, SpainAbstractComputer-Supported Cooperative Work (CSCW) is a research field where the role played by individuals asmembers of groups is fundamental. The human being is not considered as an individual entity, but it isconsidered as a being embedded into the society, where he works and interacts. From the beginning, manyCSCW systems have arisen. Some taxonomies appeared in order to find a way to classify all these tools, butthey have become more and more complicated, therefore nowadays such classifications cannot sort themcorrectly. In this paper we present a taxonomy according to these changes, so that CSCW systems can beclassified in a more flexible way.Keywords: taxonomy, classification, CSCW, communication, coordination, cooperation1 IntroductionPaul Cashman and Irene Greif organized a multidisciplinary workshop in the mid-1980s. Many people from different fields but with the same aim: knowing howcomputer science could help people who work together, computer science as a wayto solve their own necessities. In this workshop, Computer-Supported CooperativeWork was coined [5].The term groupware was used by Peter and Trudy Johnson-Lenz before theCSCW one. Its original definition is ”intentional group processes plus software tosupport them”, that is, software that supports group processes. It appeared inWe would like to thank the Spanish CICYT project TIN2004-08000-C03-01 for funding this work, whichwas a lso supported by the grant PCC-05-005-1 from JCCM.1Email: [email protected]: [email protected]: [email protected]: [email protected]: [email protected] Notes in Theoretical Computer Science 168 (2007) 237–2471571-0661/$ – see front matter © 2007 Elsevier B.V. All rights reserved.www.elsevier.com/locate/entcsdoi:10.1016/j.entcs.2006.12.0071981, in the paper ”Consider the Groupware: Design and Group Process Impactson Communication in the Electronic Medium” [9].We could say that CSCW describes the research and groupware describes thetechnology [7]. As Greenberg [6] stated, ”CSCW can also be considered as a sci-entific discipline guiding the design and development of groupware in a meticulousand appropriate way”.Due to the interdisciplinary character of CSCW [5], the number of fields where itcould be applied is very wide. If we just observe, for example, the number of differentfields in the congress CSCW 2004, that took place in November 2004, in Chicago,we can appreciate such a magnitude. There were sessions dedicated to: dynamicarchitectures, collaboration involving large displays, knowledge sharing in softwareengineering, evaluation methods, medical applications, systems, social awarenessand availability, communities, interactions with shared displays, tabletop design,organizational issues, distilling knowledge, gaming, distributed teams, operationaltransformation, gesturing, moving and talking together, bridging the physical andthe digital, information sharing and access, and synchronous collaboration.As systems in general tend to be collaborative, to ease human communicationand to be a useful tool in the processes and human coordination, it is expected thatthe number of applications, ideas, forms, etc. will grow in the years to come. Untilnow, the possibilities to classify tools, functions, etc. related with CSCW are basedon a time-space array [8], always looking for an adaptation to the innovations thatarise, to new possibilities for communication, collaboration and coordination.Taxonomies provide a way to classify different groupware tools. However, it isnot always easy to do it, and sometimes we do not know where to place a simpletool, and even less where to place a complex system. Nowadays systems are socomplicated that they need a new different method to classify them.If we consider a simple tool which has a concrete functionality to solve a specificproblem, such as a chat, the complexity to classify the tool in one of the previoustaxonomies is easy. But considering a complex system that implements severalfunctions, such as a document management system that offers tools, functions forthe version control, sharing, check-in and check-out, publication, approval, etc. theproblem of the system classification becomes more difficult.How do we classify a document management system? The control version func-tion is not considered to be only used by dispersed users. It could even include acollaboration system to create the documents in real time. Or even, it could havea shared agenda for a group of users, would it imply that they are not workingside by side? Therefore, are the classical taxonomies appropriate to classify all thegroupware applications? Existing classification methods that have been consideredso far are correct, but the complexity of the new tools and the big systems havemadetheideaobsoleteinsomeway.In this paper we present a new classification for CSCW systems based on logicalprinciples of this philosophy so as to classify them in a flexible and appropriate way.In the next section, we present some existing proposals that allow us to classifyCSCW systems. We describe the proposed solution in section 3. Section 4 offersV.M.R. Penichet et al. / Electronic Notes in Theoretical Computer Science 168 (2007) 237–247238some examples of the application of the classification method proposed. In section 5,we expose the results and sum up the conclusions extracted from the test developedto know the ”impact” of the proposed solution in the CSCW context. Lastly, section6 shows some conclusions about this work.2 Existing ClassificationsUp to now, groupware tools have been classified in many ways [2, 3, 4, 7, 8], butmost of them are based on an original matrix created by Johansen in 1988 (seeTable 1) [8]. In this section, some of the main taxonomies, as well as the Johansen’sone, are introduced as related and previous works.2.1 Johansen’s Time-Space MatrixIn this first classification, tools