Return to Nick Wallingford's CV

Literature Review

Outline of Section

2.1 The Nature of Electronic Discussion Groups and Moderation

A simple description of the purpose of electronic mail distribution lists and the process of moderation were provided by Hamman (2000). Carter (1999) described the use of LISTSERVs generally and in the educational context of communication and research specifically. A more thorough description of the process of using distribution lists, as well as commentary on the LISTSERV software used to manage most email distribution lists, was provided by the International Federation of Library Associations and Institutions (1999).

Nagel (1994) described the generalised nature of activity cycle through the life of a typical email distribution list, providing categorisations of initial enthusiasm, evangelism, growth, community, discomfort with diversity and finishing with either smug complacency and stagnation or maturity.

The concept of “net etiquette” has been described by many writers, including the International Federation of Library Associations and Institutions (1999), which described addressing conventions, the use of signature files, Subject: lines and general guidelines when working with groups.

North (1994) discussed the existence of “cultural norms” in the context of USENET discussions, and provided a series of simple strategies to assist in the understanding and acquisition of online skills required to effectively participate in electronic discussion groups.

2.2 Ethical Issues of List Research

Sixsmith and Murray (2001) examined ethical issues related to the use of distribution list archives, and discussed topics including “ethical issues of accessing voices, consent, privacy, anonymity, interpretation, and ownership and authorship of material.” While primarily intended for research of a qualitative nature, the underlying principles can be applied more widely when considering the use of material obtained from Internet archives generally.

2.3 Moderation of Lists

Paulsen (1995) described the roles and techniques of moderation and facilitation of computer conferences. While the orientation is primarily pedagogical, the concepts of organisational, social and intellectual facilitations provide a useful framework when considering aspects of control and direction in electronic distribution lists.

Collins and Berge (1997) reported on a pilot study to examine the role of moderation in Electronic Discussion Groups (EDGs), developing a methodology for the examination of the role of “moderator”.

In more recent work, Berge and Collins (2000) analysed list moderators' conceptions of their roles, their rationale for moderating or not moderating their mailing lists, where they learned their craft, and where moderating lists fits into the context of their lives. The individuals responding to the survey described the role of the list moderator variously as “filter, firefighter, facilitator, editor, manager, discussion leader, content expert, helper, and marketer”.

2.4 Mailing Lists as a Research Base and the Associated Use of Computers

Langner (1999) provided a practice oriented introduction to the use of Internet mailing lists as research material. While the orientation was toward Japanese studies for German researchers, much of the discussion was generic in nature. New opportunities and potential limitations in using mailing lists as a source of research material were outlined, and current approaches to research methodologies were presented. The use of computers in the process of mailing list research generally was described as:

“… convenient (automatic) data gathering, following are (sic) steps of data preparation, e.g. incorporating, formatting, cutting, rearranging the material. To some extent automatic classification can be achieved making use of standardized mail formats (e.g. header-fields like author, date, subject; citations; signatures).” (Langner, 1999).

Langner also identified a key issue related to the analysis of electronic mail messages in general:

“We cannot even assume that one mail address means one person. Behind one e-mail address there may be several people, other lists or software agents. E.g. “Tanaka Tomoyuki” is a wellknown figure in newsgroups like soc.culture.japan, but there have been a lot of speculations about his actual identity.”

2.5 Measures of Activity in Electronic Discussion Groups

As described by Sudweeks (2001) ProjectH “… involved a three year collaboration of a large group of people from several countries and many universities to conduct a quantitative study of electronic discussions”. The work involved both quantitative and qualitative aspects, and had a wide-ranging intent:

“The study focused on the single message, the aggregate thread and the lists themselves. A sizable chunk of publicly available, archived computer mediated group discussions was sampled and the content of the messages within the sample was analysed.”

Zenhausern (1994) and Zenhausern & Wong (1998) examined the quantitative information that can be derived from postings to an electronic mail distribution list. Both studies relied primarily upon direct access through LISTSERV utilities to the list archives. Most lists now restrict access to the full range of the software's database and statistical capabilities as a means of reducing unwanted emails to the list and direct to the members.

Rojo (1995) described the use of a similar methodology for one aspect of a study on the participation levels of scholarly electronic forums. Limitations in the use of LISTSERV utilities were encountered and described. The concept of “critical mass” as it applies to electronic discussion groups (the number of participants and contributions required for a list to be perceived as established and on-going) was developed and quantified for a range of distribution lists over a six month period.

Merrier, Duff, and Patterson (1999) examined an internal University email list, concentrating primarily upon the content of messages. They did, however, analyse the list activity for such things as time of day of postings and length of postings. They acknowledged that “… only a relatively small number of electronic mail messages representing only one calendar year and a unique mailing list from only one university” (p. 26) meant that the findings might not be applied generally.

Smith (1999) reported the “initial results of Netscan, a software tool … designed to gather an ongoing stream of Usenet messages [in order to] distill measures of activity and relationships in any collection of newsgroups selected for study”. He determined that the review of email addresses posting to the Usenet can be used to determine countries of message origination, and did extensive analyses of the posting behaviours of the contributing individuals. Several measures of list activity and quality were attempted, including poster-to-post ratios and examination of hourly, daily and weekly activity cycles and patterns.

2.6 Motivations for Participation

Rafaeli and LaRose (1993) provided a framework to describe participation and motivation for participation in dial-up bulletin board systems, given that there is something of a dilemma in that the accrual of “good” is not limited to those who actually contribute to an online discussion. Other members are also able to read the postings without the necessity to write messages of their own.

Rojo (1995) developed considered models for motivation to participate, and provided commentary on three such frameworks: two public goods-based theories (critical mass theory and the discretionary database theory) and a users' perspective of media use.

Matzat (1998) described the need for the development of effective measures of individual and collective activity, and examined issues such as group size and “the production of the collective good” within Internet Discussion Groups (IDGs). A model was proposed that “…specifies determinants of the frequency of active participation in discussions of IDGs by the researchers as their members.”

2.7 Country of Origin and Domain Naming Issues

Mockapetris (1987a and 1987b) described, in two Requests for Comment (RFCs), the details of the domain name system and protocol, and the use of a “resolver” to retrieve the information related to a particular name, allowing for the routing of data across the Internet.

Schwartz and Wood (1993) described the process of extracting host and domain information from information contained within email logs. Following on from this work, Schwartz and Pu (1994) described a software utility called Netfind, used in the early 1990’s to map email addresses to user names, a form of “Internet whitepages”. The research provided useful commentary on the issues related to domain name parsing and relationships, as well as the nature of “super-domains” – the (then common) use of individual host computer names to assist in mail routing within particular domains, such as camelot.cs.cmu.edu as one mail host within the cmu.edu domain as a whole.

Galperin and Gordin (1995) clarified the use of the Domain Name System (DNS) in interpreting second and third level domain labels, and the use of the DNS to map between “pronounceable, easily remembered names (host names)” and the 32 bit integer that comprises the Internet Protocol (IP) address of a given domain.

Zook (1999, 2000) provided analyses on the use of domain names and host counts as indicators of countries of origin and overall diffusion of the Internet. The use of generic top level domains (gTLDs), contrasted with that of country code top level domains (ccTLDs), was quantified and discussed. The increasing use of gTLDs by organisations and companies in countries other than the US indicated that other measures of country of origin, rather than relying solely on ccTLD measures, would be required in the future. Zook (1999) identified the critical issue as “…assigning geographical locations to what takes place on the “spaceless” Internet is especially difficult.”

One proposal that would have, had it been adopted, provided such geographical certainty was described by Stikeman (2001). When considering a new series of top level domains in 2000, the Internet Corporation for Assigned Names and Numbers (ICANN) rejected a proposal for a .geo top level domain to indicate “real-world locations” for individual Internet addresses. Had it been implemented, it would have been used to indicate not only country of origin for a particular address, but in fact down to the neighbourhood and city levels.

While several researchers have indicated the difficulties inherent in relating Top Level Domains (TLDs) to specific countries of origin, proposals to provide more detailed information within the domain naming system have not met with universal acceptance. Maggs (2001) expected the recent introduction of the .us domain to encounter vast increases in usage and new legal problems. Using similar arguments, Associated Press (2001) explained that the domain name system was being stretched beyond its original intentions, and the new domain names were an attempt to perform a directory function, rather than simply routing as originally designed.