Janet Verbyla, Department of Computer Science, Flinders University of South Australia, GPO Box 2100, Adelaide, South Australia, 5001. Phone +61 08 8201 3638 Fax: +61 08 8201 3626janet@cs.flinders.edu.au Home Page [HREF1]
World Wide Web, Web-site Management, Maintenance
This paper proposes seven key habits of effective Web-site managers. This paper is motivated by a desire to see Web-site management in its totality become a main-stream area of discussion and hopefully, research and development. The paper reports on the very early initial stages of a project aimed at establishing the nature of the methodologies and tools required to support effective and efficient Web-site management. One of the long term goals of this project is aimed at developing a software environment (Webman) to support methodologies for effective Web-site management.
The real test of any technology is not the intensity of the initial hype associated with it nor the strength of the buzz of first playing with it. The challenge for the technology is to survive beyond the "honeymoon" period. This means that it has to survive the critical evaluation of its real, long-term benefits against the costs associated with developing and maintaining applications of the technology. It also means that the technology has to reach a level of maturity that encompasses the complete life-cycle of serious full-scale application.
Software in general has in the last couple of decades gone through this process. Firstly the "software crisis" lead to the recognition of the importance of a systematic approach to the software life-cycle; (Dorfman et al, 1996, chapter 1) contains two key papers on this crisis. From this has evolved the significant methodologies and tools of software engineering; see (Dorfman et al 1996) or its summary [HREF14] and the CASE Tool index [HREF15] for further details. More recently, the publicising of the "software paradox" has resulted in a more critical benefit-driven rather than technology-driven attitude to the management of an enterprise's information technology i.e. software and hardware infrastructure.
In recent times it has become evident that Web-based technology is starting to move into this critical phase. For instance, this is evident in the increasing emphasis in the literature on assessing the technology's suitability with respect to the enterprise's objectives. In attempting to obtain information from the Web or even just surfing it, most users quickly recognise that there is a lack of good Web site management as evident in effective design and particularly sustained maintenance. From the other perspective, the lack of established methodologies and supporting tools makes the management of the Web-site life-cycle an ad hoc and or resource-hungry process.
This paper presents some proposals on what constitutes effective Web-site management. This is the very early initial stage of a long-term project which has as its pragmatic aim the development of a Web-site management (software) environment, nicknamed Webman. Webman will provide a suite of integrated tools that will support a systematic approach to Web-site management. Webman will be an open system to cope with rapid change in the area and it will be customizable because different Web-sites have different purposes and consequently different natures and life-cycles.
The proposals are presented in terms of Covey's seven habits of highly effective people (Covey 1989). Initially this was done as an exercise in lateral thinking. But, as hopefully the following discussion will show, the habits proved to be quite relevant when interpreted in terms of Web-site management. The personal habits described by Covey in this bestseller are of course themselves relevant to Web-site managers as much as anyone else.
The general nature of Web-site management is discussed in the next section. In the next section, the proposed habits are discussed. In the final section, the general requirements drawn from this discussion for a Web-site management (software) environment are briefly outlined.
If one was to summarise the mission statement for the Web-site management with which this paper is concerned, it would be along the lines of
This summary statement also highlights the important point that Web-site management is NOT just electronic publication just as software engineering is much more than programming although the quality of the later could be considered critical to the success of the former. Much has been published about the initial building of Web sites particularly the "internals" of pages; for instance much of the Information Quality WWW Virtual Library [HREF16] is devoted to the production of Web Documents. (This is an important site for all Web-site managers.) On the other hand management is an over-arching process encompassing all the dimensions of a Web-site's quality and lifetime such as content, functionality and purpose.
The actual nature of the stake-holders and the priority assigned to satisfying their needs will vary significantly between different Web-sites. Three common categories of stake-holders are the sponsors, the producers and the consumers. The sponsors includes the enterprise which funds the Web-site's infrastructure. Producers includes the providers of the content of the Web-site, the Web-site maintainers and of course, the Web-site manager. The consumers includes the users of the Web-site, not just the direct (on purpose) users but accidental users ("surfers") and other sites that reference this one.This mission statement theme entails stages of activities that broadly parallel those identified in the various model of a software life-cycle. The phases listed by Arch and Whitaker [HREF13] show a mapping of these stages into the process of implementing an intranet. The stages are:-
As mentioned earlier the implementation (building) phase of this cycle has received most attention to date. The material which has been published on the overall development process covers to varying degrees the first three phases i.e. analysis, design and implementation. One notable example of this is John December's quality development proposal [HREF17] that contains
Such proposals need to undergo formative evaluation so that an established methodology can be derived and useful supporting tools created.Reflecting the high level of implementation activity, there are currently many tools available for building a Web-site's content. Many tools focus on a particular aspect of implementation, for instance, Cold Fusion [HREF2] is an Web Database integration tool, HotDog [HREF3] is a commercial HTML editor and Swish [HREF4] generates search-able indexes of Web sites. There are also a number of integrated packages for building and running Web-sites. For instance there is O'Reilly and partners' WebSite Pro [HREF5] and Netscape's SuiteSpot [HREF6]. There is also advanced server architectures such as Hyperwave [HREF18] (formerly Hyper-G) which facilitate Web site administration via such features as "automatic hyperlink consistency".
There are a number of tools available for monitoring access to a Web-site, for instance, Getstats [HREF7] analyzes log output from a variety of HTTP servers, and an interesting variation on this is provided by WebPlot [HREF8] which plots the location of the clients accessing a Web server on a map.
The relationship of these tools to the Web-site management is like that of programming tools to software engineering. That is, they support important activities within the whole process but the overall success of the process depends more on the over-arching methodology, policies and procedures (and the tools which support it). This governs the whole process and glues (integrates) the component activities together coherently and cohesively. It is the lack of such methodologies and support tools which has prompted the Webman project.
Furthermore from this perspective, it becomes clearer which stages themselves lack established methods and supporting tools. It may be possible to import these from other areas such as human computer interface design and workflow analysis but this needs significant investigation. Certainly it is intended that Webman supports the integration of existing tools for the lower level tasks into its environment.
The purpose then of the remainder of this paper is to propose, at a macro level, the key habits (practices) that would constitute effective Web-site management and then to briefly describe what is required of a software environment to support for these habits.
The seven habits proposed for effective Web-site managers will be presented under the same titles as originally used by Covey (Covey 1989). However details of Covey's personal habits will not in general be restated here; instead the source publication (Covey 1989) should be consulted. As mentioned earlier, these habits may themselves be of personal and professional value to Web-site managers (as well as anyone else). Certainly employers of Web-site managers are likely to prefer people who have achieved the goals of these habits, namely,
A fundamental practice that arises from this habit is the ongoing monitoring and evaluation of a Web-site's performance. This involves proactive seeking of and response to potential problems and potential improvements. In other words, this involves proactive maintenance. It is very difficult during the "honeymoon period" of a new technology to recognise the importance of such maintenance and plan for it. But as stated at the Customer Intimacy Web-site [HREF9]
To ensure that maintenance is well done, is it important that procedures and supporting tools be developed to automate and assist in the process as far as possible. For instance, the monitoring tools should not just report on frequency of page hits but also provide path analysis information.
(Covey 1989) begins his chapter on this habit with a visualisation exercise involving the tributes paid at one's funeral. The quality of much of the Web would be greatly improved if more consideration was given to the "death" of a particular piece of information. (Indeed the mortality rate amongst Web-sites would be much higher and the Web a better place, if zombie sites, i.e. those that exist but are not alive, were killed off.) Web-site management requires methods to handle the full life-cycle of information. Unlike humans, it is easier at the "birth" of information to register when it should be modified (or at least checked for currency), retired and expired. Tools could then assist in managing the occurrence of these events.
Another practice that arises from this habit is that of first creation by conscious design. This arises from recognition that things including Web-sites are created twice, firstly by design and then by implementation. A problem that some enterprises currently face is that they have Web-sites which have been created the second time i.e. implemented, with little critical nor systematic effort being applied to the first creation. Universities are a good example of this where individuals and groups all over campus have willy-nilly implemented Web-sites often in that initial best of enthusiasm without considering that these are part of the university's overall Web-site. The effort then is re-create these sites so that they do reflect the same conscious design. This means that they not only adhere to the same style ("look and feel") guidelines but also content guidelines and life-cycle management practices. A Web-site management requires tools that assist in enforcing these guidelines, and in the worse case scenario, prohibit unacceptable material from being included in the site either directly or, more problematically, indirectly via links to other sites. Some of this functionality is incorporated in the repository navigator Ciao [HREF11] and in Waugh and McKenzie's [HREF12] Web submission tool
The previous paragraphs describe important practices aimed at achieving desired end-points. These imply that the end-points themselves have been specified. Indeed, beyond all else effective Web-site management needs a clear specification of the intended purposes of a Web-site. Such a detailed mission statement would govern all aspects of managing a Web-site just as the requirements specifications underpins all subsequent stages in the software life-cycle. Appropriate methods for developing, using and refining the mission statement may perhaps be derived from software engineering methods such as peer reviews and walk-throughs involving representatives of all major stake-holders, see (Dorfman et al 1996, chapter 4) for more detailled discussion.
In the end, at the start, and everywhere in between, effective Web-site management involves knowing who should come to the Web-site, why they should come and how to make them come.At first this habit may appear to contradict the previous one. However while the previous habit is about establishing and keeping in sight the purposes of a Web-site, this habit is about an effective process to achieve those purposes and to implement the policies that arise from these. To paraphrase (Covey 1989) this habit is about scheduling priorities (to get to the desired end-point) as distinct from prioritizing what is on the schedule.
This habit is clearly then about time and task management for which there are available a number of general approaches and supporting tools. However it should be pointed out that Covey does include under this habit a different focus and emphasis. Particularly relevant to Web-site management is the emphasis on scheduling and spending time (and effort) on important but not urgent tasks such as prevention and planning. While the need to accommodate important and urgent tasks when they arise is recognised, a methodology that helps prevent time being spent on unimportant tasks, urgent or not, is clearly invaluable. This then means that methods of assessment need to be in place to measure what is important. For instance, how does spending time on the appearance of a rarely visited page rate?
Many guidelines on building Web sites re-iterate the theme that the user must win for the Web site to win i.e. achieve its purpose. In the initial stages this involves designing the character of the Web site to attract the intended audience of consumers. Thereafter there must be in place mechanisms to ensure that the Web site enables the desired relationships to be built up with that audience. This must entail routines to refresh information and to establish patterns of interactivity. This later refers to mechanisms to involve the users in the site, customizing what the user sees as the site and extending the interaction to other forms of network communication such as electronic mail (list servers).
Obviously to achieve these aims, the Web site management must seek to understand the consumers and listen to their responses. Arguably the sponsors and the (other) producers need to be understood also. Furthermore there must be agreed measures of effectiveness against which performance can be measured.
One of the great challenges of Web site management lies not with the consumers but with the producers and the sponsors. Obviously Web sites involves significant technical input but as the above discussion and much of literature suggests, effective Web sites also draws on many other areas of an enterprise such as perhaps marketing. Effective Web-site management will seek to create a synergeny from the differences between the various sponsors and producers. This means that the processes and supporting tools for Web-site management must not speak only one language. Indeed the processes and tools can be enriched by drawing from those already in the various areas such as graphical design, statistics, flow analysis and sales psychology.
This final habit is an important one with respect to personal and professional development. However in the context of Web-site management, it takes on a slightly different meaning. Because of the pace of change in many parameters associated with the Internet and the Web in particular, it is very important that the Web-site management have in place a process to monitor and evaluate these changes. The purpose of this is to seek better ways to achieve the Web-site's mission and perhaps more importantly, to recognize when the changes actually mean that the Web-site's mission needs to be revisited and revised. One example of such a change would be a significant alteration in the profile of potential consumers who have Internet or intranet access.
This means that management methodologies and supporting tools must themselves be geared to change. For instance the tools need to be designed as open systems so that they can be modified and extended easily.
From the above description, various themes emerge as to what is required of a Web-site management (software) environment. Amongst these are requirements are
The design and implementation of the environment can draw from other areas which have similar requirements such as environment for software engineering, workflow and information management.This paper is motivated by a desire to see comprehensive Web-site management become a main-stream area of discussion and hopefully, research and development. Otherwise it is likely that the costly lessons of the software crisis will be learnt all over again. And as exponents of software engineering have found introducing good habits after the fact is extremely difficult.
S. R. Covey (1989) "The 7 habits of Highly Effective People", The Business Library, Information Australia, Melbourne.
Merlin Dorfman and Richard H. Thayer (Eds) (1996) "Software Engineering", IEEE Computer Society.
Janet Verbyla ©, 1997. The author assigns to Southern Cross University and other educational and non-profit institutions a non-exclusive licence to use this document for personal use and in courses of instruction provided that the article is used in full and this copyright statement is reproduced. The author also grants a non-exclusive licence to Southern Cross University to publish this document in full on the World Wide Web and on CD-ROM and in printed form with the conference papers, and for the document to be published on mirrors on the World Wide Web. Any other usage is prohibited without the express permission of the authors.