The CSA Newsletter began publishing reviews of web sites in the Spring, 2005, issue. The first review, published in that issue, was a second look at the Çatalhöyük Web site. It was a follow-up of a review first published in the Bryn Mawr Electronic Resources Review, now part of the Bryn Mawr Classical Review, in 1998. There have now been five reviews in the CSA Newsletter (not including the one in this issue), and a review of the reviews does indeed show some common problems. it may be instructive to see what those common difficulties are -- bearing in mind that these difficulties are no more than the growing pains we should expect from the web at this point in its history. The problems do not mean that the sites reviewed are not good but that the web is still in its infancy. We all have much to learn.

* Permanence: None of the sites reviewed has stated an explicit process for archival preservation of its content. Not one. As most readers of the CSA Newsletter will understand, this is a very distressing matter for me. I spent a number of years trying to persuade colleagues to archive digital data; so I am very sensitive to the issue of disappearing information. Lest that sentence be misunderstood, let me be very clear: the authors/editors of these sites may have archival plans. They just do not state what they are. Given the nature of the web -- information comes and goes at a very rapid rate -- I believe that is irresponsible. A scholar who places important information on the web has a responsibility to do one of three things: (1) arrange personally for archival preservation of the data, (2) post the information at a site that will accept the responsibility for the archival preservation, or (3) state clearly and unequivocally that archival preservation is not planned. No scholar wants to cite information from a source that will not be available for readers to check long into the future -- though we are, for some unknown reason, generally willing to do so with web resources.

* Peer review: Only one of the web sites reviewed has stated a specific policy regarding peer review. FastiOnline puts up articles in a portion of its site, and those articles are peer reviewed prior to posting. Once again, it seems to me that the scholars who post materials to the web have neglected to live up to their own stated standards. They may review materials on the site, but there is no clear statement to that effect.

Full disclosure requires that I admit here that there was no explicit statement on the CSA Newsletter site concerning peer review of newsletter articles although other items on the site do have explicit statements. Apparently we believed that peer review would not be expected of a newsletter. The home page of the newsletter now makes an explicit statement regarding peer review -- in multiple places to be sure it will be found. Shouldn't any site? Shouldn't any scholar know whether what he/she is reading has been fully vetted?

* Design consistency and quality: Not surprisingly in these relatively early days of the web, nearly all the sites exhibited considerable variety in terms of design consistency. Only one of the sites reviewed, one constructed by a web design firm, had a truly consistent interface. That was also the only interface that did not present obvious difficulties to users, from disappearing side-bar menus to pages that seemed to come from unrelated teams.

* Dating of information: Here again the sites tended to fail to provide consistent information about when the material was prepared. The Hacımusalar site was the only one that made a reasonably consistent attempt to date the material.

* Complexity and ambition: In general, the web sites reviewed here have shown something neither surprising nor unexpected. As sites become more complex and ambitious, they also become more likely to fail to meet their own goals. Relatively simple sites such as TradSun may not have lofty ambitions, but they do what they claim to do. In other cases the ambitions are unbounded -- and the results predictable. As sites try to provide online access to database information, for instance, they tend to exhibit more and more problems.¹ Similarly, as they broaden their coverage, sites seem to become more and more difficult to navigate. In addition, more complex sites tend to break down more often.

Equally problematic are poorly-defined boundaries and objectives for site content. Those lead naturally to sites that grow erratically and unpredictably. As a natural result, users may not be able to determine either what is available or what is intended. When a user is unable to be sure what will be found at a given site, the user tends not to return.

The list of issues is long, and it may seem that they paint a grim picture of the state of archaeology on the web. That would be a serious misreading of the situation, however, since the use of the web for scholarship is still in its relative infancy, and mis-steps are to be expected -- and we at CSA recognize that our own site must have its shortcomings. Nearly all the sites that have been reviewed have been good without being perfect, helpful without being all one might ask, innovative without being entirely successful. The purpose of this review of review is not to damn but to encourage improvement and a more self-conscious approach to the use of the web. (For an earlier appeal along the same lines, see "Scholarship and Electronic Data," by H. Eiteljorg, II, CSA Newsletter, VIII, 4; February, 1996: http://www.csanet.org/newsletter/feb96/nl029603.html.)

There is also good news, not simply good information or interesting data but thoughtful and innovative approaches. One such innovative idea that leaps to mind is the attempt of TAY to encourage user involvement in keeping data up-to-date and accurate. Although the process to get readers' comments used by TAY was relatively simple, it is an important attempt to use the community aspect of the web to great advantage. That is not easy, as Wikipedia's editing problems have shown all too clearly, but it is a very worthwhile aim. The more people who read critically and have an option to help correct errors, the fewer errors there should be. We look forward to seeing and reviewing more such innovative sites and to watching scholarly web sites grow and prosper.

-- Harrison Eiteljorg, II

To send comments or questions to the author, please see our email contacts page.

---

1. I should acknowledge a bias here. I am not persuaded that it serves anyone to put excavation or survey data up on the web while a project is still ongoing. While access to project data sets after the conclusion of the project should be standard, access prior to project completion strikes me as a very large burden on the project that produces few benefits. Return to text.