Alan Vince

Cite this as: Vince, A. 1997 Editorial, Internet Archaeology 2.

Time has passed extremely quickly in the Internet Archaeology office and it is difficult to believe that it is over six months since the publication of Issue One. Naturally enough, we have been very keen to find out how many people had read the journal, which parts they found good, which ones need rethinking, and so on. Very few readers have given detailed comments but we have thought carefully about all constructive criticism, so please don't be afraid of making your views known, either to the editors directly or through the intarch-interest mail list. In addition to the feedback we get through these means, which are essentially anecdotal, we also analyse the information given to us when users register. This is extremely useful and we are grateful to you for taking the time and for your honesty. All these data are covered in the UK by our Data Protection Act, and you can be sure that they are only ever used in the furtherance of the journal.

I gave a paper on a preliminary analysis of this data, and the log files created automatically when users access Internet Archaeology, at the Computer Applications in Archaeology conference held in Birmingham, UK, in April 1997. This paper will be published in the conference proceedings. A longer version, linked to detailed tabulation of access figures by paper, origin and occupation of user will be published in Issue Three of Internet Archaeology.

2 House style changes

Issue Two has a very similar look and feel to Issue One. We have made a few minor changes to the house style (for example, we now include details about the size of each paper and its illustrations in the issue's table of contents). We also include the URL of each page as a hotlink from the Internet Archaeology copyright notice at the base of the page. So, if you download a page to read off-line and then want to continue to use the journal online you can simply jump back to the live page through this link.

3 Automation of paper production

Another behind-the-scenes change is the increased use of automation to produce and maintain the pages. Two of these tools are modified versions of scripts available on the Internet, pnuts and htmltoc, and the third was inspired by htmltoc but manages the images used in our papers, htmlfig.

3.1 pnuts4

We are using version four of pnuts which is based on John Frank's original (the acronym is apparently Previous, Next, Up, Top and Search) which is part of the WN web server distribution. Authors producing papers for Internet Archaeology do not have to worry about providing navigation links in their papers. So long as they include a comment line in each file at the very end of the text thus: <!-- pnutsb -- > together with a complete list of each filename in the order in which they should be linked then the script will turn this into a navigation bar, plus a copyright notice. For long files we also have a shorter navigation bar which is inserted just after the first heading thus: <!-- pnutst --> . The only difference in the final output is that the copyright notice is suppressed.

3.2 htmltoc

The htmltoc script was written by Earl Hood and allows the user to choose which levels of HTML headings will be included in the Table of Contents and how they will be represented. If the script finds that a heading has been given an HTML anchor (that is, a name or bookmark used as the target for a hypertext link) it uses that anchor. If not, the script invents one. This explains the complex-looking anchors you may notice in some URLs in the journal.

3.3 htmlfig

This script was written for IA by Paul Tyers and uses an external database which contains all the information associated with a numbered figure in an IA paper. This includes not only the filename and location of the image and its caption but also data on the author, rights management, the URL of any link and so on.

4 Research Papers

This issue of the journal includes one research paper, by David Dungworth. This paper, based on the author's doctoral thesis, deals with a very complex dataset; a series of determinations of the composition of copper alloy artefacts which were found on sites in northern Britain and date to the Iron Age and Roman periods. They differ in type as well as location and date. There are, therefore, several separate strands to the work: you can approach the topic chronologically (if, for example, you want to see whether the Roman occupation had any effect on the production of copper alloys), you can approach it through the settlement hierarchy (this area remained under direct military control throughout the Roman occupation and there is a clear division in the settlement archaeology between "Roman" and "Native"), you can approach it through the artefact typology (on the grounds that variations in the methods of production and in the functions of the finished goods led the metalworkers to make different choices in their raw materials). To bring out this complexity required some considerable thought as to what the user might want and the end result is a series of interrelated tables which allow you to explore the data geographically, typologically or by individual artefact.

This paper suggests the power of electronic publication for archaeological research. It will not be long before a work like this could include links to external site reports or Sites and Monuments Records or fieldwork together with links to the online catalogues of the museums who hold these artefacts.

5 Technical Papers

The remaining papers come under the heading of technical papers. They may include the results of archaeological research but this is included as an illustration of a method of analysis. The papers are very different in their scope, ranging from the analysis of Mesolithic debitage scatters to the use of satellite imagery, by way of 3D CAD.

6 The Archaeology Data Service

Finally, one of the major events that has taken place since the publication of Issue One of Internet Archaeology (IA) is the foundation of the Archaeology Data Service (ADS). I am very pleased to be able to publish a paper by its staff, Alicia Wise and Paul Miller, which gives an insight into what the ADS is all about. Without realising it, we had, through htmlfig, actually created a metadata index to the images used in the journal, without knowing that this is what it was. As the authors say, one of the problems of research into data management is that it cuts across every field of academic endeavour and all over the world there are people dealing with the impact of computerisation and the global networking of information as it affects their own neck of the woods. Of course, metadata indexing will be done for the papers in Internet Archaeology.

Some people have looked at our journal, and our mission statement, and then looked at the ADS, and theirs, and want to know how we decide which project is appropriate to which organisation. Our rule of thumb is that if data are being used to "tell a story", it is ours. If not, then it belongs with the ADS. I can therefore see a time when an excavation might place its electronic excavation archives with the ADS but have a subset of that data available through IA as well.

One difference between the datasets might be that the ADS house the data in their native format(s), so that you can download a relational database, fire up Paradox or Access and immediately start querying the data or linking that dataset to those of your own, whereas in our case the datasets stay on the server and it would take some ingenuity for a user to acquire the entire database and get it running on a local computer. Another conceptual difference is that data held by Internet Archaeology are there to support a chain of reasoning. If, subsequent to the publication of a paper, further work took place on the excavation records I would expect ADS to hold an updated version of the archive whereas we would not alter our data but might add a note to indicate that new information was available (the equivalent to the pencilling in the margin you find in old excavation reports). I think that both the IA and ADS approaches are needed and am keen to see some pilot studies and then get feedback from our users. I therefore end this editorial with a call for papers, particularly from potential authors who would be willing to place their data both with ADS and Internet Archaeology so that we can see how this might actually work in practice.


© Internet Archaeology URL:
Last updated: Tue May 27 1997