Digital Data — Ur of the Chaldees: A Virtual Vision of Woolley's Excavations

William B. Hafford

(See email contacts page for the author's email address.)

This article covers preservation and dissemination of digital data in the context of the project currently funded under the title "Ur of the Chaldees: A Virtual Vision of Woolley's Excavations." For an overview of the Ur Project, see the presentation from the Archaeological Institute of America annual meetings, January 2013: here (www.youtube.com/watch?v=IfEtD5AphZQ&feature=youtu.be)and this page on the Penn Museum website.

It is the nature of scholarship to investigate and reinvestigate. Yet, if we do not have access to the raw data, scholarly assumptions, and methodology from the original investigation, we cannot reinvestigate that work, cannot conduct new scholarship on old studies. Thus, preserving information is undoubtedly an important task. For archaeologists, this means that we must not only preserve the physical indicators of cultural heritage and archaeological sites themselves, but we must also preserve the data and the way in which we view them.

It was once next to impossible to provide all data and analyses. Scholars explained their methods as much as possible, but traditional publication limited the ability to cover some points in detail and certainly prohibited the presentation of all of the data they gathered and investigated — especially those that they threw out as invalid.

But in many cases that information was kept somewhere — papers stored in an attic perhaps, or, more helpfully, in the archives of the institution for which the scholar worked. So we have long been generally aware of the need to preserve this material. Current scholars often seek such records to clarify their reinvestigation of older works. There are two main problems, of course. One, completeness of the records, and two, access to those records.

With the increased use of digital publication and digital storage of data, we can more easily provide access to all of the available records. This increases the capability of current and future scholars to perform their analyses, and it should improve the overall quality of their studies.

In the case of legacy paper-based data, the first step for reinterpretation is likely to be making all of the old records digital. This is not an easy, quick, or inexpensive process, but it presumes that the reinterpretation will not be the last one and that others will want access to the same information. Therefore, making old records digital has clear value. It also has very high costs.

Is it worth it?

We at the Ur Project believe that it is, and we are attempting to gather all of the paper records of the 1922-1934 excavations of the ancient city and then convert them into digital form for an online virtual research center. Although Ur was published well for its day, the publications should not stand as the sole record or interpretation of the site; revisiting the material after so many related excavations have occurred is only logical. In addition, despite ten volumes in the Ur Excavation series, not every artifact uncovered was published. Moreover, countless questions arise even concerning those that were published. The best way to investigate these questions is to look more deeply into what Sir Leonard Woolley did at the site and how he came to his conclusions. This means going back over his field notes, field catalogues, and preliminary reports as well as reexamining the artifacts wherever possible. The best approach to that process is to digitize those notes, catalogues, and reports so that access may be easier and available to a larger number of interested users.

Most of the original records are only available in the archives of the British Museum. Many scholars have investigated them for various reasons in the past decades by going to London. Being able to acquire these notes through the internet would reduce the time factor of reinvestigation, not to mention the cost, and prevent reduplication of effort on behalf of the British Museum staff, with the added benefit of protecting the original records from continued handling.

Thus, scanning the notes and putting them online seems a reasonable goal. But how many people will really benefit from that effort? The number of scholars who research Ur in depth is rather low. How many Google searches might there realistically be for 'Ur field notes'? Virtually none. That, however, does not preclude attempts to make a more scholarly resource available on the internet, one that provides the necessary data instead of allowing students, researchers, scholars and the public to find only the sensationalist claims of the less informed online. If a reliable museum site provides the original evidence for all to examine, then presumably that site will climb in the page rankings and eventually be the main source of information about Ur. Perhaps it would even encourage a general audience to look more deeply into the records for meaning, engaging in scholarship themselves.

In spite of this appeal for involving the public, the Ur Project stands firmly behind serving the scholarly community. At the core of any investigation must be the very data that scholars demand and, if that material is not provided, then the site cannot serve any audience. The first step is to digitize the evidence; the second is to provide access to that digitized evidence. The smoother search functions and interpretive remarks that may allow the public to get involved can be created later. The data at the core consist of the original field notes, catalogues, letters from the field, and links or references to all published materials about the site. Images and data on artifacts are also key, linked to the current online systems of both the British Museum and the Penn Museum, as well as to information from the Iraq National Museum when their inventories allow.

Scholars may be the primary audience, but the interested public can be involved as well, even in the early stages of this process. For instance, one possibility is already being tried in our project — crowdsourcing of handwritten records. The scans of the records have, in many cases, already been obtained, but making the text of those records searchable requires time in transcribing them. Computers can do a good job of converting typewritten materials to digital text, but handwriting is much more difficult, particularly when written somewhat haphazardly on note cards that have seen the further abuse of having been carried in the field and shuffled about for decades. Thus, we began UrCrowdsource.org to get help with transcribing the records. More than 500 pages of letters and notes have thus far been transcribed by the effort, mainly through a handful of interested volunteers. The site continues to grow and, though on a small scale, it is making a difference in our work. The handwriting is definitely not easy to read, and Woolley used many abbreviations and words that the public are not familiar with. However, many of the volunteers have some archaeological or historical background and have been keen to work out the meanings. We have also created a page covering the observed quirks of Woolley's handwriting to assist. Quality control remains a concern, but any transcriptions have to be checked, and it is easier to go through transcribed materials than it is to begin from scratch with the handwritten.

By allowing the public to assist in the collection of the data, we further hope to promote knowledge of the resource and increase interest in seeing it completed. Word travels once a seed is planted, and by consistently displaying what we are doing in conference talks as well as on social media and blogs, we are appealing to a variety of audiences.

But we cannot lose site of the primary goal of providing good data and making it easily searchable without placing too much interpretation in the records. This drives us to the question of terminology. Archaeologists will probably never agree on set terms for classes of objects, particularly not over large regions. So how do we classify objects to make them easy to find and yet objectively described? For example, should figurines be listed under votive objects? Were they really dedicatory? And what is a figurine anyway? Is it only defined by its dimensions as different from a statue or a statuette? Or is it solely a material issue, clay vs. stone? Different answers can and will be offered. It may be best for us to attempt to cover them all, though perhaps impossible. Nevertheless, a list of alternate terms discussed by archaeologists in general that connects us as a discipline would be very helpful. In the absence of such a general list, we can only cover the most used terms and define the ones we use, both so that our usage is clear and so that others might connect to our terms.

In the end, we hope that the Ur site can be linked to similar sites as they are created. More and more, archaeologists are putting their data online from both current and legacy excavations. If these sites can be interconnected through machine readable formats and stable Universal Resource Identifiers into a series of linked open data, far-reaching research can be performed. This envisions a much wider scholarly network online, but one that we are slowly approaching. It also implies opening scholarship to more people, even those untrained in particular disciplines.

Regulating or moderating open access systems may be difficult, but as has been seen in projects such as the Real Science Online work of the Citizen Science Alliance (CSA, ironically) — the website is called Zooniverse.org and includes sections on Ancient Lives and Galaxy Zoo — many amateurs, retired professionals, students, and general members of the public are keen to assist. Engaging them can only broaden knowledge and education in general. Balancing the needs of two primary audiences, however, is a challenge that must be addressed as the work progresses.

-- William B. Hafford

An addendum. The Ur of the Chaldees Digital Archaeology Project faces an enormous number of challenges. Members of the project team would greatly appreciate comments from readers about the issues involved. In particular, readers' responses to the following questions would be appreciated.

1. Must we create standardized archaeological terminologies before we can create this type of site or the network of sites envisioned for the future?

2. In a research resource such as this one for Ur, how much emphasis should be placed on human interface and site presentation issues? Is simple access to the data enough, at least at the outset?

3. How many audiences can we really reach? Is attempting a scholarly and public interface too much?

4. What level of participation should the general public have in online scholarly pursuits?

If you would like to offer assistance with these questions, please either use the commentary feature of the CSA Newsletter (emailing the editor) or email the author (using this page for his email address).