|
A set of articles in the March, 1997, issue of Scientific American was brought to my attention by a colleague. The articles, under the collective title, "The Internet: Bringing Order from Chaos," (pp. 52-83) concerned finding information on the Internet, and there were several views about how that will or should happen in the future. The inadequacy of current search procedures was, of course, the starting point for the discussion. Interestingly, the contributors seemed to assume that solutions to the problem of locating information on the Internet would be automated solutions. They considered ways to attach descriptive data to Web pages, for instance, so that search programs could locate the appropriate descriptive data in response to a search request. (The descriptive data-often called metadata, data about data-can be included in a Web page without showing up on the browser screen.) There was also some discussion of ratings for Web sites, with rating information either attached to the Web pages or available through other sources -or both. The ratings could be for nearly any aspect of a Web site or page, though the perceived need for ratings now stems from concerns about obscenity on the Internet.
A much shorter article caught my eye in the May 27, 1997, edition of PC Magazine ("Can the Web Hurt Schools?" by Bill Howard and Sebastian Rupley, p. 30). This one dealt with a presentation by Michael Crichton, author of Jurassic Park, among other things. Mr. Crichton described a school report given by a youngster-a multi-media report on the Maya, with information taken from the Web and presented on a computer to his teacher. Though the report was praised by the teacher, Mr. Crichton characterized it "'as heavy on glitz and dead wrong on facts.'" He used that experience to support his contention that "'There is a host of data supporting the idea that technology does not make a difference in the classroom.'"
Yet a third article on this subject appeared in PC Week for August 25, 1997 ("Developing a Card Catalog for the Expansive Web," by Eamonn Sullivan, p. 34). Here again, the problem of finding appropriate material on the Web was discussed, and developing Web standards were suggested as a potential cure, permitting (or, in the long run, requiring) Web pages to include descriptive information about their own content. Similarly, indications of appropriateness for specific ages and other ratings might be included.
These three magazine references obviously have something in common, though the Scientific American and PC Week articles were concerned with finding what is there, and Mr. Crichton derided the accuracy of what was found. All the articles pointed out the need for help with information on the Internet. The Scientific American articles were direct in suggesting automated searching, using metadata in Web documents to help find materials, and a reference to one academic effort to standardize the metadata categories was mentioned. Mr. Sullivan in PC Week also suggested standards, but from the organization that oversees the World Wide Web. Mr. Crichton, according to the reporter, was not looking for help; he hoped to slow the advance of the computer into the classroom. None of the writers saw libraries or librarians as aids to the problems, much less as solutions, though all who were concerned with finding materials mentioned parallels to library functions.
The traditional library and librarian, narrowly conceived, may not be the solutions; so other names or broader definitions may be required. The names aren't the issue, but I believe that some people or organizations must perform some of the gate-keeping services traditionally performed by libraries and librarians-not only making subject indices to lead people to appropriate materials but providing some form of quality control as well.
Emphasizing the library function, one large, federally-funded project to assist with finding and distributing geospatial information is called the Alexandria Digital Library (ADL). The ADL project is intended to make a wide variety of maps and images available, and two indexing schemes are being used, U.S. Machine Readable Catalog (USMARC, the standard U.S. library indexing system) and the Federal Geographic Data Committee (FGDC) extensions to USMARC. With these indexing aids, users can find geospatial information by place name, spatial footprint, data theme, and date of compilation. (See http://alexandria.sdc.ucsb.edu for more information.)
The Alexandria Digital Library, unlike most of the indexing notions referred to earlier, is based on constructed, not automated, indices. I believe that such prepared indices are preferable to automated searches of the Web. There are many reasons for that. One, I think the discipline of an outside evaluation of a resource is important; having constructed indices rather than searches of all Web documents requires a more self-conscious process, one that will, in my view, result in better and more accurate searching. Two, I am concerned about a wider array of electronic data than the Web, including CD-ROMs and files for distribution via the Internet, for example, in addition to Web documents-and not about all Web materials, but only those that have scholarly value. Three, I do not want to build a system for current Web technology and then find it in need of reconfiguration when some new generation of communication technologies arrives. Finally, I think the existence of an index, not just a search process, will encourage those who create resources to submit them to indexing authorities for inclusion, making it more likely that their resources will be found when they should be. Given my preference for prepared indices -and for evaluations of electronic resources-I believe some gate-keeping organization(s) are required. The scope of the work suggests that we will need separate indexing groups and indices for various subjects.
Among the institutions that might take on the indexing role for archaeology are the Archaeological Institute of America and the Society of American Archaeology. Either could review Web sites and other Internet sources; either could solicit reviews to assess quality and accuracy. Both could lead people to interesting materials about archaeology. Both could, therefore, serve the library/librarian functions for archaeological information on the Web. The Archaeological Data Archive Project (ADAP), the Archaeology Data Service (ADS), and other archaeological archives might also do this work; archival organizations will certainly be involved in indexing- indeed, they already are-since they must assist with access to their own materials and those held in other archives.
Work on determining the appropriate metadata categories for network resources has been going on for some years, and the development of specific standards was begun at one early meeting. "The March 1995 Metadata Workshop, sponsored by the Online Computer Library Center (OCLC) and the National Center for Supercomputing Applications (NCSA), convened 52 selected researchers and professionals from librarianship, computer science, text encoding, and related areas, to advance the state of the art in the development of resource description (or metadata) records for networked electronic information objects." ("OCLC/NCSA Metadata Workshop Report," by Stuart Weibel, Jean Godby, Eric Miller, and Ron Daniel, http://www.oclc.org:5046/oclc/research/conferences/metadata/dublin_core_report.html, Section 1.0, Executive Summary, citation as of September 15, 1997) The meeting was held in Dublin, Ohio, and resulted in metadata categories now called the Dublin Core. (This was the academic set of categories to which one of the authors of the Scientific American articles referred.) Similarly, indexing for the ADAP was first discussed in the CSA Newsletter in August of 1994 ("A Database About Databases,").
Personnel at the ADAP have tended to stick with the term index, while many others have used the term metadata, The terms are not interchangeable, but the aims are the same. Those who use the term metadata generally start with the need to describe a resource for others - to provide data about data, usually within the file itself, hence metadata. Those who prefer to speak of indexing start with the need to find information through resources that are, more or less, complex indices. In both cases, the aim is to find the categories of information that should be recorded about any electronic resource. Some consider those categories to be the ones required of any resource provider; others take those categories to the ones needed by the user of an index to locate appropriate resources. In either case and whether talking about metadata or indices, the point is to provide information that allows access not only for scholars within a discipline to the data of their discipline but for people in any discipline to information from any other. For instance, archaeologists may want to use geophysical data to help help locate ancient metal or clay deposits; geologists might want information about ancient metal working or ancient trade to try to locate existing metal deposits. The possibilities for cross-fertilization are enticing.
ADAP personnel considered indexing needs very early in the development of the archive but began working more urgently on these problems again after the Computer Applications in Archaeology conference last spring. Starting with the proposal from the Metadata Workshop mentioned above, the Dublin Core, personnel at the ADAP created a set of index categories and tried them out on a group of resources either in the Archaeological Data Archive or on the net, applying the indexing ideas to real resources so that difficulties could be found sooner rather than later. We tried to use as diverse a group of resources as possible, and we discovered that, though we started with the Dublin Core categories, we could not satisfy ourselves with them alone.
The categories of the Dublin Core and the ones we added are more complicated than those in the kinds of indices we are accustomed to consulting. For instance, books have authors, editors, contributors, illustrators, and so on A full index of electronic resources should not only include all contributors to the resources, it should also make explicit the job of each for the specific resource referenced. Thus, our index of names is not simply that; it is an index of names and job titles - Jane Jones, principal investigator; John Doe, database designer; Mary Smith, CAD modeler; Jack Brown, ceramics specialist. Similarly, institutions that support archaeological projects may play different roles; they, too, need complex index categories.
Our most vexing category turned out to be a group of categories - one would not do the job - to indicate the location of an archaeological site or survey area. We believed that the index should provide the means to search by political names (countries, states, etc.), geographic regions (river valleys, mountain ranges, and the like), cultural regions (which may be indistinguishable from cultural period designations), or precise mapping coordinates. Furthermore, political names could be ancient, modern, or related to any period between. (Someone should be able, for instance, to find sites in what is now Russia but was once the Soviet Union without retrieving information about all sites in the USSR. Similarly, a search for sites the Rio Grande Valley should retrieve sites in the U.S. or Mexico.)
We have assumed a database environment for the indices; such a structure makes it easy to allow multiple entries for any given resource. For instance, a geographic indicator might be Arizona, Southwestern U.S., and Colorado River Valley for the same resource. Similarly, the cultural indicator might be Pueblo III and Anasazi, ancient Greek and Hellenistic, etc.
It will be obvious that most categories of information apply only to the data under study (the sites, artifacts, survey data, etc.) or to the study itself, but some may apply to either one. The latter is the case, for instance, with chronological indicators. An excavated site has a first and last date of occupation; the process of excavating that site also has a beginning and ending date. There are also dates attached to the publication of electronic resources.
To the extent possible, the authors or other persons responsible for a given resource should be the ones to determine the appropriate entries for their resources. Those responsible for the index should decide as few such questions as possible. But they must evaluate and approve the information provided by resource providers.
Some of the problematic categories may be of interest to the general reader and are discussed here, but, on the assumption that most readers will need less detail, the complete list of the indexing categories is not included here. That list will be attached to the Web version of this article, as will actual index entries for a selected group of resources. We encourage readers to examine the list and the samples and to indicate to us their thoughts and suggestions. Readers should also examine the information on the Web site of the Archaeology Data Service (http://ads .ahds.ac.uk/ahds/welcome.html) on this subject as well as http://www.oclc.org:5046/oclc/research/conferences/metadata/dublin_core_report.html for the original report on the Dublin Core.
One problem area is determining what makes an individual data item and what is a collection, each part of which must be treated separately. Breaking down individual resources into units may yield collections that have individual items that are themselves collections of individual items that are themselves collections . . . . Some rather arbitrary decisions will doubtless be made in order to determine the smallest item to be indexed.
In some cases, doubtless more and more as time goes on, an electronic resource will have been taken from a non-electronic original. The nature of the original, as well as the processes used to convert the resource, should be stated. The original will not be indexed; only the digital copy will be.
Materials will often be available in multiple file formats. Users must know not only what formats are available but what the original format was and what processes were used to translate the data from that original format to any other.
For each resource there must be an explicit statement of the review process for the material before it was posted. Much material on the Internet has been made available by the author(s) without the sort of prior review that would be taken as axiomatic with a paper publication (see "Electronic Publication - The New Vanity Press" in this issue). The review process, therefore, should be explicit so that potential users may know what prior review, if any, was carried out. There should also be some way to indicate that the materials have been reviewed since being made public (especially important in the absence of prior review) and where such reviews may be found.
Unlike paper publications, materials made available on the Internet may be changed at a moment's notice. Therefore, users must know when files were created and last updated; they must also know if there is any intention to change or modify the files.
There will ultimately be many resources that are directly related to others-not simply resources on similar topics but additions to earlier data sets, amplifications of other resources, and corrections of earlier errors. Indices must include bi-directional pointers to and from those related items; there must also be descriptions of the relationships. We consider this to be among the most helpful of the kinds of data to be included, since the pointers will help people to learn of the many small corrections and additions to the record that might otherwise have remained obscure because of their relative unimportance. These kinds of corrections are also more likely to be made a part of the public record, I believe, if there is a constructed index that potential contributors of data can count upon to link their corrections and additions to the original files.
Web addition to printed document.
A complete list of the indexing categories we have defined follows. Of course, not all resources will require all categories. We hope readers will respond with their suggestions and comments. After the list of categories I have added examples of some resources to show how they might be indexed with these categories. The indexing information is not necessarily complete, but the resources included were chosen to provide a variety of resource types and to stress the indexing scheme rather than the actual index data.
Indexing categories:
1. Unique identifiers - Every resource must have a unique identifier. Since the resource may be a particular item or a group of items - for instance, a Web site or an individual Web page; a data table or an interconnected set of tables, there must be two potential identifiers, one for a specific resource and another for a collection of resources. Breaking down individual resources into units may yield collections that have individual items that are themselves collections of individual items that are themselves collections . . . . Some rather arbitrary decisions will doubtless be made in order to define the smallest item to be indexed.
2. Title, subject, and description - Each resource must have a title, a subject, and a description, and, here again, there may be differing approaches for collections and for individual parts of collections. Descriptions must be long and full enough to be truly helpful. Subjects will eventually be chosen from a limited list of possibilities, though no resource should be limited to a single subject entry.
3. Type of scholarship - Any resource may be based on a field survey, a site, a catalog raisonné, etc. That must be made explicit. Text material, the Web version of an article or monograph, must also be identified as such.
4. Location of the resource - Users must know where to find the resource, both to gain access (URL, ftp site, etc.) and to know the person or institution responsible for maintaining the resource.
5. Archiving - If a file has been archived, the fact will be noted, along with the name/location of the archival home. This may seem superfluous, but many indexed items will not be in archives but nonetheless available on the Internet or as CDs. (The Dublin Core categories include Publisher, but we have separated the location of the resource from the archive. Each could be considered the publisher.)
6. People - There will be information about the people who worked on a resource, including a job title. Address, telephone number, e-mail address, and institutional affiliation will be recorded when available but may be considered optional.
7. Institutions - Similar information about the institutions that supported the resource creation will be indexed.
8. Language - The language(s) used in the materials must be stated.
9. Original source & conversion processes - In some cases, the electronic resource will have been taken from a non-electronic original or from an electronic original that required conversion. The nature of the original, as well as the processes used to convert the resource, should be stated.
10. Data types - Potential users must know the forms of the data, including both the generic identifications (e.g., Web documents, data tables, text files, etc.) and the specific file types (HTML, DBF, DOC, RTF, etc.). When multiple file formats are available, the additional formats will also be included. (Note that users will know whether they are using the originally supplied format or a secondary one.) To the extent that the files were manipulated to migrate them from one format to another, that must also be noted.
11. Instructions/documentation - Resources will require considerable documentation if they are to be used intelligently. The availability of and access to such documentation will be included.
12. Permissions for use - Copyright information and any other restrictions on use will be explicitly stated.
13. Creation & modification information - Users must know when files were created and last updated; they must also know if there is any intention to change or modify the files.
14. Review process - For each resource there must be an explicit statement of the review process for the material before it was posted. Much material on the Internet has been made available by the author(s) without the sort of prior review that would be taken as axiomatic with a paper publication. The review process, therefore, should be explicit.
15. Reviews - The files will include bibliographic information for reviews of the resource, though it is clear that this may be difficult. It will be very important, though, for those resources not subjected to prior peer review.
16. Geographic location - Maximum and minimum latitude and longitude of the site, survey area, or artifact locations will be included. (These categories may be changed to UTM designations.) In addition, there may be entries for river valleys, mountain ranges, or other regional designations. Other geographic terms and hierarchies will also be used so that users may search by broad geographic areas (continents, for instance) as well as narrower ones.
17. Political location - Country, state, county, township, and city or town can be included. This may be somewhat problematic with a resource such as the Ming Dynasty vases in U.S. museum collections. Which political entity is intended, China or the U.S.? Since multiple entries are permitted, both would be included in the indices. Users would be expected to use other categories in a search to prevent turning up inappropriate items.
18. Cultural location/time period - There will be designations of either location or chronology - or both - with terms like Anasazi, Classical Greece, etc. As with the political location, certain studies might require multiple entries; antiquarian studies, for instance, might require multiple entries.
19. Absolute chronology - We will include designations of chronology with absolute dates (expressed in years, with a negative sign to indicate a date before the common era) for the material under study. Needless to say, such dates must be taken with great caution, but omitting them would make access to the data by non specialists very difficult. There will also be explicit dates to define the time when the research was carried out as opposed to the material under study.
20. Related resources - There will ultimately be many resources that are directly related to others - not simply resources on similar topics but additions to earlier data sets, amplifications of other resources, and corrections of earlier errors. Indices will include bi-directional pointers to and from those related items; there will also be descriptions of the relationships. We consider this to be among the most helpful of the kinds of data to be included, since the pointers will help people to learn of the many small corrections and additions to the record that might otherwise have remained obscure because of their relative unimportance. Indeed, such small additions and corrections are more likely to be made public if such indices are available to bring them to the attention of readers.
Some resources and how we would catalog them (more may be added in the future; these are the examples listed as of October, 1997):
1. The ADAP Web materials from Michael Adler's The Prehistoric Pueblo World: A.D. 1150-1350.
Title: Site Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350
Identifier: ADAP C1 and ADAP U1 - This resource needs to be treated as both a collection (there are 12 tables, each on its own page, that can be reached from this starting point) and as an item in its own right (access to the group is through this starting page, and documentation is here as well). Therefore, there are two identifiers, a collection number and an individual number.
Subject: Prehistoric/Ancestral Pueblo sites
Description: A Web page with basic documentation for and links to tables of Pueblo sites known to have existed between 1150 and 1350 A.D.. There are 12 individual tables (each on its own Web page), each for one of 12 defined geographic regions. Terms and definitions are supplied in the base Web page. Additional specifications, as required, are defined for individual tables.
Type of scholarship: Site survey, literary (This was not the result of new on-the-ground surveys.)
Location of the resource: http://csa.brynmawr.edu/web1/prepw.html is the starting page for these materials; Web site of the Archaeological Data Archive Project
Archiving: Archaeological Data Archive Project
People: Michael Adler is listed as the editor, Amber Johnson as assistant for data compilation in appendix, and each of the contributors as a contributor. (Further information about each, including home institution, where appropriate, would be on file in another data table.)
Institutions: University of Arizona Press, publisher of original source; Crow Canyon Archaeological Center, sponsor of original meeting to begin project; ADAP, Web publisher/archive (Further information about each institution would be on file in another data table.)
Language: English
Original Source: Appendix of The Prehistoric Pueblo World, A.D. 1150-1350, University of Arizona Press, 1996; text file altered to make HTML files by Archaeological Data Archive Project personnel; some additional text included in site home page to describe data handling.
Data types: Web site; HTML
Other data types: ftp-available (through browser) data table with all 12 tables combined, DBF file, generated automatically from text file; data table with all 12 tables combined, comma-delimited ASCII file, generated automatically from text file; data table with all 12 tables combined, DBF file, alteration (carried out by ADAP personnel) of automatically generated file to separate numeric and non-numeric portions of single data entries; data table with all 12 tables combined, comma-delimited ASCII file, alteration (carried out by ADAP personnel) of automatically generated file to separate numeric and non-numeric portions of single data entries
Instructions/documentation: included in basic Web page and ftp-available (through browser) text file
Permissions for use: original copyright held by University of Arizona Press
Creation & modification information: posted 1996, modifications limited to typographic corrections, but modifications of ftp-available data tables to separate numeric and non-numeric data were carried out in 1997
Review process: original source subjected to normal peer review by publisher
Reviews: Alan Osborn, American Indian Culture and Research Journal, 21(2): pp. 313-317, 1997; Tammy Stone, American Anthropologist, 99(2): 456-7, 1997; Dean Saitta review, in New Mexico Historical Review, vol. 72, pp., 279-280, 1997; no author stated, New Mexico Magazine, Nov. 1996; no author stated, Books of the Southwest, A critical checklist of current Southwest Americana, Vol. 40 (11), Nov. 1996.
Geographic location: maximum latitude: 38 degree north, minimum latitude: 32 degrees north; maximum longitude: 115 degrees west; minimum longitude: 105 degrees west.. (Individual Web pages would have additional geographic specifications to indicate the limits of their coverages.)
Political location: southwestern U.S., Arizona, New Mexico, Colorado, Utah, Nevada; northern Mexico, Chihuahua (all apply to the area under study)
Cultural location/time period: Anasazi Pueblo III, & Pueblo IV for the sites under study
Absolute chronology: min. 500 A.D.; max. the date of the study (note that this seems to differ from the dates in the study title, but these are the minimum and maximum dates of the sites that were inhabited at some time between 1150 and 1350 A.D.) - project dates 1989 to 1995
Related resources: none
2. One of the individual Web pages included in the Web materials defined in number 1.
Title: Southern Rio Grande Region (Rio Abajo, E. Border Pueblos) Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350
Identifier: ADAP U12 within C1 - This resource must be treated as an item in a larger collection (it is part of the Site Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350, above). Its item identifier is U12. That is sufficient to find the item, because it is a unique identifier. It is also identified as within C1, meaning that it is part of the collection labeled C1.
Subject: Prehistoric/Ancestral Pueblo sites of the Southern Rio Grande region
Description: A Web page with a table of Pueblo sites known to have existed between 1150 and 1350 A.D. in the southern Rio Grande region. Terms and definitions are supplied in the base Web page (http://csa.brynmawr.edu/web1/prepw.html). Additional specifications are defined with the table.
Type of scholarship: Site survey, literary (This was not the result of new on-the-ground surveys.)
Location of the resource: http://csa.brynmawr.edu/web1/lowerrg.html; Web site of the Archaeological Data Archive Project
Archiving: Archaeological Data Archive Project
People: Michael Adler is listed as the editor, and Katherine A. Spielmann as contributor and author.
Institutions: no institution is associated with this Web page other than the ADAP as Web publisher/archive
Language: English
Original Source: Appendix of The Prehistoric Pueblo World, A.D. 1150-1350, University of Arizona Press, 1996; text file altered to make HTML file by Archaeological Data Archive Project personnel
Data types: Web site; HTML
Other data types: ftp-available (through browser) with related tables in a single data file; see Site Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350 (C1 - the number of the parent collection)
Instructions/documentation: individual documentation included in basic Web page; for general documentation for all tables see Site Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350 (C1), http://csa.brynmawr.edu/web1/prepw.html
Permissions for use: original copyright held by University of Arizona Press
Creation & modification information: posted 1996, modifications limited to typographic corrections
Review process: original source subjected to normal peer review by publisher
Reviews: none listed (see C1)
Geographic location: southern Rio Grande region, maximum and minimum latitude and longitude have not been included but will be.
Political location: New Mexico
Cultural location/time period: Anasazi, Pueblo III, & Pueblo IV for the sites under study
Absolute chronology: min. 1000 A.D.; max. 1830 A.D. (note that this seems to differ from the dates in the study title, but these are the minimum and maximum dates of the sites that were inhabited at some time between 1150 and 1350 A.D.); project dates 1989 to 1995
Related resources: this is one of 12 tables that make up the Web site, Site Data from Appendix of The Prehistoric Pueblo World, A.D. 1150-1350 (C1), the other 11 would be linked here as well
3. The ADAP Web materials from Michael Cosmopoulos' The Early Bronze 2 in the Aegean.
Title: Artifact data from The Early Bronze 2 in the Aegean
Identifier: U14 and C2. This resource must be treated as a collection (there are several data tables that make up this one resource) and as an individual item (it is a single Web page with providing access to ftp fownloads).
Subject: artifacts that were used for an analytic study of the artifactual assemblages of EB 2 Aegean sites (mainland Greece, the Cyclades, Crete, and the East Aegean, including Troy); the seven tables contain information about metal, ivory, stone, bone, and clay artifacts as well as figurines and seals.
Description: A Web site with ftp-available tables. Terms and definitions are supplied in the base Web page.
Type of scholarship: Artifact survey, literary
Location of the resource: http://csa.brynmawr.edu/web1/eb2.html; Web site of the Archaeological Data Archive Project
Archiving: Archaeological Data Archive Project
People: Michael B. Cosmopoulos is listed as the author.
Institutions: Jonsered, publisher of original source; ADAP, Web publisher/archive
Language: English
Original Source: The Early Bronze 2 in the Aegean, Studies in Mediterranean Archaeology 98, Jonsered 1991; Lotus WK4 files constructed by author and placed on the University of Manitoba server (http://www.umanitoba.ca/faculties/arts/classics/BronzeAge2.html)
Data types: Web site; HTML
Other data types: ftp-available (through browser) data tables; each of the seven tables was downloaded from the University of Manitoba server, put into database format, and altered (extraneous material removed) by Archaeological Data Archive Project personnel; files are available as comma-delimited ASCII or DBF files
Instructions/documentation: included in basic Web page (and in publication)
Permissions for use: original copyright held by Jonsered
Creation & modification information: posted 1996, modifications limited to typographic corrections and removal of extraneous material
Review process: original resource subjected to normal peer review by publisher
Reviews: none listed
Geographic location: Eastern Mediterranean; maximum and minimum latitude and longitude have not been determined but will be included.
Political location: Greece, Troy (1), Anatolia (Troy (1), not simply Troy, because there are other cities with the same name, and they must be distinguished.)
Cultural location/time period: Early Bronze 2 Greece
Absolute chronology: we will ask the author to specify his definition for the span of EB2 and for the dates of his work on the project
Related resources: none
For other Newsletter articles concerning the Archaeological Data Archive Project or the use of electronic media in the humanities, consult the Subject index.
Next Article: Archaeological Institute of America Annual Meeting items
Table of Contents for the Fall, 1997 issue of the CSA Newsletter (Vol. X, no. 2)
Table
of Contents for all CSA Newsletter issues on the Web