The Computer-Assisted Anthropology News (CAAN)

Edited by James Dow and Lee Sailer

Vol. 1, No. 4 July 1985

ARCHAEOLOGICAL CATALOGING SYSTEMS FOR MICROCOMPUTERS

Dennis H. O'Neil

Archaeological Certification Program Palomar College San Marcos, CA 92069

(Although the following article is written for archaeologists, the problems and techniques discussed have much wider applications in anthropology. We are sure that other anthropologist will find it very useful.)

ABSTRACT

The suitability of commercially available microcomputer data base management systems for archaeological artifact cataloging is discussed. File management and relational data base management systems are defined and compared in terms of their strengths and weaknesses. Nine considerations for selecting a program are presented. Finally, twenty leading software packages are evaluated.

INTRODUCTION

Most artifact catalog systems in use today no doubt still consist of card files or some other hard copy form. Typically, each object is carefully described on a separate card in terms of a series of locational and morphological attributes. When it is time for making sense of the data, the cards are sorted manually. Tables and graphs summarizing the data trends are created by hand as well. The entire process of sorting catalog data until significant relationships are apparent may take weeks to months.

Nothing is likely to substitute for the experienced archaeologist's skill in assigning artifacts to types and perceiving meaningful patterns in data. However, computers can save a great deal of time in the simple mechanical sorting of catalog listings and in creating tables and graphs. They also potentially reduce clerical errors and result in a significant storage savings. The fact that computerized catalog lists can be quickly duplicated on other magnetic memory media or printed out on paper provides inexpensive security from fire loss and transportability for data sharing. Additionally, some of the catalog systems provide protection from data tampering by requiring the entry of passwords before files can be examined or modified.

Computer based cataloging is now practical using the current generation of microcomputers. They are more or less portable, sufficiently powerful, and relatively inexpensive (O'Neil 1984). The software packages that are designed for cataloging are called "data base management systems" (DBMSs). Over two hundred of these programs are available for microcomputers and more come onto the software market every year (Alperson 1984a:234). They differ significantly from one another in their capabilities and limitations. Unfortunately, it is not usually possible to rely on the claims of vendors in determining which package is best suited for your specific needs. This report compares the leading DBMSs for microcomputers in terms of archaeological applications.

BACKGROUND

It is assumed that the reader already has a passing familiarity with microcomputers and such terminology as floppy disk, random access memory, and dot matrix printer. However, DBMSs have a vocabulary all of their own. To begin with, the data that would be equivalent to what is found on a single catalog card is a "record." Individual attribute categories on that card are "fields." Artifact length, width, and thickness would each be a different field. All of the records together from a particular site catalog might make up a "file." Actually, which records make up a single file is determined by the user at the time that the data is recorded. The file could be defined easily to include the records from part of a site or even several sites. The length, nature, and quantity of fields and records as well as the means of searching for data in the records are dependent on the specific DBMS. Each program is different in these capabilities.

The simplest DBMSs are file management systems (FMSs). They deal with a catalog in much the same way that you would by hand but more quickly. Each file consists of individual, distinct records which are user formatted for the specific application. These records are sorted one at a time by the software. The systems generally create an index of one to several "key fields" which allow simple data searches to occur much more rapidly by identifying records for retrieval. Unlike a hand sort of catalog cards, these file managers also readily permit complex searches involving a combination of attributes. For example, one can get a list of records for lenticular projectile points, made from obsidian, weighing 2-3 grams, and found associated with burials. Such complex searches take more time than searching only by keyfields. File managers have three significant limitations. Data searches are restricted in nature by the predefined record structure. Also they can search only one file at a time. Lastly, the maximum number of records per file is sometimes relatively small.

If these limitations are not a problem for your particular application, file managers are an ideal choice. They are generally very easy to set up and use by even a computer novice, since they provide simple menus of commands for selection with one or two key strokes. They are also comparatively cheap and several are available for Apple computers in addition to the more expensive IBM PC and other machines employing the MS-DOS disk operating system.

More sophisticated DBMS's are usually of the relational data base management system type (RDBMSs). These systems store data in tables (i.e., "relations"). The fields may be thought of as vertical columns and the individual records as horizontal rows. This structure allows the entire data base to be modified quickly by a single operation. For example, the "project" command constructs a new relation from user selected columns in an existing relation. Such projection would allow building any number and variety of site catalog subsets. The "join" command makes a new relation out of two or more separate ones (Kruglinkski 1983: 13-18). This would allow combining different site catalogs into a single new one. This essentially means that these RDBMSs set up data in such a way that there is an almost unlimited number of different searching, comparing, and manipulating approaches that can be employed. A relational catalog system can still be useful as research interests shift, since you do not have to anticipate all of the questions that will be asked when the data base is created. While RDBMSs overcome the query limitations of the FMSs, they do so usually at the cost of increased complexity in set-up and use. Some require learning a program language in order to customize the data base file structure. Once established, many use a lengthy list of commands to manipulate the data. In other words, in selecting a RDBMS, there often is a trade-off between increased power and decreased ease of use. They generally are higher priced as well. Most of the existing relational software was written originally for computers using the CP/M disk operating system. Subsequently, they were modified for use on the IBM PC and its clones because of their faster operating speeds and larger random access memories.

Beyond the scope of this paper are still more sophisticated software such as the network/hierarchical data base management systems. At present, they are prohibitively expensive and require programing skills for tailoring their use to archaeology. Also not discussed here are the actual program techniques by which the different DBMSs store and access data. These are quite variable and significant for speed of operation. Kruglinski (1983:20-25, 39-53) provides a succinct discussion of these matters.

CONSIDERATIONS IN SELECTING A DATA BASE MANAGEMENT SYSTEM

Choosing a DBMS for artifact cataloging sounds simple. Initially, it would seem that the best system would be the most powerful and the most versatile. On further reflection, it becomes readily apparent that at least nine factors have to be taken into consideration:

1. Who will use it? How willing will the users be to learn how to manipulate a complex system? This mostly comes down to a choice between menu or command driven operations. Menus do not require memorization, but their operation is often slower.

2. What microcomputer, if any, is available already? Does the right DBMS software exist for this machine?

3. Is programing required to set up the particular application? If so, who will do it? Programing is time consuming and relatively difficult to do well.

4. Is there a need for compatibility with other software so that graphing, statistical, and word processing programs can utilize the catalog data?

5. Is there a need to compare and exchange data between different files set up with your DBMS? Most relational systems have this capability, but few file managers do.

6. Is there a need for data to be uploaded or downloaded from other types of DBMSs? Not all programs import and export data easily. Chances of success are much higher if both systems store or can convert data to ASCII code or Data Interchange Format (DIF).

7. How long do fields, records, and files need to be? These are critical concerns that must be considered at the outset since there is much variation in maximum limits set by different programs. Generally, relational systems are less restrictive. However, caution is in order regarding the maximum file length claimed for each DBMS. In fact, the real limitation is often disk space. Using typical artifact catalog data from Palomar College, a 360 K double sided, double density 5.25 inch floppy disk used on an IBM-PC would allow ca. 3,800 records. Using a 10 megabyte hard disk drive, the number of records permitted could increase to ca. 105,300. Any DBMS that allows billions of records or any other de facto unlimited capability cannot be taken advantage of currently with microcomputers. However, some programs partially overcome disk size limitation by allowing one file to be "chained" or "spanned" over many disks. A drawback to this approach is that constant switching of disks becomes tedious. Hard disk drives are much easier to use when dealing with large data bases and they can cut the record sort time by half or more.

8. Can all of the questions that will be asked of the data be defined at the time that the file system is set up? In other words, what kinds of query capabilities will be necessary now and in the future? Again, the relational systems generally are the most flexible.

9. How much money is available to purchase software and a computer if necessary? This should be the least important question. Keep in mind that most of the DBMS software for microcomputers is now being written for the IBM-PC and its clones.

In the final analysis the best DBMS to choose is the one that does what you want and is still users friendly. It would be a mistake to select one that does far more than is needed since it is likely to be more expensive and more difficult to set up and use. Regardless of the type of DBMS that you employ, it won't save time initially. In fact, it will take more time since the record structure has to be created and the data carefully entered. Only later do the benefits appear.

DESCRIPTION AND EVALUATION OF TWENTY LEADING DATA BASE MANAGEMENT SYSTEMS

For comparative purposes, the characteristics of twenty of the leading DBMSs are briefly described in Tables 1-2. They are separated into two tables on the basis of whether they are of the file management or relational type.

The ten file management programs described in Table 1 vary widely in their capabilities and limitations. The industry leader at the low end of the spectrum is PFS:file. It is exceptionally easy to learn and has a straightforward manual. Record forms can be created quickly by simply typing them onto the screen. Records can be added or modified even more rapidly. Entire files can be reorganized and new fields readily added, although these operations are very time consuming and require two disk drives. PFS:file can sort on multilple fields, however, this is much slower than sorting on the one permitted key field. The query capabilities are limited but logical and easy to use. Report formatting is enhanced by the addition of the PFS:report and PFS:graph programs. A major problem with PFS:file is its limit of one floppy disk per file. This is particularly handicapping with the Apple computer version.

Friday is a program with mid level capabilities among the file managers. It is not as user friendly as PFS:file, but it is more capable in terms of file size, the number of allowable key fields, and report formatting. Friday speeds up the process of shifting betweeen tasks by using a circular (rather than nested) set of menus. These allow moving to a new task without retracing steps. Friday is also more versatile in exporting and importing data to and from other programs. It is especially suited for exporting to d:BASE II. Friday can be used to initially build screen layouts and to carry out data processing when the number of files is small. The files can then be transferred to dBASE II for large amounts of data and for special manipulations. Friday has two potential drawbacks. Its limit of 32 characters per field may be too restrictive for catalog fields requiring text rather than simple measurements or other numeric data. Finally, the report formatting procedures are complex for mid level FMS's.

At the upper range of file managers in terms of capabilities are DB Master Version 4+ and Advanced DB Master. They have solved the problem of limited file size by allowing single file "chaining" over many disks. Earlier versions of DB Master were cumbersome to use, but recent improvements make it the optimal FMS for use on an Apple computer and a very useful system for the IBM-PC. With the addition of the three optional utility disks, DB Master becomes a highly capable yet moderately friendly system that even has sophisticated statistical capabilities and three levels of security entry codes. Advanced DB Master has the addition of "macro" or "smart" keys which allow up to 99 sets of user defined commands to be initiated with single command. It also has a "print buffer" which allows the user to continue with other work on the data base while printing is going on.

Perhaps the most elaborate file management system is FMS-80. This program is best suited for well defined cataloging and retrieval with moderately long files. However, it is slow and cumbersome compared to many of the fully relational systems. It uses extensive nested menus for most operations though these may be bypassed with direct commands. File templates are layed out on the screen with your own word processing program run from within FMS-80 through a shell program. A built in programing language can be employed for multiple file access and to design very complex reports. While this makes FMS-80 much less user friendly, it need not be resorted to. The manual is rather intimidating and difficult to use on a casual basis. Compensating for this drawback are numerous help screens that can be called up when needed. The normal query capabilities are nearly comparable to those of fully relational systems with the notable exception of lacking project and join commands.

               TABLE 1 PAGE 1 - FILE MANAGEMENT SYSTEMS
                                                          
                  Concentric        Database         DB Master
                  Information       Manager II       Version 4+
                  Processor (CIP)   (The Integrator)  

OPERATING        PC-DOS            PC-DOS            Apple II+
SYSTEM           MS-DOS            MS-DOS            and IIe


REQUIRES            no                no                no
PROGRAMING

EASE OF SETTING      4                                   
UP FILE (1)

EASE OF USE (1)      4                                   

MAX CHARACTERS      50                60               100
PER FIELD

MAX FIELDS          40                40               100
PER RECORD

MAX RECORDS     65,000            32,767           24 megabytes
PER FILE                                           (200 disks)

MAX KEY FIELDS      40                 5                  

MAX FILES            1                 1                 1
OPEN AT ONCE

FILE IMPORT      ASCII, DIF        ASCII, 1-2-3     DIF
AND EXPORT       PS-file (import)  SYLK, WordStar       
                                   MailMerge

COST              $395              $295              $350

COMPANY          Concentric        Alpha Software   Stoneware
                 Data Systems      Corp.            Inc.

PUBLISHED        Bryan (1984a)     Bryan (1984b)    Brevdy (1983a)
REVIEWS          Kruglinski (1984) Exner (1984)     Robertson (1984)
                 Marshall (1984)   Needle (1983)




_____________________________________________________________________
(1) Scored from 1 to 5, with 5 = most easy and 1 = most difficult.
    Only programs that were personally tested have been scored.



               TABLE 1 PAGE 2 - FILE MANAGEMENT SYSTEMS

Advanced         FMS-80            Friday           Nutshell
DB Master        Version 3.1       Version 1.0      Information
Version 1.0                                         System

PC-DOS           CP/M, PC-DOS         PC-DOS
Macintosh        MS-DOS, MP/M II,     CP/M
                 MmmOST, TurboDOS

    no               no                no               no
            (except for full usage)

                      2                 4                5


                      2                 3                5

   250              255                32            16 million


   250              255                32           60,000


16 megabytes     65,535            65,000             2 billion
(44 disks)

                 unlimited             40           unlimited

     1               19                 1                1


ASCII            ASCII             ASCII, 1-2-3     ASCII
                                   WordStar
                                   dBASEII

  $595           $395-$495           $295             $150

Stoneware        FMS Software      Ashton-Tate      Leading Edge


Brevdy (1983b)   Kruglinski                         Kennedy (1984)
Robertson (1984)  (1983:123-166)                    Myers (1984)
Rubin (1984a)



               TABLE 1 PAGE 3 - FILE MANAGEMENT SYSTEMS
                                                          
                 PC-File III       PFS:file         Ultrafile
                 Version 2.0


OPERATING        PC-DOS, MS-DOS    Apple II+        PC-DOS
SYSTEM                             and IIe


REQUIRES             no                no               no
PROGRAMING

EASE OF SETTING       5                 5                4
UP FILE    

EASE OF USE           5                 5                4

MAX CHARACTERS       65              1679              100
PER FIELD

MAX FIELDS           42               100               50
PER RECORD

MAX RECORDS      10,000             1,000 (Apple)   32,000
PER FILE                           22,000 (IBM) 

MAX KEY FIELDS       42                 1                5

MAX FILES             1                 1                1
OPEN AT ONCE

FILE IMPORT      ASCII, DIF        ASCII            ASCII, DIF, FCM
AND EXPORT       MailMerge         (export only)    1-2-3, WordStar
                 PC-Calc

COST             $59               $140             $195

COMPANY          Jim Button        Software         Continental
                                   Publishing       Software

PUBLISHED        Watt (1984)       Kruglinski (1984) Jacobson (1984)
REVIEWS                            White (1984)
                                   Zarley (1983)

Ultrafile is another extensive FMS that approaches the capabilities of relational programs. It is solid and relatively foolproof. It is particularly versatile in handling numeric data and even has the ability to produce three-dimensional bar, pie, and line graphs. The report generation capabilities are better than average and it supports disk chaining. It is menu driven, however, the wide range of functions to choose from makes it as cumbersome as FMS-80.

Nutshell is a quite different sort of file manager. Its extremely long maximum field and record length along with the ability to have variable length records makes it ideal for written text storage and retrieval. It is one of the easiest systems to set up and manipulate, which makes it ideal for casual use. It is menu driven and has many context sensitive help screens. Every word in every record is indexed in order to facilitate data retrieval. It does not have all of the capabilities that FMS-80 and Ultrafile do, but it is far less rigid in the way it modifies and expands records. Compared to PFS:file, Nutshell is nearly as user friendly and certainly more powerful and flexible in the way it manipulates data. These characteristics may make it one of the future industry standards for FMS's.

PC-File III is in a class by itself. This intermediate level FMS is distributed directly by its author on an almost non- commercial basis. There is no advertising and the price is quite nominal. Despite this unorthodox distribution method, PC-File III is a surprisingly good menu driven program. It is easy to set up and use. Macro keys can be created to save key strokes and data can be security encrypted with entry passwords. The manual is the shortest and most concise of those evaluated. An auxiliary program called PC-Graph is currently under development. The one drawback of PC-III is its somewhat limited number of records permitted per file.

The ten relational systems described in Table 2 also vary considerably in their capabilities and limitations. One of the more venerable RDBMS's is Condor 3. Since its introduction in 1977, it has been revised enough to become a highly stable, crash-proof program. It can be set up and operated by both comparative novices and experienced users without programing. There is a choice of employing already established menus, custom created menus, or direct commands. Data entry and retrieval is straight forward and logical. It also sorts data a little quicker than average for full RDBMS's. In addition to custom reports, it will produce line, bar, and pie graphs as well as scatter charts. The manual is easy to read and there are numerous help screens that can be called up from within the program to answer most questions. A possible drawback to Condor 3 is its limit of 127 characters per field. This might be a problem if much text is likely to be used. Despite this and other minor limitations, Condor 3 could be an ideal cataloging system when data will be entered by students or other casual users.

Another of the first fully relational DBMS's available for microcomputers was dBASE II. It has become the de facto industry standard despite the fact that it requires tedious programing to use it fully. Many users have to resort to hiring a consultant to customize the system to their own needs. Others fall back on user support groups and one or more of the numerous third party dBASE II related books to help them sort out the intricacies. The major problem has to do with the fact that it is highly command oriented and has a limited built-in report generator. The strength of the system is that custom help screens and menus can be programed to creatively overcome these problems. Once that has been done, dBASE II becomes a relatively user friendly yet powerful system. An excellent series of tutorial disks come with dBASE II. A recent sequal program written specifically for the IBM-PC, dBASE III, has vastly improved capabilities and relative ease of use. Most significantly, it increases the maximum number of records per file to an essentially unlimited level and eliminates most of the need for programing. dBASE III currently has the drawback of having a non-copyable system disk. It also is not entirely crash proof. These minor problems will no doubt be dealt with in subsequent versions.

A likely future industry standard for extensive RDBMS's is R:base 4000. Compared to dBASE II, it is a breeze to set up a file structure and it has the capability of handling far more records in a file. It is also quicker at sorting and more flexible than dBASE III, especially when it is used with two auxiliary programs, Clout and Extended Report Writer. Clout is a very forgiving query system that allows use of conversational English. Extended Report Writer permits the generation of complex report formats with enhanced computational abilities. Like Condor 3, R:base 4000 has abundant on-line help texts and combines menus and logical, plain English commands. While Condor 3 is slightly more user friendly, R:base 4000 allows more and larger records. It also provides for password security and extensive validity checking of entries. R:base 4000 would be appropriate for archaeological uses requiring unlimited file lengths and multiple users with different security levels.

Personal Pearl might be a good choice among relational systems if there is a need for accounting and other general business applications in addition to cataloging. It comes with on-disk templates for (1) appointment calendars, (2) phone lists, (3) cash disbursement journals, and (4) personnel profile systems. An additional integrated accounting package can be purchased to use with Personal Pearl. This includes programs for (1) general ledgers, (2) payrolls, (3) inventory control, and (4) accounts payable and receivable. These capabilities would be useful for a small museum or a salvage archaeology firm. Personal Pearl is a relatively easy to use system due to a good on-disk tutorial and many on-line help screens. However, its relational capabilities are rather rigid and limited compared to Condor 3, R:base 4000, and d:BASE III.

DATAEASE is a surprisingly user friendly menu driven program. Records are designed on the screen in an intuitively simple manner, similar to Condor 3 and R:base 4000. DATAEASE is unique, however, in making report formatting easy by providing constant on-screen timely prompts with all possible choices from which the user can select. The program also has the capability of creating custom menus similar to those allowed by d:BASE II. DATAEASE additionally provides for security passwords. This impressive level of user friendliness runs counter to the general trend of increased difficulty going along with increased versatility. However, the relational query capabilities are minimal. In fact they are not much better than FMS-80 and Ultrafile.

Knowledgeman might be an appropriate choice when there is a need for a lengthy text oriented relational file system. It has extensive capabilities including an integrated electronic spreadsheet and a security system. However, it is not particularly user friendly for a novice since it is highly command oriented. Revelation and 10-Base offer much the same capabilities with the addition of an essentially unlimited number of records per file.

           TABLE 2 PAGE 1 - RELATIONAL DATABASE MANAGERS
                                                          
                 10-BASE           CodeWriter       Condor 3



OPERATING        MS-DOS 2.0        PC-DOS, MS-DOS   PC-DOS, CP/M   
SYSTEM                             CP/M  MP/M-80           
                                   C-DOS, Turbo
                                          
REQUIRES            no                no                no
PROGRAMING                                  (except for full  usage)

EASE OF SETTING       3                 5                5
UP FILE (1)

EASE OF USE (1)       3                 4                4

MAX CHARACTERS    9,999                79              127
PER FIELD

MAX FIELDS       unlimited            100              127
PER RECORD

MAX RECORDS      unlimited         36,000           65,534
PER FILE         

MAX KEY FIELDS   unlimited              2                8

MAX FILES            16            limited by            1
OPEN AT ONCE                       operating sys.

FILE IMPORT      ASCII, DIF        none             ASCII
AND EXPORT       1-2-3, WordStar
                 MailMerge

COST               $495              $200             $650

COMPANY          Fox Research      CodeWriter       Condor
                                   Corp.            Computer Co.

PUBLISHED        Derfler (1984)                     Cohen (1984)
REVIEWS          Zachmann (1984)                    Kruglinski
                                                    (1983:55-90)
                                                    Perotti (1984)


_____________________________________________________________________
(1) Scored from 1 to 5, with 5 = most easy and 1 = most difficult.
    Only programs that were personally tested have been scored.


 
            TABLE 2 PAGE 2 - RELATIONAL DATABASE MANAGERS

DATAEASE         dBASE II          dBASE III        Knowledgeman
Version 2.12     Version2.41       Version 1.1


PC-DOS, MS-DOS   PC-DOS, CP/M      PC-DOS, CP/M     PC-DOS, MS-DOS
                                                    CP/M       


    no               no                no


     2                2                 4                


     4                3                 3

   255              254               254           65,535
                                   (memo fields can
                                   have 4,000)
   255               32               128              255


65,535           65,535            1 billion        65,535


unlimited             7                 7           unlimited

    32                2                10           unlimited


ASCII, DIF, SYLK   ASCII           ASCII            ASCII
1-2-3, MailMerge                   MailMerge
Multimate                          dBASE II

  $600             $495              4695             $500

Software         Ashton-Tate       Ashton-Tate      Micro Data Base
Solutions                                           Systems Inc.

                 Cohen (1984)      Jenkins (1984)   Cohen (1984)
                 Goley (1984)      Kruglinski       Kruglinski
                 Kruglinski        (1984)           (1984)
                 (1983:91-21)



            TABLE 2 PAGE 3 - RELATIONAL DATABASE MANAGERS
                                                          
                 Personal Pearl    R:base 4000      Salvo
                 Version 2.0       Version 1.15


OPERATING        PC-DOS, MS-DOS    PC-DOS, MS-DOS   PC-DOS, MS-DOS
SYSTEM           CP/M              CP/M-80, CTOS    CP/M
                                   BTOS
                 
REQUIRES            no                no  
PROGRAMING

EASE OF SETTING       4                 4
UP FILE (1)

EASE OF USE (1)       4                 4

MAX CHARACTERS   unlimited          1,530               78
PER FIELD

MAX FIELDS          400               400               16
PER RECORD

MAX RECORDS      8 megabytes       2.5 billion      32,767
PER FILE         

MAX KEY FIELDS   unlimited            400           variable

MAX FILES        5 (8-bit machines)    40               16
OPEN AT ONCE     10 (16-bit machines)

FILE IMPORT      ASCII             ASCII, DIF       ASCII
AND EXPORT       (export only)     SYLK, 1-2-3      (export only)
                                   dBASAE II        DIF

COST               $295              $495           $496

COMPANY          Pearlsoft         MicroRim Inc.    Software
                                                    Automation

PUBLISHED        Cohen (1984)      Alperson (1984b) Cohen (1984)
REVIEWS                            Cohen (1984)
                                   Kruglinski (1984)
                                   Rubin (1984b)


Salvo and CodeWriter are more restrictive RDBMS's. Salvo's limit of 78 characters per field along with the need to learn a difficult programing language makes it much less attractive than Condor 3 or R:base 4000. While CodeWriter does not require programing, it is comparatively limited in most of its capabilities. However, it has the advantage of being available for the less expensive Apple and Commodore computers.

Due to insufficient descriptive data, two new programs written specifically for archaeological purposes are not critiqued in Tables 1-2. The first of these is MINARK, an extensive cataloging system from Austrlia (Johnson 1984a; 1984b). It is available for PC-DOS, MS-DOS, CP/M-80, and CP/M-86. The second is a CP/M based Canadian automated field recording and cataloging system known as ADACS II (CompEngServ Ltd. 1984). Both programs are comparatively expensive (i.e., $1900 and $1024 respectively), though reduced prices apparently are available to academic and other non-profit research institutions.

CONCLUSION

The use of data base management software in artifact cataloging can ultimately result in time saving, clerical error reduction, and greatly enhanced data manipulation. It is also apparent that such software can cost time, money, and a great deal of frustration if the users are not familiar with the system or if it does not do what is needed. The decision as to which program to select is very important. It is always a good idea to soberly ask at least three questions. First, are you able to use the system? Second, will it do the job? Finally, what features does it have, compared to other programs, that are potentially useful to your application.

Casual users with files of 1000 or fewer short records per file would be best off relying on a highly user friendly file manager such as PFS:file. If there is a need for a larger capacity, yet still relatively easy to use, file management system, then PC-File III, Nutshell, and possibly Friday would be appropriate. If the user needs a file manager to handle very large amounts of data using a floppy disk system, then DB Master would be suitble.

Relational data base systems are generally for heavy users who are willing to learn a relatively complex set of commands. However, they are generally much more versatile and can search and sort large data bases more rapidly. Condor 3, R:base 4000, and possibly dBASE III currently offer the greatest advantages for most extensive cataloging tasks.

For further general assistance in selecting a DBMS, it is helpful to read Alperson (1984a), Bond (1984), Gabel (1984), Gaines (1981), Krajewski (1984), Miller and Goley (1984), and Sirota (1985).

ACKNOWLEDGEMENTS

A debt of gratitude is owed to the following companies for their generous donation or loan of the software reviewed for this report:

Ashton-Tate
CodeWriter Corporation
Concentric Data Systems, Inc.
Condor Computer Corporation
Continental Software, A Division of Arrays, Inc.
DJR Associates, Inc.
Jim Button
Microrim, Inc.
Leading Edge Products, Inc.
Pearlsoft, Inc.
Software Solutions, Inc.

REFERENCES CITED

Alperson, Burton L.
  1984a  Choosing Data Management Software. PC
    World 2(7):234-241.

  1984b  A Champion Among Middleweights. PC 
    World 2(7):151-155.

Brevdy, June
  1983a  Review: DB Master Version Four. InfoWorld
    5(52) and 6(1):52.

  1983b  Review: Advanced DB Master. InfoWorld
    5(52) and 6(1):52.

Bryan, Shawn
  1984a  Concentric Information Processor.
    Microcomputing (8):52-53.

  1984b  Data Base Manager II--Alpha Software's
    Friendly Integrator. Data Based Advisor 2(10)

Cohen, Adam
  1984   Relational Revelations. Digital Review
    1(11):59-66, 106.

CompEngServ Ltd.
  1984   An Archaeological Data Acquisition System:
    ADACS II, (advertising pamphlet), CompEngServ
    Ltd., 925 Parkhaven Ave., Ottawa, Ontario, 
    Canada

Derfler, Frank
  1984   10 Base: Flexible Power. PC July 24

Exner, Ron
  1984   Data Base Manager II--The Integrator.
    Popular Computing 3(8).

Gabel, David
  1984   How to Buy Data-Base Software. Personal
    Computing 8(2):116-125, 206-209.

Gaines, Sylvia W., editor
  1981   Data Bank Applications in Archaeology.
    University of Arizona Press, Tucson.

Goley, George F. IV
  1984   dBASE II. Popular Computing 3(8):182-189.

Heintz, Carl
  1983   Personal Pearl. Interface Age May.  

Jacobson, Bill
  1984   UltraFile. Creative Computing 10(9):68-72.

Jenkins, David
  1983   dBASE II V2.4. PC World 1(10)139-143.

  1984   dBASE III. PC World 2(13):67-173.

Johnson, Ian
  1984a  MINARK Archaeological Database System In-
    formation Sheets. MINARK Research, Suite 1206,
    MLC Centre, 239 George Street, Brisbane,
    Australia.

  1984b  MINARK: A Database System for Archaeolo-
    gists. Advances in Computer Archaeology 1
    (Spring):12-28.

Kennedy, Don
  1984   Nutshell. PC Week 1(23)

Krajewski, Rich
  1984   Database Types. Byte 9(11):137-142.

Kruglinski, David
  1983   Data Base Management Systems: a Guide
    to Microcomputer Software. Osborne/McGraw-
    Hill, Berkeley.

  1984   Update on Six Database Managers: a Com-
    parison of Six Popular DBMSs for the IBM PC.
    Guide to the IBM Personal Computers, special
    issue of Byte 9(9):187-196.

Marshall, Jonathan
  1984   CIP: a File Manager for the IBM PC that
    More than Holds Its Own. InfoWorld 6(20):60-61.

Miller, Michael J. and George F. Ggoley IV
  1984   Making Sense of Database Software: Some
    Information Management Programs are More Equal
    than Others. Popular Computing 3(8):106-109,
    194-197.

Myers, Jay K.
  1984   Nutshell Information Manager. PC Products

Needle, David
  1983   Integrate DBM II with other Software. Info-
    World 5(18).

O'Neil, Dennis H.
  1984   Archaeological Uses of Microcomputers with 
    'Off the Rack' Software. American Antiquity
    49(4):809-814.

Perotti, James
  1984   Condor 3: Mellow and Marvelous. PC 
    3(16):170-171.

Robertson, Barbara
  1984   DB Master and Advanced DB Master. Popular
    Computing 3(8):166-180.

Rubin, Charles
  1984a  Easy Data Management: Advanced DB Master.
    Personal Software April:92-96.

  1984b  Data Base Power. Personal Computing
    8(5):25-26.

Sirota, Warren
  1985   Data Management. PC World Special Edition:
    1985 Software Review 86-105.

Watt, Peggy
  1984   Software for a Donation. InfoWorld 6(24):
    36-38.

White, Ron
  1984   PFS:file and Report. Popular Computing 
    3(8):155-163.

Zachmann, William F.
  1984   Pick of the Month--'10-Base'. Software 
    News (10):12, 20.

Zarley, Craig
  1983   File But Never Forget. Personal Software
    1(1):79-80.

SOFTWARE DISTRIBUTORS

Alpha Software Corp, 30 B St., Burlington, MA 01803. Ashton-Tate, 10150 W. Jefferson, Culver City, CA 90230.

CodeWriter Corp., 7847 North Caldwell Ave., Niles, IL 60648.

Concentric Data Systems, 18 Lyman St., Westboro, MA 01581.

Condor Computer Co., 3290 W. Bayshore Rd., Palo Alto, CA 94303.

Continental Software, 11223 S. Hindry Ave., Los Angeles, CA 90045.

FMS Software Ltd., 54 north Main St., Spring Valley, NY 10977.

Fox Research, 7005 Corporate Way, Dayton, OH 45459.

Jim Button, P.O. Box 5786, Bellevue, WA 98006.

Leading Edge Products Inc., 225 Turnpike, Canton, MA 02021.

Micro Data Base Systems Inc., POP 248, Lafayette, IN 47902.

MicroRim Inc., 3380 146th Place SE, Bellevue, WA 98007.

Pearlsoft, P.O. Box 638, Wilsonville, OR 97070.

Software Automation Inc., 14333 Proton Rd., Dallas, TX 75234.

Software Publishing Co., 1901 Landings Dr., Mountain View, CA 94043.

Software Solutions Inc., 305 Bic Dr., Milford, CT 06460.

Stoneware Inc., 50 Belvedere St., San Rafael, CA 94901.


ERRATUM WORLD CULTURES DATABASE

In the last issue of CAAN La Jolla was incorrectly spelled La Lolla. Thus the correct address for subscriptions to the World Cultures Database is

World Cultures, P.O. Box 12524, La Jolla, CA 92037-0650

The World Cultures Database is a journal of cross- cultural codes, codebooks, bibliographies, sampling frames, atlas information, reviews, and articles published on microcomputer diskettes. The cost is $60 annually plus a one-time database entry fee of $95. See CAAN Vol. 1 No. 3 for more information.


HUNTING AND GATHERING TALES

This section of the CAANews is devoted to reports of what anthropologists are doing with com puters. Please send the information on what you are doing, your successes and failures, and what you are interested in to the editors. We will publish your reports with your name and address so our other readers can get in touch with you if there is a possibility of mutual assistance. The reports are going into a data-base that can be searched in the future if someone needs to find a person to help with a CAA problem. We will also publish comments here.

Please let us know if you would also like to participate in on-line conference to exchange this information. For this you need a terminal, a local network port, and at least $6.00 per hour.

CALL FOR ELECTRONIC MAILBOX ADDRESSES

CAAN is asking for readers to send in their electronic mailbox address. We will publish these regularly in a CAAN directory of electronic mailboxes so that our readers may communicate via computers. Send in addresses on national mail and conferencing systems, computers linked to national networks, and even local computers, if anthropologists have access to them.

CENTER FOR ARCHAEOLOGY FIELD TRAINING PIMA COMMUNITY COLLEGE

After acquiring our first microcomputer in 1979, the decision was made to shift as many as possible of our office field and research operations to personal computers.

Hardware: Word processing and large database projects are done on a Northstar Horizon that has been upgraded to a multiuser, multiprocessor, l5MB hard disk system. Up to 8 users can be accomodated running a mixture of 8 and l6 bit software under the TurboDOS operting system (CP/M compatible). Communication to our other computers is done via modems or hardwiring. Cartography, CADD, and instrument interfacing are accomplished using various Hewlett-Packard series 80 microcomputers. This system supports plotters, digitizers, electronic calipers and digital balances. It also is set up to upload information from our field data collectors which are Hewlett-Packard 4lCV programmable calculators. Software: We utilize a variety of commercial programs a well as programs developed in-house for specialized applications. A more detailed discussion of our work with personal computers can be found in the July/August (1984) issue of ARCHAEOLOGY magazine.

I hope that our latest project, ANTBBS, will be operating in the next few weeks. A survey of the available bulletin board systems led us to write our own. The following features are planned for the bulletin board system: general message area, current, fieldwork, job listings, ongoing research, and a section for abstracts of current but unpublished research. The abstract section will run under DBASE to allow keyword searches. The major intent of ANTBBS is as a source of information on research that has yet to be published. In addition, it can serve to promote communication between anthropologists interested in personal computer applications. Contact person: David V. M. Stephen, Director, Centre for Archaeological Field Training, Pima Community College, 2202 W. Anklam Rd., Tucson, AZ 85709-0001. Tel: (602) 884-6022.

A RUGGED FIELD COMPUTER

Leland Gilsen, Historic Preservation Office, Oregon Deptartment of Transportation, Salem, Oregon

Anthropologists interested in a rugged field computer should keep their eyes on the Delta Data Systems Corporation's portable Tempest microcomputer. It is designed for the military and is known as the Tote-PC - the transportable operations tactical equipment personal computer - quite a mouthful. Tote is a small portable that is MS-DOS compatable and runs IBM software. It is expandable to 640K and is 256K standard. It has 5 expansion slots and weighs 32 lbs. It has a full size IBM keyboard and is designed for battlefield use. Any micro that goes into a battlefield should be looked at by field researchers. U.S. forces in Europe are field testing Tote at the current time. It is designed for NATO and field command use. It runs packages like dBASE, Wordstar and other standard applications. It can run off a battery pack or a car battery. It can hold a 10MEG hard disk or dual diskettes. Its boards include an RS232, printer port, etc. It will sell for around $10,000. Just wait around for a while and see them come onto the market as surplus equipment...or make a deal to field test them under conditions that would simulate true battlefield conditions in "primative" areas. The company might be interested in such tests. I assume it is designed for a number of current sources for U.S. and European voltage systems...but have no hard data on that aspect.

If it is designed to work in the dirt and the heat and the humidity of world-wide battle conditions...then it may be the perfect micro for anthropology. The IBM compatability is important too. If anyone out there has better data on this system...they should keep everyone informed through CAAN.

COMPUTERS IN SOCIAL WORK

Computer Use in Social Services (CUSSN) Network is a non profit association of professionals interested in exchanging information and experiences on using computers in the social services. Members participate in the Network by

A. Sending materials for the CUSSN Newsletter such as (1) member needs interests, hardware/software use activities (2) information resources and (3) longer reports/articles on conferences, surveys, vendor products, ideas, experiences, computer applications and events. Those wanting longer pieces to be anonymously reviewed by CUSSN advisory board members should so indicate.

B. Participating in the skills bank and software clearinghouse.

C. Distributing Newsletters to friends and at workshops and conferences.

D. Referring vendors. if you think a vendor/consultant could benefit by exposure to CUSSN members, tell them, so they can advertise their services and products in the CUSSN Newsletter.

E. Holding local CUSSN meetings. Local meetings in Dallas, Ft. Worth, Chicago and Baltimore have been successful. For those in a foreign Network dues are $5 for students and the poor, $10 for individuals and $10 + for those willing to provide additional support. Those interested in joining the Network should write to Dick Schoech, CUSSN Coordinator/Editor, The University of Texas at Arlington, Box 19129, Arlington, Texas 76019. Make checks payable to CUSS Network. Please indicate if you do not want your name provided to those interested in using the CUSSN mailing list.

The CUSSN Newsletter is published approximately 4 times a year and is sent free to all network members. Institutional and library subscriptions are available for $15 a year. For overseas all mail, add an additional $5 for postage. All prices are in U.S. dollars. Back issues of the newsletter are available for $2.50 each. Volume 1 has 2 issues.

The CUSS Skills Bank allows members to locate or share specific knowledge, skills and experiences.


HOW TO SUBMIT MATERIALS TO CAAN

Material should be submitted to the editors at the addresses which follow. News, letters, notes, and articles are welcome. We prefer material submitted electronically on 5 1/4 diskettes or sent to User 70150,266 on CompuServe. Send short items for "Hunting and Gathering Tales" to Jim Dow. Subscriptions should be sent to Lee Sailer. Anything else can be sent to either co-editor.

  Lee Sailer               James Dow
  CAANews                  CAANew
  Dept. of Anthropology    Dept. of Soc. and Anth.
  Univ. of Pittsburgh      Oakland Univ.
  Pittsburgh, Pa. 15260    Rochester, Mich. 48063
  Tel: (412) 624-3388      Tel: (313) 370-2430

We can accept diskettes in the following formats: A.B. Dick Magna II, Actrix (SSDD, DSDD), ATT 3B2, Chameleon CP/M-80 (SSDD, DSDD), Chameleon CP/M-86 (SSDD, DSDD), Cromenco CDOS (SSSD, SSDD, DSSD, DSDD), Cromenco IT CP/M (SSDD, DSDD), Datavue, DEC VT-80, DEC (Rainbow, Pro), Epson QX-10, Heath/Zenith Magnolia CP/M, HP-125, IBM-PC CPM- 86(SSDD, DSDD), IBM PC-DOS(V1, V2, SSDD, DSDD), IMS 5000, Kaypro (2, 4, 10), LOBO MAX-80, MAGIC, Morrow (MD-2, MD-3), NCR Decision Mate V, NEC PC-8001 (SSDD, DSDD), any Osborne, Otrona, any PC clone with MS-DOS, Sanyo MBC-1000, Superbrain (Jr. 35 track, 40 track, QD) Televidio (TS802, TS803, TS806, TPC-I), TI Prof. CP/M-86, Toshiba T-100, TRS-80 I Omikron CP/M, TRS-80 III Memory Merchants CP/M, TRS- 80 IV Montezuma CP/M, Xerox (820, 820-II (SS, DS)), Zenith (Z-90, Z-100 (SS DS)).

Those with formats not listed here are invited to call: we can probably figure out some special arrangement. This offer even applies to Apple users.