ARCHIVAL INFORMATICS NEWSLETTER

Part 1of Archival Informatics Newsletter & Technical Reports

ISSN 0892-2179 WINTER, 1987 Volume I, #4

ANOTHER YEAR As an inducement to prompt renewal, an With this issue, the Archival Informatics index to the 1987 Archjval/nformatics Newsletter completes its first full year of Newsletter w1l1 be sent to those who renew publication. I'm pleased to report that next for 1988 or purchase back issues as part of year it, and the Archiyal Informatics anew subscription in 1988. Technical Reports, w111 be available not only I want to thank those who have contributed through Archives and Museum Informatics, to the publication this year, and especially but also through the Society of American thank Tom Brown whose "Machine-Readable Archivists in Chicago and the Museum Views" column has been consistently DocumentatIon Association in cambri£t;Je, stimulating. Tom has agreed to continue his England. Individual members of SM. .ar~ now column In 1988. entitled to a 10% discount on Subscflptlon prices whether they renew through Archives and Museum Informatics or sign up through TABLE OF CONTENTS SM.. All European orders will be handled through the !"IDA from now on. I am extremely pleased to have the SM. and MDA ArUcles: signing up subscr'lbers and distributing MARCON: ASoftware Review p.66 publications. Along with the mail I receive, David Bearman this is further testimony that the Newsletter Machine Readable Views p.70 and Technical Reports are serving auseful Thomas E. Brown purpose. Edltorlal Commentary p.71 In 1988. the Archival Informatics SAA-CGAP Planning Group Report p.72 Technical Reports will include: * AD1rectory of Software and Systems for Regular Features: Archives and Museums * Proceedings of a seminar on Authority Conferences Control Museum Computer Network p.74 * ACase Study of Appraising Online Am. Soc. for Info. Science p.75 Governmental Records Systems with Authority Control seminar p.76 discussion papers. and Conference calendar: 1988 p.77 * Functional Requirements for Exhibit In-Box p.78 Management Systems, Projects & Proposals p.82 The Newsletter will continue to have the Software Forum p.84 same current awareness content it has had Software Products p.86 and hopes to include software reviews of new 5tandarets p.87 systems scheduled for release In 1988 by . Technical Report Exec. Summary p.88 IME and the MDA (TINmus), Vernon Systems (Collection), and Cactus Software as well as others as they appear. MARCON PLUS: TEXT RETR IEVAL SOFTWARE A REV IEW MARCON and MARCON Plus are versions of asoftware package from AIRS, Inc., 335 Paint Branch Dr" College. Park. MD 20742; It runs on an IBM PC-AT, COMPAQ 386 or 286 or compatib les with hard dlSk, under DOS 2.1 or later (3.1 for networked version). It requires aminimum of 640K RAM and at least a 10MB hard drive. MARCON operates with acolor or monochrome

. ~. monitor. outputs to most standard impact printers and accepts input from the DEST PC-Scan ond other OCR devices. MARCON is available without security, cross database searching or saved search features for $495. MARCON Plus, includes the above three features for $695. ALAN version of MARCON Plus for five workstations costs $3200 ($6400/10; $11,000120), MARCON Plus is also used as aretrieval system for custom developed CD- ROM; prices negotiab leo This review of MARCON Plus 3.01 was conducted on an IDR 386 with 1MB of RAM and 201"'IB of disk,

MARCON is text retrieval software Qm: tool. this review examines two major aspects excellence. It 00es wonders with large of its use: how easily it can construct an databases of irregular, variable length application (and how complex that records including significant text. It builds application can be) and how well it performs databases with sophisticated authority data entry, retrieval and reporting from the control and retrieves from any or all fields resulting appl1cat1on. of records in one or more files using equally sophisticated online retrieval approaches. BUILDING AN APPLICATION Archives and museums will be attrooted to Getting Started: When we first MARCON by the ease with which they can encounter MARCON, it consists of five disks build and search complex databases but it is of highly compressed program data and a not without weaknesses. While MARCON is a manual. Installation proceeds smoothly, powerful. generic text retrieval system, although it may require modifications to ideally suited to constructing and searching your "config.sys" file and will consume about files comprising aresearch catalog, it is less 4MB once finished "unchrunching", well suited to public use as acatalog, or to "unpack'ing" and "unsQuashing" (messages collections management and/or project which appear on your screen as administration applications. decompression routines owned by PKWARE shareware, go to worl<), Initially. you will MARCON is atool. The system is driven by get directly to the main menu, but as systems adata dictionary which defines every field administrator you will be free to introduce used in any record in any file (AIRS calls security control requiring others to prOVide files "collections"). Because MARCON is a apassword at this stage to view, or update, tool, and not aready made app lication, any collections and you may direct them to a everything in MARCON begins with the fields different inWal menu. and the collections which they comprise. Documentation and User Interfoce: When we want to define anew database. we At the main menu I ifyou select "Data Entry name acollection (a file), identify any and Design" you will be asked to identify a previously defined fields which will be used. collection in which you want to enter data, or and define new fields in the collection. When that you want to design. About this paint you we want to add data, we define acollection and will encounter one of MARCON's weaknesses: are permitted to add data to fields according its online help often goes on for several to the rules which we previously established densely packed screens, and the printed in the data dictionary. When we want to documentation, which now comes in avery search acollection, we identify the fields professionally printed and bound package, is which will comprise the search argument, not as good as It was in earlier releases, and then fill in the values which we want to Indeed, it often neglects to give essential search in each field. Because MARCON is a instructions, or prOVides them in the most

66 Archival Informatics Newsletter vol.! #4 unlikely places. Occasionally it conflicts with the online help; and it rarely builds from the same structure. The documentation Menus can alw~s be navigat9d by the no longer follows the w~ in which a user is cursor. tab, or a unique first character, likely to encounter MARCON, and Ifound the whichever the user prefered. Items can be logic difficult to follow. though the index was transferred from one window to another by (1)Od. Apparently, AIRS hired a technical "point &. shoot" (locating the cursor on the· writing firm to produce this mess; they desired item, and hitting ENTER or RETURN) advised me that they have now decided to which is an especially nice feature when return to their in-house writer and will combined with the winOOws MARCON employs soon republish the oocumentation, but for indexes. Of course. consistency does not beware. mean that the interface is intuitive; and the Fortunately, one of MARCON's strengths is 23 commands created by combining the Alt that it has avery consistent user interface. and Shift keys with ten function keys, Although my documentation failed to explain provide numerous features which users must that F7 would delete collections, records, learn, or they will fail to take advantage of fields, etc.; I eventually found it in atable MARCON's power. It would be helpful, explaining the function keys (below). and especially for occasional users, if the most was flne from then on. Similarly. once I basic of these functions were defined on learned that F8 would perm1t me to change relevant screens in an options bar. the item on which the cursor was located. and Design: The power of MARCON becomes that FlO saved my changes, Icould move evident os soon as we begin to define the first through the design process with ease. Held of the first collection. We can def1ne a highlighted screen box of any size, anda field Summary of the Function Keys of any size. so that the field may be required to fit in, or scroll through, the box. We can define the data as the usual alpha-numeric. Fl Help numeric, and date. or as adate stamp from ALT-Fl User comments system time, and also have the option to F2 Index list window define it as calculated from adefined ALT -F2 Index list evaluation equation. We may restrict the field to one F3 Page to previous record occurrence or define 11 as multiply ALT -F3 Copy from previous record occurring. in which case it may have many SHIFT -F3 Saved query - use values within a single record, or we may F4 Page to next record define it as inverted text. so that eoch word, ALT-F4 Import from DOS file (excepting those on an editab1e stop word . f5 View enlarged field list), will be indexed separately. We can ALT -F5 View all fields define a Ollisi!' or index, which will hold SHIFT -F5 View query values of the field. and restrict data entered F6 Print in the Held to values pre-existing in that ALT-F6 Export Dos file list, or to unique values (not in the list), or SHIFT -F6 Reports permit the list to grow by addition of new F7 Delete values, thus constructing an index on the fly. F8 Edit design Ultimately we will discover that many fields SHIFT -F8 Ed1t ANY IALL query can be defined to the same list, that the list F9 Define a new ... will be viewable in a window at data entry ALT -F9 Define new nested field and searching. and that values may be F-l0 Save/Go entered. or used in seorches, by poi nt and ALT-Fl0 Sovequery shoot from that window, thus reducing SHIFT -F 10 Save/Go and begi nanother ... errors. The 11st can be viewed in asorted or

Winter. 1987 Copyright by Archives & Museum Informatics 67 unsorted fashion. And we have a huge range Boolean AND or OR with any other hits. Any of value checks that can be applied to the number of values may be searched in afield, field, either as awarning or as an error and each may be com bi ned as an AND, OR, or condition, each with its own system NOT. V61ues may be specified as exact ooministrator provided messages. Clearly, matches, as wild cards of one or more designing each field will tal

68 Archival Informatics Newsletter vol.1 #4 decisions in the data entry module keep it acknowledges that there have been seriolJs from being among the best For example, bugs and is working to repair them, after you enter a record, the screen defaults apparently with success, but buyers should to a "choose arecord to modify" screen, evaluate this feature fully. rather than directly to the next blank MARCON fi les may be un loaded, and template. Fields can't be defined as reloaded, in a MARCON specific format. incremented nor can default values be defined Further improvements in file output are in the data dictionary - both of which would planned, inclUding implementation of MARC make data entry ~ faster for some Helds. In record input and output, but apparently it is my copy, warning messages acted like error hard for acompany which makes what might messages and cou ld not be over- riliten as be the premler PC-based text retrieval documented package to invest enough in reporting. For Reporting: For two years, reporting has the present, the inability to report across been the crippling weakness of MARCON. This collections is aserious limitation for anyone release, 3.0, was designed to acilress this designing acollections management system, handicap with a new report writer. If the where multiple files are linked in ways report writer performed as oocumented, the wh'lch cannot entirely be satisfied even by module would be weak, but acceptable. nested fields and repeat occurences. However it doesn't, and isn't. Two exceptionally useful kinds of "default" UTILITIES: reports are proviOOd by MARCON. The first Database management: MARCON provides is a printout, or write-to-disk, of ~h some useful utilities to assure that the record in one collection. The user selects databases being constructoo are consistent, which fields will app~r , up to thirty ( !) up-to-date, and error free. One function sort fields, and whether each record will updates all the inverted text indexes for print on aseparate page or three lines after newly enteroo, or alteroo, records. One the previous record. Field labels will print, feature alluded to earl1er permits global each field will app~r on a new line, and data chonges in records from indexes, from on ly one collection will be reported. automatically replacing values in records The second default report is a printout of any with new values in the list, and allows of the index lists, complete with the number indexes to be deleted. A"database manager" of occurences per term if desired. function reorganizes the storage of adatabase MARCON now allows athird kind of report, which has been heavily edited, saving space one that can be defined using the new report and increosing searching efficiency. Abatch writer. The only reporting capability that loOOing fac'ility provides a method to reload works flawlessly here will print atable of MARCON data or load an ASCII file into afield fields and values for selected records or all or collection. These functions, which take a records, again from one collection. All three reasonable amount of batch system time, of these reports are easy to use (tables are have useful on-screen countltlwns indicating more complex to define, but not much) and the number of records left to process ond are useful. summary screen reports when the process is Unfortunately, for the present, we must completed. stop here. The report writer oocuments the Securityadministration: In MARCON method for writing a report in which the Plus, the systems administrator is page location of each Held selected is defined empowered to establish user accounts with by coordinates and fields may be displayed passwords for the application and each of its with or without labels anywhere on a page of major modules, for acollection, for afield, any size (effectively permitting formats ond for a function (ociI, edit). This level of such as mail labels, card catalog cards etc.). security is MeQuate for most applications In my copy, these reports oon't work. AIRS and works well. OCXlly I within this module

Winter, 1987 Copyright by Archives & Museum Informatics 69 the administrator also sets the colors for each users' screens. While I don't imagine this will be customized by most systems MACHINE-READABLE administrators, it is worth noting that VIEWS MARCON has done an excellent job using color, which contributes to ease of entry and by THOMAS E. BROWN searching, but that its highlighting is also fine for amonochrome monitor.. To conclude the first full year of these Thesaurus: MARCON developed out of a ruminations on archives and machine­ system that AI RS president, Tad Durr, readable records, th1s column w111 harken designed for management of documents at back to earlier comments. These remarks BRISC, an urban studies research center in will probably tie up some loose ends. and, at Baltimore known to many for the.!.lJ::!lan. the same time. unravel some tied ones. Information Thesaurys which Durr authored The first issue of the Archiyal Informatics there. The system was cmigned to handle a Newsletter (vol. 1# 1) contained atechnical thesaurus, and 1s st111 marketed with a information leaflet on commercially module for thesaurus development, but my available database management systems copy d1d not contain this module. It has been ( DB MS's) which B111 Reader and I wrote. I used by the Art & Architecture Thesaurus have since had the opportunity to discuss the and other archives and museum projects, but leaflet with atechnical representat1ve of a is not being pushed by AI RS at present vendor whose product the leaflet discussed. because of the strain 1t places on the system. Happily, the leaflet accurately outlined the AI RS expects to reissue the thesaurus as part technique to extract information from the of anew architecture for the overall system, DBMS and put it into asoftware and in 1988. hardware independent format. Further, the vendor representative estimated that SUMMARY someone famillar with the DBMS would need only about 15 minutes to write the program MARCON is an exceptionally full featured to unload the data in the desired format. information retrieval system, well suited to However J he adm1tted that processing time the construction of online research catalogs, would be inordinate. He estimated that it and databases containing large Quantities of would take about 12 hours of run time (3 text. It is less well suited to structured hours of CPU time) to unload adatabase of database applications, especially those about 5 m111ion characters. Thus, for this requiring substantial printed reporting, for DBMS. it is not aQuestion of whether off­ which it will be found wanting. loading the data is technically poss'ib1e, but whether it would be feasible. DavId Bearman Inthe second issue (vol. 1#2), John MacDonald of the National Archives of Canada NOTE: After announcing that I would write urged that archivists should view themselves this review. I was contacted by AIRS to assist as corporate users who need to include them in designing aMARC I/O module for archival concerns 1n functional the1r system, and have accepted acontract requirements during the system design for that work. This review has not been seen phase. I note that about ayear ago. the by anyone at AIRS and I will not begin work National Bureau of Standards issued on the MARC I/O module until after it is guidelines to U.SJederal Agencies on how to : . published, so I feel there is no influence write functional specifications for database from them, but I do want to clarify what management systems. The obvious Quest10n could be seen as aconflict of interest. is whether this guideline, issued as aFederal Information Processing Standard (FIPS),

70 Archival Infor matlcs Newsletter vol. 1#4 addresses the archival concern over . disposition do not apply to machine-readable transportability. records with similar information. The guideline aludes to transportability 2) While machine-readable records have twice. First, it defines conversion capability been appraised primarily for their as one of the "global data f~tors". informat1onal value, they may have Unfortunately, the samp Ie specification here evidential value. Thus machine-readable is for inputting fixed length or comma records may be appraised as having delimited data into the DBMS, not for evidential value worthy of continued outputting similar formats. The second preservation although textual records wah reference appears under "other comparable contents exist. specifications" as a requirement for 3) The overall goal of the National "database transporting". Here the Archives is to accession only records in a specification 00es include requirements for software and hardware independent format. offloading data. It says: However, in unusual situations, the "The DBMS must have afacllity that can information may have such value that selectively dump portions or the entire destruction is not justified and m6'y' be so database (i.e., the defined schemas and software and hardware dependent that the associated data occurences). The dumped file resource commitment for reformatting is not must be loadable by the DBMS. The DBMS justified. In such asituation, the National must allow copying the database record Archives can accession a data file in a occurrences to standard files for information hardware and software dependent format and interchange. " leave 1t to afuture researcher to bear the Interestingly, the vendor whose DBMS costs of reformatting. needs 12 hours to unload a 5MB database into 4) During the scheduling process, the atransportable format would meet this appraisal should include atechnical analysis specification. of any software or hardware dependency. My discussion of the evolution of appraisal However, the full extent of these technical of electronic records (vol. 1#3) elicited problems m6'y' not be apparent until the time considerable comment. The review alluded to of scheduled transfer. If significant a meeting on May 27, 1983. of a National technical problems become evident at the Archives Appraisal and Disposition Task time of accessioning, these problems can Force. At the conclusion of lengthy warrant the destructlon of information discussions, aconsensus developed around which was earlier appraised as permanent. four positions. Their articulation was not formally incorporated Into the final report T.E.B. of the Tosk Force, but Ithink these statements may still be basically valid and deserve to be disseminated and discussed. I EDITORIAL COMMENTARY welcome your reactions or comments. 1) The disposal provisions for machine­ Irecently encountered acase of the readable administrative or housekeeping consequences of the result of follOWing Tom's records may not always be the same as those recommendation #3. University of which have been developed for textual Massachusetts project archivist Judy records. The long- range strategy is to Hensche is looking for firms which can incorporate appropriate disposal provisions reconvert a massive collection of microform into the General Records Schedules which representing one million captured Vietnam now concern administrative records in only war documents whose index entries were textutll form. Until this is achieved, the (once) in machine readable format. and are short term solution mw. be to state within now recorded os microfilm images of those General Records SChedules that the "Flexowriter" punched cards.

Winter, 1987 Copyright by Archives & Museum Informatics 71 I I

Documentation of the Intelllgence Data SM-CGAP PLANNING GROUP REPORT Handl1ng Systems (IDHS) of the U.S.Military Assistance Command in Vietnam, obtained In February 1987, the SM Commltteeon from the Notional Archives, is helpful as Goals and Priorities created a planning group background to understand the system which to develop adetailed action agenda to produced a huge collection of microforms at implement goals ClOd objectiVes relating to ,the University of M~husetts Boston automated records and techniques outlined in Harbor campus, but it ooes 11ttle to help Planning for the Archival Profession Juay. and her researchers to access the (commonly known as the GAP Report). The documents on the f11ms. What can be gleaned planning group met in Ann Arbor on July 28, from the ~umentat1on is that the 1987. Its report consists of two sections ­ Department of Defense used an "Automated" Automated Techniques and Automated Records. Document Storage and Retrieval System Reprinted below are "Proposed AcUvities" manufactured by FMA Inc. and sold statements from the planning group report. commercially during the 1960's as the The full report includes ack:Iitional text on FlleSearch (Flexowriter) system, to index Purpose and Significance, Potential Actors captured oocuments. and Possible Resources, and aTimetableJor Index terms were highly COOed to seve each proposed actiVity. space, and the cooes were keypunched on the Flexowrlter. The document, with its AUTOMATED RECORDS: associated index worksheet and machine­ A PROPOSED ACTION AGENDA punched card, were f11med together. The Fl1esearch system used a unique, seven bit, I. "Establish aclearinghouse or means of binary code, which complicates matters exchanging information about activities in somewhat, but the system as awhole is the area of electronic recordskeeping reasonably well ~umented. practices and techniques. The clearinghouse Unfortunately, the cards were not read onto could contain information about existing or magnetic media and into standard data planned activities in this field, examples of representation cooes, as these new and non­ general records schedules, examples of proprietary appr08Ches supersecEd the legislation, information about access and Flexowriter system. As aconsequence, the privacy issues, and technical material about costs of re-indexing the entire collection, or preservation, data transfer, and new of converting the data from the micrographic technologies. " image of the Flexowriter cards wm fall to II. "A regUlar series of articles should be future researchers. written and publ1shed in the SAA Newsletter It is intriguing to consider whether this on various aspects of automated records and represents good social po11cy, is cost techniques. The articles would summarize effective and is fair. Ifind myself thinking activities and projects that are being that even though tt is obviously more costly undertaken in these two areas by the Society to convert the data now than it would have itself, by information professionals in other been in the early 1970's, it seems on first associations and organizations, as well as examination to be rEmSOnable social policy within archival institutions throughout since it places the burden on those who want canada and the United States. These articles to use the materials. But Iwould not be should be mene available to other w111ing to accept the consequences if the data publlcations, such as the newsletters of were U.S. voting returns or information regIonal archival associations and specialized "essent1al" to a knowl8O;Jable c1t1zenry. I newsletters, e.g. Archivallnformat1cs." would welcome your comments and other III. "A research agenda should be developed examples. Juay. would appreciate offers (or for the archival profession in the area of bids) to convert the data. D.B. automated records and techniques. The

72 Archival Informatics Newsletter vol. 1#4 research agenda should identify activities and revised in their particular area of interest projects that are currently being undertaken and expertise. Such groups could then in these two areas by the Society and its integrate such publications into their own members, by other associations and plans and activities, suggest additional organizations, as well as by archival materials (such as workbooks) that might institutions throughout canooa and the United supplement the publications, and suggest States. The research agenda should also distribution mechanisms and avenues for identify the outstanding issues that have yet maximum impact." to be addressed by the archival profession. IV. "Develop teaching techniques to The agenda should describe in considerable disseminate basic information to beginning detail the various actors or players who archivists. " should be involved in these activities, the V. "Propose amechanism to influence NARA resource implications, and propose anumber and LC to provide leadership on automated of options for coordinating the work of the techniques. " various actors, thereby ensuring that long­ VI. "Develop functional specifications for term objectives are attained." local archives and manuscript control IV. "Develop aworkshop on the life-cycle systems." management of automated records. " VII. "A. Develop amechanism to evaluate V. "Workshops for archival educators to archival descriptive standards. introduce or improve knowledge of concepts B. Undertake retrieval studies in and methods of teaching about automated automated systems usi ng the MARC AMC records and techniques in basic archival format to develop better descriptive education courses," standards. VI. "Influence the interim certification C. Work towrds clarifying definitions board to ensure that there will be adequate for MARC AMC f1elds 655 (Genre/form coverage of machine-readable records and terms) and 755 (Physical characteristics techniques included in the certification access). Develop guidelines for applying process." terms. VII, "Identify college and university D.Appoint aLibrary of Congress organizations and associations whose Subject Headings Task Force to suggest members create and maintain records. changes and modifications of LCSH to the Develop guidelines with these organizations Library of Congress. Develop gUidelines for for preserving archival documentation, archival use. especially machine-readable records of the E. Produce afrequency distribution colleges and universities." list of terms used in the various access points in the MARC AMC 6xx fields in the AUTOMATED TECHNIQUES: RLI NAMC file. A PROPOSED ACTION AGENDA F. Revise and expand Archives, Personal Papers, and Manuscripts. I. "Continue to offer workshops developed by the Automated Records and Techniques Task Planning group participants were: Larry Force and develop additional advanced topic Dowler (Harvard), Fynette Eaton (NARA), workshops, " Tim Ericson (SM), carolyn Gada (ICPSR), II. "Develop aworkshop for planning for Margaret Hedstrom (N,Y. State Archives), automated techniques for archival David de Lorenzo (Gallaudet), Harold Naugler administrators. " (Nat. Archives of canada). Donn Neal (SM), III. "Develop mechanisms to ensure that the Lee Stout (Penn. State), Richard Szary l:Ippropril:lte SAA groups (committees, task. (Smithsonian), Bill Wallach (Michigan), forces, etc,) are consulted about official SM and Lisa Weber (SM) publications that are to be developed or

Winter, 1987 Copyright by Archives &Museum Informatics 73 users groups, attending to the legal issues in CONFERENCES software acquisition, managing risks in implementations. MUSEUM COMPUTER NETWORK The keynote address by David Bearman of Archives & Museum Informatics argued for The Annual Meeting of the Museum increased attention to development of Computer Network held OCtober 13-14, in standards for information exchange and cambridge, Massachusetts, witnessed a introouced the concept of "open" or smorgasbord of papers in what seemed like a permissive standards which codify practices single unending session, distinguished by the but do not require uniformity between in­ numerous important projects discussed but house systems which use different subsets of hampered by lack of formal discussion ­ the standard at their option. periods. Exhibitors and project . Two research reports stand out for me. demonstrations competed with the formal· Dr. Jane Stone (Dept. of Computer SCience. session and with numerous ~ discussion Montana State University) reported on a and special interest groups, and were, as a simple technique she deVeloped to measure consequence, much less successful. items depicted in an archeological imagebase Abstracts of papers presented at the meeting and, potentially, to scale digitized images so will be published in the next issue of that they could be displayed together. She SPECTRA, and some of the full papers will locates two points on each images for which appear in subsequent issues (available as a the distance is known (a penei 1or shoe or membership benefit for $25 from MCN, Box person or object of k.nown length) and the Ill, East Winthrop ME, 04343). machine then computes the scale and Participants heard reports on many major calculates the dimensions of other objects 1n international projects, including papers by the image. To date the technique has been D. Andrew Roberts on the Museum applied only to videodisc images, but Dr, Documentation Associatfon of the U.K., Toni Stone intends to extend it to digitized data. Peterson on the Art 8< Architecture Dr. D.D.Hilke reported on audience Thesaurus, Jim Blackaby on the revision of research she conducted in conjunction wHh Nomenclature and on the "Common Databases the SITES, "laser at 25" exhibit. Concern for History Museums" component of the has often been expressed that interactive AASlH Common Agenda project, Marilyn components in exhibits might compete with Schmitt on the Getty/ADAG and other other means of presentation, so Dr. Hnl

Winter. 1987 Copyright by Archives & Museum Informatics 74 computer was on, they spent more time in Dr. Adams tried to place the "information the exhibit, but no less time on the rest of society" in an evolutionary, rather than the exhibit just because the computer was revolutionary, relation to prior eras in his available, and they engaged in 24~ more philosophical opening remarks. calling on "information seeking" behaviors when the historical evidence, he OOwnp1ayed the computer was on. ability of technology to increase knowleO;je, Two developments within the MCN itself noting that knowledge, unlike information, is should also be noted. The MCN extended a personal acquisition, requiring a human voting privileges to all individual members intermediary, but his broad perspective did of the organization (voting was prevlosly not damper the enthusiasm for technological limited to organizations) and it elected to fixes and razmataz shown by others. hold afull day "vendor/user" forum in Evidence of recent information management conj unction with its meeting in Los Angeles, evolution was reported by James McKinney OCtober 26-28, 1987. The forum will and his colleagues at Harvard Business provide an opportunity for venoors to School. McKinney et. al. have been stUdying discuss strategic decisions in their software internal corporate communications in great and contrast their products and for users to detail for the past six years. Their intensive compare systems. Acall for papers will be observations of ml communications between issued in January; preliminary proposals employees of several firms on at least two for papers or sessions should submltted by occasions of at least five months duration March 1, to David Bearman, 1988 MCN each I provided a basis for an exceptionally Program Coordinator, Archives & Museum rich analysis of the impact of electronic mall Informatics, 5600 Northumberland St., systems on bus1nesses. Their findings that Pittsburgh, PA 15217. E-mail decreases the percentfq3 of time managers spend in meetings and increases AMERICAN SOCIETY FOR the amount of face to face contact are less INFORMATION SCIENCE interesting in themselves than the fact that use of electronic mail works. It work.s by ASIS held its 50th annual meeting in improving preparation for meetings (the Boston, the week before MCN. The event was largest reason for the decrease in t1me spent staged in agrand fashion, with plenary in meetings), improving service response aaJresses daily from astellar cast including time, increasing the depth of employee Robert McCormick Adams (secretary of the understanding of manfq3ment priorities, and Smithsonian), Joseph Duncan (Chief shorten'lng development time 1n projects (1n Statistician, Dun & Bradstreet), part by reducing the number of unanticipated Congressman George Brown, Arthur Miller bugs). For electronic mail to succeed, (Harvard Law SChool) and Jacques Vallee however, McKinney and his colleagues found (EUROLINK International). seventy other that organ1zations must change the way 1n sessions were held in concurrent slots, so which they work - bUilding informal this report wm be impressionistic and networKS in place of formal rev1ews and reflect my choice of sessions and the budget cycles and by managing the competing demands of exhibits and my communIcatIons process 1tself. In the committee actiVity as in-coming Chairman of question period, McKinney acknowledged that the Arts & Humanities Specia'i Interest Group one area of corporate structure that has not (SIG-AH). Ileft the meeting impressed by accommodated itself to the electronic ideas Iheard articulated around two broad environment is records manfq3ment and that themes: the societal and oraanizational mechanisms for ensuring corporate impeot of information systems end empirical accountab1lity have yet to emerge 1n the stUdies Of retrieval methoos and indexing firms he has studied. theories. Jamie Grodsky of the U.S. Congress, Office of Technology Assessment, reported on another aspect of corporate behavior which

Winter, 1987 Copyright by Archives & Museum Informatics 75 has yet to adjust: dissemination of press weighted terms and vectorized searching information and information to the public over Boolean Query. through the Freedom of Information Act. Two papers of special interest to archives Grodsky pointed to the plethora of and museums, both dealing with new information outlets maintained by agency technologies are printed in full in the press offices, noted the resulting reduction Procedings. Howard Besser presented the of access to information, and hinted at the experiments in digital imaging at the likelihood of a federal wire service in the University Museum in Berkeley in an near future. She then turned to the FOIA and exceptionally rich paper I and George Thoma examined the problems which the law discussed design issues involved in the (enacted when there were fewer than 3000 implementation of cost-effective image based computers in the entire Federal government) document management systems at the and the case law which has grown up around National Library of Medicine's Lister Hill FOIA litigation, have created for access to Research Center. Thoma documents second information in electronic formats. The case by second that optical scanning throughput, law for records in traditional formats has the human labor component of imaging, will clear ly established that agencies are continue to be the critical variable in required to release "records in being", not to capturing texts and images using digital create new records, but requests for technologies, but that many minor electronic information pertain to records engineering adjustments can be made which, which are created in response to the request in the aggregate, significantly affect costs, and are not "in being". The law has established that records must be retrievable ARCHIVES AND AUTHORITY through a "reasonable" search, which case CONTROL law has defined in afairly constrained way; with electronic systems, it may not be unreasonab le to demand more extensive Aseminar on Archives &Authority Control searches. Finally, because the law limits was sponsored by the SIBIS-Archives Users release of proprietary materials, the Group and the Office of Information Resource software in which data resides is not Management (OIRM) at the Smithsonian on releasable, and this may effectively make OCtober 27. Full procedings of the many data sets unavailable through FOIA. conference, including discussions, will be OTA is studying these and other information published as an Archival Informatics dissemination issues and invites comment. Technjcal Report in the summer of 1988. Alarge number of papers grapp led with Over a hundred Smithsonian archivists, empirical analysis of the relationship curators, information systems support staff between indexing methods, retrieval success, and guests attended the sem inar to hear and other factors. Fortunately, many of Jack ie Dooley (U.C.San Diego) introduce the these papers are published in full 'in the concept of authority control, Tom Garnett (SI Conference Proceedings, available from Libraries) discuss the experience of a Learned Information (members $32, others library implementing authority control, $40). Marion Matters (M innesota Hist. Soc.) In two valuable papers, Marcia Bates contrast it with the experience of astate reintroduced the content of the databases as a archive, Lisa Weber (SAA) datai J the state of too often 'ignored variable in users studies authority control as an issue for the archival and Carolyn Eastman examined the role profession and Rich Szary (01 RM) assess the negation plays in Queries. In achallenging relationship between technical capabilities session on automatic indexing, Susan of the software system and needs of the Humpreys reported further on the frame­ archives users community. David Bearman based indexing methods in use at NLM and (Archives &Museum Informatics, led a Gerald Salton expanded on his arguments for discussion of the speakers and the audience.

76 Archival Informatics Newsletter vo1.1 #4 CONFERENCE CALENDAR 1988

February 4-11, Art Llbrar1es Society of July 20-23, Nat10nal Associat1on of North Amerlca, Annual Conference, Dallas TX Government Archives and Records Managers. [ARUS/NA,3900 Tlmrod St.,Tucson.1V. Annapolls NO [NAGARA. New York State· 8571 tl Archives,Room l0A75. CUltural Educat10n Bldg. Albany, NY 12230] March 21-24. RIAO-88 "user-Or1ented Content-Based Text and Image Hand11ng". August 22-26, "Museum Arch1ves", cambrltt;Je. MA [M.I.T.• COnference serv1ces. Wash1ngton. DC, [5mtthsonlan Institutlon. Bldg 7 Rm. 111 •cambr1dge, MA. 02139] Office of Museum Programs. Wash1ngtonDC 20560] Aprl1 13-17. Soc1ety of Architectural H1stor1ans Annual meetlng ChlC8/1) [~H. september 14-17, Amerlcan Assoolat10n for 1232 Plne St. Phll~lphla PA 19107] State and Local H1story, Annual meetlng [MSLH, 172 Second Ave. North, SUIte 102, Aprl1 21-22. Nott1ngham I England Nashv111e, TN 37201] . "Computer 1n Museums seminar. sponsored jointly by the Museum Documentation Assoc. september 21-24, 5econdAnnual Museum amrthe MuseumsAssoo.• [MDA, 347 Cherry Documentat1on Ass001aUon Conference, Hinton Rd.. cambridge. CB 1-4DH. UK] "Term1nology for Museum Dreumentation"; C8mbr1d;Je (UK); preceded by asttKtf tour May 15-18. American Society for for the interested [MDA, 347 Cherry Hinton Information SCience-Mid-Year meet1ng on Rd., cambr1dge CB 1-4DH, UK] ArUficialintell1gence,Ann Arbor. MI. lAStS. 1424 16th St. Nw, Wasll1ngton. DC 20036] september 29-~tober 2. SOC1ety of Amer1can Archivists Annual COnference, May 26-29, International Assoc1at10n of Atlanta Georgla [~, 600 S. Federal St., Social SCience Informatton Systems & SU1te 504, ChiC8':J), IL 60605] TechnolD;JY (IASSIST-88), Washington. DC [Pat Doyle, Mathemat1ca Pol1cy Research ~tober 2-6. 2nd Annual Library & Inc. , 600 Maryland Ave. ,SW •SUlte SSO, Informat10n TechnolD;JY Assoclat1on Washington. DC 20024] COnference, Boston MA [LI TA. 50 East Huron St., Ch1C8/1), Il 60611] June 3-7I American Association of Museums Annual COnference, P1ttsburgh, PA [MM, ~tober 23-27, AmerIcan SOCIety for 1225 Eye St. ,NW, SuUe 200, Washington Informatton SCIence. Annual Conference, DC. 20005] Atlanta, GA [ASIS, 1424 16th St., NW, Washington, DC 20036] June 6-1 0, Associatton of canadian Archivist~ Annual meeting, Windsor, ON ~tober 26-28. Museum Computer Network, [ACA, P.O;Box 2596. Station D, ottowa. Annual Conference, los Angeles [MCN. K1P-5W6, CANADA] . P.O.Box 111, East Winthrop,ME 04343] Conference 1ncludes Venoor/User forum on June 12-16, Nebraska Videodisc Design­ ~tober 26. To partiCIpate as avenoor or Workshop. 8as1c workShop dates for 1988 propose sess10ns or speaKers, wrtte DavId are March 20-24;May 22-26; July 17-21. Bearman, 5600 Northumberland St., [ 1800 ~rth 33rd. St.. L1ncoln,NE 68583) P1ttsburgh ,PA 15217 .

Winter, 1987 Copyright by Archives & Museum Informatics 77 IN-BOX

REPORTS

ACommon Agenda for History Museums: Report of the First National Conference on Conference Proceedings, February 19-20, Issues Concerning ComputeriZed Public 1987; edited by Lonn Taylor, Nashville,TN, Records, Public Records Division, American Association for State and Local Massachusetts Secretary of State, 1987,2 History, 1987, 53p. vols. This volume contains several thoughtful The conference reports on the conclusions articles which are important in their own of four working groups consisting of state right, not just as acall to action, James freedom of information act officials, Blackaby's categorization of the types of commercial and public advocacy information found in history museums is of representatives and legal and political particular interest. The Common Agenda figures, which addressed acommon set of project has recently received additional issues. The consensus report seems to open funding to enable it to pursue the criteria for access by concluding that release development of common data standards. of automated records under FOIA constitutes release of "pre-existing" records, not Government Information: An Endangered creation of new records, but to restrict Resource of the Electronic Age, papers criteria by agreeing that access to certain presented at the first Special Libraries kinds of records, such as property Association, State-of-the-Art institute, assessments or professional licenses, which October 19-22, 1986, Washington DC, SLA, involved individual citizens could be denied. 1986, 277pp. $21. 75 The details of the working group discussions Papers in the on-going debate about the are informative for anyone concerned with appropriate role of government and the access issues. Afollow-on conference is private sector 'in dissemination of public scheduled for 1988. information. Virginia Museum of Fine Arts Information Program Reporting Guidelines for Systems Framework, Richmond VA, VMFA Government Records Programs, National and Ottowa, Ontario, CHIN I 1987 l unp. Association of Government Archives & This gigantic report on the results of a Records Administrators and the Council of collaborative effort by the Canadian Heritage State Governments, Sept. 1987, 26pp.[ CSG, Information Network and the staff of the Iron Works Pike, P.D.Box 11910, VirginIa Museum of FIne Arts to apply CHIN's Lexington. KY 40578) "Corporate Information Systems This protocol for statistical data collection Framework" (a methodology akin to IBM's represents the culmination of years of Business Systems Planning) is worth careful efforts by NAGARA to arrive at measures of study. Its scope is all the informatlon activity which could represent cumulatively systems of amoderate size museum ( 106 the impact of public records programs. One separate data entities, 23 major activity hopes It w11l be adopted and tested widely and groupings each comprised of numerous sub­ evolve over time so that reliable data can be activities), and its objective is total compared and programs can be assessed, Very intellectual integration. It is the best useful definitions and measurements are museum systems foundation document I know proposed which other repositories will want of, including those on which I worked while to study. at the Smithsonian.

Winter, 1987 Copyright by Archives & Museum Informatics 78 NEWSLETTERS ART ICLES & BOOKS

CD Datfl Report (ISSN-8755-5727), Barnett, Patricia &Lucker, Amy E., Langley Publications Inc. 1350 Beverly Rd., procedural Guide to Automating an Art Suite 115-324, McLean, Va 22101 , $225 Library, occasional Papers #7, Tucson, Al. p.a. Art Libraries of North America. 1987 40p. Editor Linda Helgerson has oone asuperb $15.00 job of following this emerging market and The title is misleading, but the papers in reporting on it monthly from all angles; this volume. Qn retrospect1ve conversion, hardware, software, users, products. Issues online catalogs, authority control and the are lengthy ( 40+ p,). well-written and relationship between object and timely. bibliographic information systems. are useful additions tQ the literature and the Information Retrieval & Library AutQmatiQn "DirectQry Qf PrQjects and Systems" (ISSN-0020-0220). LQmQnd Publ1cat1Qns compiled by the Clearinghouse on Art Inc., P.O.BQX 88, Mt. Airy. MD 21771. Documentation and ComputerizatiQn will $66.p.a.. Commencing VQ1.24 in 1988. provide at least one new, and essential, Its edited press releases and publication contact for anyone. notices include some items which are news tQ me every month. Bearman. David, Towards National Informat1on Systems for Archives and Registrar, Newsletter Qf the Registrtar's Manyscript Repositories: The NatiQnal Committee of the American AssoclatlQn of Information Systems Task Force (NISTn Museums ($10 plus AAM membership). Papers. Chicago. SM. 1987, $10, 119pp. The Fall 1987 1ssue of Registrar 1s

Research L'ibraries Group News ( ISSN- Crawford. Walt; Patron Access: Issues for o196-173X) is afree. occasional, house Online Catalogs, Boston, G.K.Hall, 1987 Qrgan which often has articles Qf interest to 250p. + index the archives and museum community. The This small book contains so much 5eptember 1987 issue (#14) contains a uncommon sense about online catalogs that it report on the King Papers Project and its use should be required reading for anyone of MARC-AMC in apr'lmary publ1catiQn. designing areference database. From the historical editing, project. first dictum, that "online catalogs should serve patrons at least as well as card catalcgs SCOPE: Humanities Computing Update ( ISSN­ 00" aseries of assertions are made (and 0735-8296), Paradigm Press Inc., Osprey, amplified by examples) which are, Flo 34229-1057, bi-mQnthly, $52.p.a. collectively, the best guidance yet published. SCOPE contains awealth of bibliographic, I am part1cularly impressed by the references, references to hardware and implications Crawford draws from, his software reViews. and software nQtices in assertions that the online catalog is "not a addition to an occasional short article. thing, but aperspective" ( ie.• aview of a ~electea past Items from SCOPE software, database), and that it "should not be a courseware reviews and software reviews computerized card catalog" but rather a are issued on floppy, as DiSCSCOPE ($5). source of dynamic "status" information.

Winter, 1987 Copyright by Archives &Museum Informatics 79 Djrectory of Information Management valuable statement of the wtrYs in which such Software for Libraries. Information Centers. subject term authorities can relate to Record Centers. 1987-88 Edition. Ed. by traditional library practice. John Kazlauskas and Pamela Cibbarelli, Studio City, CA Pacific Information, 1987, Overmire, Rozell J Records Management in 300DD •. $49. Dbk. Museum Travelling Exhibits: Planning a Two page, systematic descriptions of 118 Computerized System, Masters Thesis, July packages, alist of non-respondents, and 1987 J John F. Kennedy University, san indexes by vendor. hardware, modules and Francisco,CA, 119p. & appendices product name. The information should be Asurvey of computerization projects in adequate to decide whether to evaluate a the travelling exhibits field; rich in deta11 product further, but more comparative from which to develop requirements but tables would have been useful. While the short on advice, design considerations. or scope is appropriate for the repositories in analysis of the sources of successes and the title, archives and museums w1ll find failures in the field. some products of interest here as well. Parker, J.J.N., "The Hull Doomsday Project Dryden. Jean. "Subject Headings: The PAASH Database". Humanjstjske Data, 2-87, pA­ experience" ,Archivaria, 24, Summer 22 1987, p. 173-180 An introduction to the work surrounding The development of the Provincial Archives the analysis of the Doomsday book as a of Alberta SUbject Headings is the most machine- readab Ie data file. complete account we have of the challenges to archivists of modifying lCSH and the pitfalls Rush, James E. & Tannehi II, Robert S. Jr. , and potential benefits of such an unrertaking. "Bibliographic Data Elements for Computer The article, and the headings themselves, Programs" , Library Hi Tech, # 14, p.79-93 must be required reading for archival I failed to cite this essential article in a descriptive catalogers. previous discussion of cataloging of software, and hasten to correct my oversight. Dunlap, Ellen S. & Reed, Kathleen, "BorrOWing of Special Collections Material Sh1llingsburg, Peter L.; SCholarly Editing in for Exhibition: ADraft", Rare Book and the Computer w: Theory and Practice, Manuscript librarlanship. v.2# 1Spring Univ.of Georgia Press. 1986. $11.95 1987 p.27-37 The real of building archives and museum This draft guideline. prepared for the Rare data and image bases is to use them to Book and Manuscript Division of ALA has interpret our past. Sh1l1ingsburg combines uses for all archives and museums, and is practical advise with discussion of the being proposed as the basis for astandards premises of documentary editing and development process. potentials for transforming texts by use of electronic techniques. LC Thesaurus for Graphic Materials: Topjcal Terms for Subject Access, Woshington DC, Sturges, Poul, "Policies ond criterio for the Prints & Photographs Division. library of archiving of electronic publishing", Journal Congress. 1987,591 pp. $30, from CDS of Librarianshlp, vol. 19#3, July 1987 Subject access to image material has long p.152-172 been aproblem for libraries, archives and The author takes abroad view of the museums. This structured thesaurus, practices of librarians, archivists, data containing ahuge hierarchy of terminology, archives, media centers and others and w1ll be of great assistance to anyone makes recommendations, preViously providing such access. Jackie Dooley's advanced in his report to the British National introduction to the volume stands alone as a Bibliography Research Fund.

80 Archival Informatics Newsletter vo1.! #4 "Lasers in Graphics Arts: White Paper" EPHEMERA avallable free from Compugraphics Corp., Marketting Communications, 200 Annual Revjew of OCLC Research. Jyly Ballardvale St., Wilmington, MA 01887, is 0 1986-June 1987, DUbl1n,OH, DeLC, 1987, technical introouction to lasers in Image 53pp. setting (type setting), scanning and printing. OCLC continues to conduct exciting basic It is stra1ghtforward and non-commercial. and appl1ed studies of network1ng, 1nterfaces, automat1c indexing and Optical Discs for Storg and Access in ARL cataloging, image and information capture libraries, SPEC Kit 133, April 1987, and retrieval, speech recognition, new media [Systems Procedures Exchange Genter, Office and ahost of other issues of concern to of Manaoement Stud1es, Association of cultural repositories. Their annual report Research Libraries, 1527 New Hampshire is clear, informative and includes references Ave., NW, Washington, DC 20036] contains to more detalled pUbl1cations. optical systems planning reports from MIT and the Smlthsonian, proposals for projects The Beckman Genter for the History of at Kent State, MIT, U. Michigan, Syracuse Chemistry: Five Year Report. 1982-87, and Vanderbilt, progress reports from LC Philadelphia, PA, The Beckman Genter, and MIT, and evaluations from Brown, 1987,36pp. U.lllinois and Vanderbllt, mostly from late Neither museum nor archive, this 1985 and 1986. Useful for anyone pursuing disciplinary study center sponsored by the similar ventures. American Chemical Society and the American Institute of Chemical Engineers ot the Resources for the History of comput1ng: A University of Pennsylvan1a is engaged in Gy1de to U,S. ond C6nOOion Records, is both enterprises, and 1n informat10n available from the Charles Babbage Institute, exchange about the h1story of chemistry. 103 Walter Library, Univers1ty of Minnesota, Minneapolis, MN 55455, for $9. Guide to fynd Master fund Ra1s1ng Software, It contains apprOXimately 350 entries. 1987 Edit10n, Master Software Corporation, 8604 A111sonv1lle Rd. , Suite 309, Trinet and the 1987 SIC Revision is free Indianapol1s,lN 46250, 85pp. from Trinet Inc., 9 C6mpus Dr., Parsippany, Although this booklet specifically sells NJ 07054. It contains an overv1ew of the Fund Master software, its discuss10n of the Standard Industrial Classif1cation system, its issues involved in fund raising is an h1story and structure, and the 1987 excellent analys1s Of funct10nal requ1rements revisions to the SIC, with only amodest plug for this actiVity and a~ place to begin for Tr1net. defining local needs and assessing local systems. Visual Imp Transmission: An examination of electronic (ill1yery of visual images and Information Packet on Library Technical text from the l1brary to the ~mic Standards, ALA, Library & Informat10n community, Final Report to the Council on Technology Association, 1987, $3.50 Library Resources, Merrill W. Smith, L1TA has collected into apocket folder Patrict< A. Purcell and Christopher copies of ANSI, NISO, SISAC, BISAC, ALA ond P.Thorman, MIT, Rotch Visual Collection, L1TA publications brochures on standards. sept. 1986 26pp. Whether this is usefu1depend5 on whether Report on experiments in image you have picked them up yourself at meetings transm1sslon aM avarlety of real and or went to save the effort of writing these hypothet1cal configurations to support such organizations. efforts.

Winter, 1987 Copyright by Archives &Museum Informatics 81 I PROJECTS & PROPOSALS SCholarly Commun1cat1ons Off1ce The ACLS Newsletter, Summer 1987, Getty Conservation Information announced that American Council of Learned Network Societies is closing its office of SCholarly The J.Paul Getty Trust has announced the Communication in Washington DC. The Office rJdvent of the Conservotion Informotion published SCholarly Communications and was Network, the r.esult of more than two years a useful switching circuit. It, and Director of planning. Its three online databases Herbert Morton, will be missed. Meanwhile, feature access to technical literature, ACLS has just published its first "Occasion6l conservation materials and products & paper", "A Life of Learning" by carl E. suppliers, as well as an electronic mail SChorske, the delightful and fascinating text system enabling conservators to consult each of his Charles Homer Haskins Lecture on other. The bibliographic database w1l1 April 23, 1987. Contactthe ACLS at 228 provide access to almost 100,000 citations East 45th St., New York, NY 10017-3398. inclUding ephemeral technical reports. The materials database provides access to 1,000 Canadian Strategic Planning properties of materials relevant to The Association of canadian Archivists conservotion practice. Network access (ACA) has established a Strategic Planning (including one account number, one E-mail Committee charged with submitting a report box, and a full set of oocumentation) is $50, for consideration at the 1988 meeting in and includes access to a help-line and the Windsor, Ontario. The Committee has "Network News", a bi-monthly report on its circulated aQuestionnaire to ACA services. Users pay $40 per hour, prorated, membership. Asecond Questionnaire, on the for access including communications charges. potential role of the Bureau of canadian E-mail users pay $70 per 1Kbytes sent or Archivists, is being distributed by the ACA received. For subscriptions, contact, User and by the Association 00s Archivistes du Services, Conservation Information Quebec.. Contact Colleen Dempsey, President Network, 450-3 61encoeAve., Marina Del and Committee Chair, ACA, P.O.Box 2596, Ray, CA 90292 Station 0, Ottowa, Ontario, K1P-5W6, Canada U.S. Government Information Inventory Canadian Archival Certification The Office of Management and BUdget The Bulletin of the ACA (Sept. 1987) F~ral (OMB) Bulletjn 87-14 requires contains the text of adiscussion paper on Agencies to inventory their information certification drafted for the ACA. U.S. dissemination products and services. NTIS is archivists will be interested in its to collate these submissions into a similarity to (and difference from) their government wide database available for sale. own recently adopted plan. OMB intends to revise Circular A-130 to incorporate this as a permanent British Archival Descr1pUon requirement. The Arch1val Description Project at the University of liverpool, which produced the Image Management Needs Assessment Manual Qf Archiyal Descrjotion (MAD) in New Jersey is conducting a Statewide image 1986 has been awarded funds by the British processing survey to identify requirements l1brary R&D Department to revise the rules of administrators for traditional microform, tQ include in them description of such special unitized microform, CAR systems. electronic documentary formats as title deeds, imaging systems, and COM. Contact ceasar cartographic records, photographs, lacovone, Director, Div. of Archives & architectural and planning archives, sound Records Mgmt., CN 307, 2300 Stuyvesant archives, and moving image and machine­ Ave. ,Trenton, NJ 08625 readable mater1als. It 1s also charged With

Winter, 1987 Copyright by Archives & Museum Informatics 82 testing the suitability of MARC AMC in Geological Survey is disseminating earth conjunction with "MAD2" and with exploring sciences data, and is readying adistribution the implications at' optical digital media. of Global Digital Seismic Network data from The project team 1s arranging the National EarthQuake Information Center. experiments, demonstrations and discussions The Census Bureau is exploring distribution with as many UK colleagues as poss'ible and is of demographic data on cd's, while NASA's desirous of communicating with others who Vayaqer Planetary Data System and the have examples of cataloging instructions, in­ National Oceanographic and Atmospheric house rules for description. or proposed Administration's database of solar flare standards. Contact Margaret Proctor. actiVity are alr~ available in this format.

Fieldwork Coordinator I Archival Description Project, Archives Unit, University of Consortium to Develop Nineteenth Liverpool, Liverpool, L69- 3BX, England Century Painting Image Base Six major American art museums have International Museum Database received funding from the Pew Charitable Survey trusts to develop avideodisc of paintings and The International Councll of Museums. weorks on paper created between 1860 and International Documentation Committee, is 1914 by European and American artists conducting asurvey of museum databases. To which they collectively hold. The Museum of be included or see results, Contact Mary Mcoorn Art (NY), the Metrpolitan Museum Case, Director. Office of the Registrar. of Art ( NY). the Brooklyn Museum. the Art Smithsonian Institution, Washington, DC INstitute of ChicagJ, the Museum of Fine Arts 20560 (Boston) and the Philadelphia Museum of Art are cooperating in aproject aimed at Smithsonian Institution Information evaluating the potential of videodiscs for Architecture public education, database searching and An executive summary of the Information authoring. and at building afoundation for artchitecture project at the Smithsonian was future collaborative efforts. issued in October 1987 by the Office of Information Resource Management. While Automatton Surveys the project has adistance to travel, its Lisa Weber reports on the results of the success to date is encouraging. For SAA automation survey in the ~ information about the museum collections Newsletter. November 1987. Rebecca Buck architecture. Contact, Jane Sledge, Office of reports on the results of asurvey of Information Resource Management. collections management software vendors in Smithsonian Institution, Washington DC the fall issue of Registrar. 20560 Smithsontan CD-I Publication Plans U.S.Government Databases on CD-ROM The Smithsonian Institution Press has In addition to databases such as Medlars, announced plans to publish aCompact Disc­ formally published by Government Agencies. Interactive (CD-I) version of Edwards many agencies are now using CD- ROM for ParK'S Treasures of the Smithsonjan. The "intern~" dissemination of information. The disc, created by American Interactive Media U.S. Patent and Trademark Office is Inc. in Los Angeles. is to be available early in distributing its "Classification and search 1988, Narratives about 300 of the Support Information System" (CASSIS) in Institution's most famous artifacts are to be this fashion and using it in its reference viewed under interactive user control, with rooms in Crystal City, Va. The Air Force is branching prOVided for those desiring compiling adatabase of acquisitions related greater detail. Future titles are planned in materials to publish on disc. NAVSEA is what is to be alibrary of CD-I publications looking at CD-ROM for parts lists and according to SI Press Director Felix Lowe. servicing data for the fleet. The U.S.

Winter. 1987 Copyright by Archives & Museum Informatics 83 SOFTWARE FORUM

Jon Reynolds. University Archivist at level data. The system finds Waugh in the Georgetown University (37th &. 05t.. NW. lookup file. skips it if it isn't tagged as a Washington DC 20057). writes: creator. and passes to the folder level file if it is. and prints out, or displays. the "' started using dlJ8SB II in t983. and have appropriate data if found. Since you are made qUite a bit of progress. We use rJb8se using indexed fields. this would be found IIIPlus and Clipper now. and I may get the almost instantly. and would start printing or chance to try out FoxbtJS8 Naturally. I was screen writing in about a second and a interested in your statement about db8se in haIf..... 1f you wanted a photo of Waugh. the the last newsletter. Depending on the wrry system would t~ke you to ~ photo file. ~nd so you define powerful and generalizable on. If you had asked for collection level data. archival application. you may be right. but I it would take you to that f11e. I will make the hope to prove you wrong. I am NOT making collection level file as MARC-like as I can. bets. or promising delivery dates. but I am hopeful. My present system was designed to meet my most pressing neoo. which was (is) file folder access to current accessions. It works I\CCESS!ONS FILE rr r-PIlCY_IO_fl_LE_-----i nicely. but is not space efficient. and is not a Ace. ",urmER ~ " t f f r-----1 complete system by any means. I am EXTENT f t t t f------I enclosing a ROUGH PRELIMINARY diagram SOURCE f t indi~tes f t f------I which what I think is possible NOTES t t t t '-- --J using dbase pr(XJrams compiled w11th 1max_ 128) f t t I~r------, Clipper. Each box represents a db8S8 file. ~ I SOU.,O RECORDINGS As you know. each file may have up to 128 t t t t r-----1 fields ( 1024 in Clippen. 254 characters COLLECTION/RECORD f t KEYWORD GROUP LEVEL FILE t f r-----1 per field. 4000 total per record. exclusive FILE t t jH COLLECTION I ,.;44H

Winter. 1987 Copyright by Archives & Museum Informatics 84 Now Ithink this will be clear ly powerful; actually used. Idecided to base the searches the generalizable part is more problematic. on the "BEGINS" command which uses indexes Iam confident of the searching part. Idiot to search the database, rather than read each proof data entry will be 0 little harder, but record. This will prove particularly useful not hopeless. Ithink it w1l1 prove possible in searches by hierarchical classification, in to look terms up in a thesaurus during data which Ican gather all lower offices (e.g., a entry, but we sha 11 see. search by record group "31" would yield all If I00 get it all to work, I'm not sure how departments of the College, e.g. Art, History, to go about sharing it. I'm as poor as the next Dance, etc,). By the same token, we enter archivist, but Idon't h~e time to set up a the thesaurus codes, rather than the words, real business. I w'i11 probably continue to of the SPA C&U thesaurus and can 00 give awrry printouts of the ~ for the cost of inclusive searches by simply entering the copying and mailing, but w1l1 try to make first portion of code to gather narrow terms some money on advising people how to go with broad ones. about putting it to use. If it ends up more or As atest of the database and of my less fool-proof, I mrry try to figure out how programs, I have entered some 150 records to sell it for a modest sum. and searched them successfully. Data entry P.S. Your reooers may be interested to is as rapid as Icould hope, partiCUlarly know that the SABA handscanner actually because in the case of file folder level entries worksl Iwas able to try one out for several we use adata file to read redundant but weeks, and found that it reads Courier and necessary information into the record, e.g., several other typewriter fonts Quite well, as box number, fo1oor number, collection dates, long as the page is clean and the impression locat1on. is good. It is not perfect, but it reads much At this point, Ithink I have only one more better than Itype. We didn't buy it because major task: the creation of menus. all our typed finding aids are in Bookface After all of this work, I00 think that we Academic; but for less than $600 it is will no longer process in the traditional amazing." sense. What we are doing is making use of. the inventories that offices prOVide, and entering the data directly into the database. Raimund Goer1er, University Archivist, Ohio envision that processing would no longer State University, ( 169 COnverse Hall, 2121 mean consolidating accessions from the same Tuttle Park Place, Columbus, OH 43210­ office, Weeding, and creating a new 1169) writes that he has ordered release inventory. Instead, what is likely is that the 4.0 of PROGRESS. but that using his current weeding will take place and that the records version, 3,2: in the computer will be adjusted." "... we continue to make, IUl'lnk, significant advances in our application. As of this date, Reaoors are encouraged to report on thei r the data dictionary has been created and own development efforts, comment on the programs to add records at the collect1on software designs proposed by others, or level, the box level and the item level have suooest generalized requirements that these been written and successfully tested. Most and other systems should meet. Iw1l1 recently Iwrote programs to search this continue to run aforum, so that design issues database of accessions by means of keywords, can be aired. The forum is open to software thesaurus terms (from SAA's Thesaurus for vendors who wish to explain their system COllege and University Archiyes) keywords architectures as well. I may delete parts of and thesaurus, classification, and record i.d. discussions that are too lengthy or too Printed reports consist of an alphabetical promotlonal. Ed1tor listing of all keywords actually used in the database (with provisions for see and see also) and 6 listing of thesourus terms

Winter, 1987 Copyright by Archives & Museum Informatics 85 random, access to the OED is an entirely new SOFTWARE PRODUCTS experience for dictionary-philes.

Advanced Projects International, Inc., Vernon Systems Ltd. [P.O.Box 6909, [8000 Towers Crescent Dr., Vienna, VA Auckland, New Zealand) has announced 22180], astart-up firm headed by Hernan COLLECT/ON a multi-user, PC-based Otano who lead experimental projects in application for collections management built videodisc and digital imaging at the National on the Reve/titlon DBI'1S by Cosmos Inc, that Air and Space Museum, has announced four is consistent with MDA data standards. The models of its DigitalArclJivalRetrieval first complete release was not available System ( DARS). The basic components of when I saw the system operating at the MCN these systems are image capture devices, meeting, but what I saw was extremely storage systems, and information processing impressive. software. The image capture devices range COLLECT/ON supports a rich data from flat-bed scanners for uniform paper structure with numerous data groupings and photcxJraphic oocuments and over- heoo (composed of "multi-valued" fields) that camera scanners for exceptionally large and serve as authority data. Data entry is vastly small items, to automatic microfflm readers simplified by these segments which can and video camera's for color oocuments, permit exception only data entry for much of paintings and three dimensional objects. a record. Atrue thesaurus capability, with Storage is accommodated on magnetic both the use/use for and broader/narrower erasable disks and optical WORM disks. The structures supports as many thesauri as the retrieval system is proprietary, but user wants. Indexes are constructed for 28 promises to extr~t text from images and access paths. Navigotion between any permit output to high resolution laser screens is aone step process and the user can printers. return to the place of departure with a tooole. Maximum record length is about 65K Chaclwyck- Healey Inc., [ 1101 King St. , per record and all fields can be variable in Alexandria, VA 22314], has been length and occurence. Some unfinished demonstrating a CD-ROM product that aspects, such as security control, were not heralds a new W&/ of presenting a wide demonstrated, but other features, such as variety of statistical information. reporting and query, are handled by S(JPERI1AP combines 1980 U.S. census data Revelation which has an impressive set of with maps of the U.S. to the county level to resident features. I look forward to the enable users to displ&/ relationships Spring 1988 release. between pre-aggregated datasets cartographiCtllly. Susan Severtson, ABACUS Data Group Inc. [8 12 S. W. President, expressed interest at a recent Washington St., Suite 600, Portland, OR meeting in extending the proouct to allow 97205) recently sent me documentation of local data sets to be displayed together with their Photograph Management System and the census data. Historical Society Membership System, Chadwyck-Hea1ey has also announced software prooucts aimed at the WANG-VS February 1988 availability of "The market. ABACUS also makes mailing list and Architectural Landscape of North America: accounting software aimed at historical An Image Bank on Videodisc" by Alan Gowans, societies, museums, and archives. The consisting of 28,000 color images, indexed Photograph Management System features and captioned in two separate volumes. item or lot level description and multi-layer hierarchical subject assignment. The Tri-Star PUb1i5hing [475 Virginia Dr., Membership System features member Fort Washington, PA 19034) is offering the history, gift and aff11iate society OXford English Dictionary (without membership (third party billing) and supplements) on CD-ROM. Interactive, interface to word processing.

Winter, 1987 Copyright by Archives & Museum Informatics 86 rather than usi ng standard codes. STANDARDS Presumably they wouldn't make up codes for states instead of using postal codes or for X.400/X.500 chemical elements? Perhaps they oon't know One of the silent events of 1987 was the of standards. If so, allow me to introduce the acceptance by alarge number of computer Standard Industrial Classifications (SIC) and telecommunicmions information venoors codes, asystem managed by the U.S. Feder~l of the X.400 standard which is now several Government, which relates to U.S. economlC years old. X.400 is astandard for electronic and social data the way that census tracts messages which makes possible exchange of relate to demographic data and electoral data between any systems. Acceptance was districts relate to poJit1cal systistlcs... which marked by two staged events - avendor is to say •you can't correlate anything with demonstration at the Hannover Fair in March data which has not been so coded. 1987 involving British Te1ecomm and the SIC is ahierarchical schema in which 11 Deutsche Bundespost along with many of the one digit"Divisions" (Agriculture, Mining, world's largest computer companies and a Retai 1, Services etc.) are divided into 2 digit demonstration at Telecom '87 in OCtober "Major Groups", three digit "Industry involving the Swiss PTT, the French PTT and Groups" and 4 digit "Industries. The system AT&T and many others. is simple, comprehensive and powerful. A X.400 would provide aworld-wide free gUide to the 1987 update is available electronic mail service by defining just how from TRINET Inc., 9 Campus Drive, messages would be switched and relayed on Parsippany, NJ 07054. the world's public telecomm services In England, aStandard Industrial (providing an organizational infrastructure cLassification used by the central Statistical to carry out the potential transfers ~nd Office serves the same function. It has been implement the host of local, e.g., natIOnal, incoporated into Social History & Industrial tariffs) and how the contents of an envelope Classifjcation published the University opf would be described. X.400 can support all the existing telecommunications methros ­ Sheffield in 1983 and available from the telex, facsimile, teletext, videotex and l'1useumn Documentation Association which straight data exchange and therefore has incorporated its use into the Museum Documentation System. promises to make electronic document interchange afeature of the 1990s landscape. Missing from the standard is a Background & Projections "Telecommunications and Information means oJ handling directories for remote Systems Standardization -- is America addressees - an oversight being rectified in Ready?" May 21, 1987 (87-458) written x.500, aU,S. standard being proposed to the by David Hack, Analyst in Information CCITT early in 1988. Science and Technology at the COngressional This one is very real, and almost upon us. Research Service contains auseful Virtually all the major vendors have discussion of the background to international announced support and it is un1'ike1y that any will be able to resist and survive into the standards efforts and their potential 1990s. Archivists need to consider the economIc and political implications. implications uninhibited, universal, instantaneous electronic communications CD-ROM log1cal File Format will have for work within their On OCtober 5, the InternationaI Standards Organization passed the Volume and F1le organizations and plan accordingly. Structure of CD-ROM for Information Interchange standard (ISO/DIS 9660) which Usi ng Standard Codes grew out of the "High Sierra" group I frequently find myself dismayed to discussions and had previously been adopted discover yet another project making up codes for languages, countries, or industries as ANSI X3.B 11. While access remains to be standardized. the format now seems stable.

Winter, 1987 Copyright by Archives & Museum Informatics 87 AUTOMATBD SYSTBMS FOR ARCHIVES AND MUSEUMS: ACQUISITION AND IMPLEMENTArION ISSUES

ARCHIVAL INFORMATICS TECHNICAL REPORT VoU, #4, Winter 1987

Executive Summary:

Selecting and implementing an automated system is a complex and expensive process which few archives and museum directors have experienced. This report examines the technical requirements of each of the major types of automated systems installed in archives and museums. Specific technical features of membership and development, exhibits management, collections management, accounting and personnel (including volunteer) systems are reviewed. The emphasis in all cases is on analysis of options.

The report aims to educate the professional staff of an archive or museum who possesses the expertise to define functional requirements for their information systems. but who have little or no experience defining technical and performance requirements or selecting hardware and software systems. The report is intended to be used as a workbook. It presents a step-by-step guide to writing a request-for-proposal and developing evaluation criteria with which to select the system that comes closest to meeting institutional needs. It includes checklists of features. forms which can be used in evaluating systems and vendors, and project planning guidelines for the entire system life-cycle. And it discusses the issues which a cultural repository will face in implementing automated systems.

In addition, the report identifies important implementation issues and suggests methods for staffto assess implementation impacts and to plan realistically for installation, training, operations, maintenance and replacement. While this report will not obviate the need for additional professional help, it will prepare staff to provide on-going support for the automation of archives and museum information systems.

Available in January, 1988. Archives &. Museum Informatics, 5600 Northumberland St.) Pittsburgh, PA 15217. Separate copies $45 pre-paid, includes handling, or by subscription to Archival Informatics Newsletter &. Technical Report ($160 p.a. includes all reports for the year plus the Newsletter' $180 O.a. foreian subscriotions includes airmail.).

Winter, 1987 Copyright by Archives &Museum Informatics 88