Page images
PDF
EPUB

A set of computer programs written in the FORTRAN V Level I programming language has been developed which allows the indexing, classifying, retrieval, and editing of citations of research papers through the use of a computer terminal in a research laboratory or office. Complete listings of the application programs may be found in Appendix D.

The system is designed to help user's of medium and small computer systems organize, edit, file, and retrieve copies of research papers. A major advantage of implementing a system for personal copies of research papers, is that it allows the user to consider the references to form a research data base, and allows the user to search through them using new search strategies.

The current version of this data base system does not provide publication abstract capabilities. Enhancements to the system are planned during fiscal year 1981 and provisions for abstracting will be included.

This report describes the overall systems, its operation, and the computer program. It allows a user with some familiarity with computers to use or duplicate this system. The software system described enables the user to access the entire data base or subsets of the data base.

A Perkin-Elmer1 Model 7/32 minicomputer was used in the development and implementation of the system along with its associated peripherals and software routines. Other minicomputer systems can be used to implement the data base system, although minor adjustments in the programs may be

necessary.

2. COMPUTER SYSTEM ENVIRONMENT

2.1 Hardware

The memory capacity of the Perkin-Elmer computer used is one-half million bytes. Processing is performed using thirty-two bit word formats.

The computer system hosts a variety of peripheral devices to include rotating magnetic disk, magnetic tape, cassetter recorders, electrostatic printer/plotter and a medium speed line printer (see Fig. 1).

Telecommunications access methods are achieved in one of three ways: 1) commercial telephone network; 2) dedicated short-haul modems; and 3) dedicated cable where short distances (less than four-hundred and fifty feet) are involved.

1

Reference of computer manufacturer does not constitute an endorsement by NBS.

system and provides a universal hardware communications interface. The ASCII* character data format is used in the transmission of data between the computer system and its remote devices.

2.2 Software

The software operating system installed in the computer is the "Perkin-
Elmer OS/32MT" operating system. The system permits multi-user access
to the computer's hardware and software resources. Priorities are
assigned to a computer program during the development process, and deter-
mines the execution of the program with regard to peripheral and central
processor availability.

The data base management software system is entitled the "Kaetzel, Glass,
Smith Data Base" (KGS). Software sub-systems used with the KGS data
base system are: Perkin-Elmer Multi-Terminal Manager (MTM) and the
Perkin-Elmer FORTRAN V Level I compiler. The function served by MTM is
to provide a remote software link between the user's terminal and the
production and development facilities of the computer system.

2.3 Data Base Storage

Data elements (fields) of every KGS related-publication are stored in the data base in a prescribed format. Each publication represents one data base record, (see Fig. 2). Decisions relating to the representation of data within an element are the responsibility of the user. For example, a recommended format for the author name element could be; last name-comma-space-first initial-space-middle initial. The importance of using a consistent format is reflected in the retrieval of information when the "field" search criteria is employed.

Information retrieval is available in two formats (see Fig. 3).
Optionally, the operator may have the information displayed on the CRT
terminal or direct to a line printer device in the computer facility.
Criteria used in selecting data base subsets are key word, author's
name, and publication title. In addition, provisions are made to allow
the entire data base to be displayed. Other features found in the
retrieve mode are field versus character string search, and the union
versus the intersection of a selected group of key words.

2.4 Data Base Sorting

The sequential ordering of the KGS data base is performed through the Perkin-Elmer Data System's "Sort-Merge" Software Program. The operator activates the sort program prior to entering the KGS data base system. Features of the sort routine include ascending or descending sequencing and sorting capability up to seven levels. Command files have been

* American Standard Code for Information Interchange.

[ocr errors]

and file code. By using this procedure, the entire data base is sorted and remains in that sequence until the next sort is performed.

3. MODES OF OPERATION

3.1 Data Base Supervisor Program (PDBSUP)

The function served by the data base supervisor program is to allow the operator to select from several data bases and to select the desired mode of operation. Refer to Figure 4 for software configuration. Program Modules are stored on disk and are loaded into memory only, when it is called by the operator (see Table 1). Upon completion of a run mode, control is returned to the supervisor program which idles waiting for another operator command. Error conditions which may occur during the execution of a run mode are reported on the operator's terminal, resulting in control returning to the supervisor program.

3.2 EDIT Mode (PDBED)

The user may edit the KGS data base by typing the command "EDIT". Execution of the EDIT mode begins by requesting one of the five EDIT sub-modes: 1) Search; 2) Update; 3) Append; 4) Delete; and 5) Return. All editing is performed by referencing the data base record number. Records are numbered from one to the number of the last record in the file. There is no maximum number of records except that which is dictated by disk storage capacity. Each data base record consists of fifteen lines (elements). Maximum line (element) sizes are preassigned for each record and are shown in Figure 2.

3.2.1 SEARCH Mode (PDBED)

During the process of editing the KGS data base, an identifier (record number) is used for accessing record information. The SEARCH mode is available to assist the operator in locating the desired record number. The program asks for the author's last name which is used in matching against each data base record. When a match is detected, the record number, author's name and publication title are displayed. An option to continue searching or terminate the search is available.

3.2.2 UPDATE Mode (PDBED)

The UPDATE mode is used to modify an element or elements in a data base record which is already in the data base. The program first asks for the record number, then displays the fifteen lines of text which are currently stored in the record. In addition to the text, the line or element number is affixed at the beginning of the line. These numbers, from one to fifteen are then used to identify the element (see Fig. 2).

By typing the line number, the program displays the current contents and asks for the updated information. The new text which has been entered is

character string capability is available. The program then asks for the next line number to be updated. When updating is completed, the operator types an asterisk, at which time the entire record is displayed and the user is given the option to save the updated record by typing "Y" or "N". A "Y" response will cause the information to be saved. An "N" response will cause the previously recorded text to be retained. The program then asks for the next record to be updated. When update is completed, the operator types an asterisk and another edit operation is requested.

3.2.3 APPEND Mode (PDBED)

The
The

The APPEND mode is used to splice new records into the data base. In the APPEND mode, each element is requested by first displaying the ele1 ment name and then a prompt to indicate that input is necessary. operator enters the appropriate information for the entire record. information is then displayed and the option to save or discard the information is given. A "YES" response will result in the record being stored on disk. A "NO" response will result in the request for the next record. By typing an asterisk, the APPEND mode is terminated and another mode is requested.

3.2.4 DELETE Mode (PDBED)

The nature of data base manipulation dictates that a method be provided for deleting unwanted records from the data base. The DELETE mode provides the operator with this capability. This process is achieved by marking the elements in the specified record with the characters "DELE". This suppresses the record for all future data base manipulation.

3.2.5 RETURN Mode (PDBED)

By typing "RETURN", the control of the system is returned to the data base supervisor program (PDBSUP).

3.3 KEY WORD Mode (PDBKEY)

The KEY WORD mode is designed to search each data base record and produce a sorted summary report of key words recorded in the key word fields. This program is activated in the run mode by typing the word "KEYS". option is also provided which allows output to be displayed on the user CRT or on a hard copy device.

The report is generated in an alpha-numeric sorted order. In addition to the list of key words, the number of occurrences is indicated.

A summary

1

A special character displayed by the computer to indicate that an operator command is expected.

the number of "deleted" records, and the number of different key words.

3.4 SELECT DB Mode (PDBSUP)

The purpose of the SELECT DB mode is to allow the user to transfer control to another data base without terminating the KGS data base software package. This is useful if the user wishes to maintain publications on two or more areas of research. By typing "SELECT DB" the command "ENTER DATA BASE NAME" will be displayed. The data base name entered is subsequently used to activate the appropriate retrieval mode for data base manipulation. The valid data names are maintained in the supervisor program and are modified by deleting or inserting as needs arise.

3.5 ONE KEY Mode (PDBQK)

The purpose of the ONE KEY mode is to assist the data base operator in querying the data base for specific publication topics (key words). By typing a key word value when requested, the program searches the data base and displays those records (if any) which contained that key word. This is useful if a "snapshot" of the data base is desired, without displaying each data base record.

3.6 RETRIEVE Mode (KGSEX)

Information retrieval and display of the KGS data base is accomplished using the RETRIEVE mode. Various search criteria such as key word or author name may be entered and subsets of the data base extracted. Also included is the option for a field or character search of a data base record. A field search attempts to match the characters entered by the operator with the characters contained in the data base element. A string search will search the entire element for a match of the characters entered by the operator with any contiguous character match contained in the data base record element. The maximum number of characters in a string search is one hundred. In the case where the key word search option is used, each of the six key word fields plus the title field will be employed in searching. Optional printer hardcopy output is available. An option for the "UNION" or the "INTERSECTION" where more than one key word is also provided. In the UNION ('OR') condition, records must have all key words (maximum six) in either the title or key word field. In the INTERSECTION ('AND') condition a match on any of the entered key words will result in the record being retrieved.

When a string search criteria is employed the operator may enter the root of a key word which allow records to be retrieved having that root plus variations. For example:

Entering the characters "COL" might result in records having key words such as:

« PreviousContinue »