A set of equations is introduced which describes the inhibitory interaction between the receptors in the eye of the horse-shoe crab, Limulus. The terms primary inhibition field, and inhibition field are defined, and the results of a digital computer study of their interdependence with respect to size and to the type of k[subscript i,j] and t[subscript i,j] functions employed are given. It is concluded that a large receptor array can be simulated by solving the set of equations for the output of the central receptor in much smaller sub-arrays. If the sub-arrays are chosen so as to cover the large array, then the resulting set of central receptor outputs provides a good approximation to the output of the large array. A device which simulates a large receptor array by repetitive solution of the set of equations for 9×9 sub-arrays is described. Various switching arrangements which reduce the complexity of the device are detailed. The circuit diagrams for all the required elements are given. In order to evaluate the possible utility of the device as a tool in pattern recognition studies, the output from the device for certain simple patterns was simulated on the digital computer. The results of this study are given, and some possible applications to the field of pattern recognition are mentioned. It is pointed out that the design of the device makes it a versatile instrument for simulating various types of receptor arrays. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

Accessibility of the forest is a basic requirement for a commercially managed forest. Logging in British Columbia often is carried out in remote areas where log transportation is primarily by truck hauling on forest roads. Forest road construction and particularly earth moving are significant elements in the total cost of forest products. Proper choice of road design elements will determine the optimum economy of any forest road. The derivation of road design elements is shown in detail as an introduction to the earth allocation problem. Minimization of costs of main forest access roads is studied in this thesis and alternate methods of mass allocation are presented. A semi-graphical method of mass allocation (mass diagram) is compared with a method employing the electronic computer and the tools of operations research (linear programming). The theory of linear programming (LP) is shown as the optimization technique used for minimizing the earth moving costs. The LP assumptions and limitations are discussed. The two methods were tested on the forest main haul road C in the University of British Columbia Research Forest. Calculations of volume distribution and the required intermediate calculations are carried out with an electronic computer for comparison with traditional methods. The mass diagram method might be used for a long time due to its simplicity, whilst LP provides a more precise solution. The costs of earthmoving and planning are $84.00 or 0.6% of the total earthwork and planning costs less by using LP rather than the mass diagram in the example calculated. The use of dynamic programming (DP) to determine the optimum road lay-out is suggested as a topic for further research, a preliminary step for optimization in mass allocation. / Forestry, Faculty of / Graduate

Pedology, like most other sciences, is facing a data explosion in which it is becoming increasingly difficult to organize, summarize and interpret large quantities of data. Coupled with this is an unprecedented demand for soils information for consideration in resource management and environmental considerations. Since decision making in these areas increasingly has to be justified by economic criteria, the need for evaluation of soils information for land use considerations in economic terms is paramount. In this study, the British Columbia Soil Survey Data File was used. The file contains only routine soil survey data. It was improved and modified to correct problems arising from earlier experiences. The identification, organization and coding techniques used in the data file are presented. In view of the interest in establishing national data bank systems a modified hierarchical organizational and identification system is proposed, which should be equally applicable at regional, provincial and national levels. The Data File was also used to explore by statistical techniques the inter-relationships between soil properties, the feasibility of predicting the values for dependent variables by multiple regression equations and to study the modal concept of soil which is basic to soil classification and subsequent statistical analysis. Numerical taxonomy techniques were used to determine the feasibility of using objective statistical techniques in the development of a model by which soils could be rated for a specific land use, as well as, determine on the basis of correlation and regression analysis an estimate of hypothetical treatments and costs that would make a given soil behave more like an ideal soil for the use in question. The study showed that, using cash cropping and road bed construction as two contrasting soil uses as examples, the derivation of cost estimates for soil manipulation is feasible. However, the derivation of the ideal soil (model) is critical. / Land and Food Systems, Faculty of / Graduate

Due to the introduction of modern computing technology to planning, it is anticipated that the techniques for quantitative analysis in planning will be changed radically; the expanding use of model and the increasing importance of inter-city data collection may have an effect in redefining the process of planning itself. In this thesis, ah effort has been made to investigate the current problems of building and using models in planning so that a methodology, combining several multivariate statistical methods and the modern computation algorithms, is developed for planning analysis and prediction from the comparative point of view. It is hypothesized that simple prediction models can be improved by the use of these comparative techniques. As a testing mechanism of the proposed methodology,a simple regression model for CBD floor space prediction is devised, and the result shows that the comparative analysis is effective for better prediction. / Applied Science, Faculty of / Community and Regional Planning (SCARP), School of / Graduate

This study used the word processor as a tool for written output to examine the effects of an experiential 'Write to Read' program on typing performance, decoding strategies and successive processing in learning disordered children with motor dysfunction. A case history approach was taken in view of the small number of subjects available, and in order to adequately describe each individual's unique and complex cognitive motor profile. Subjects involved in the study were three male students in a Junior Learning Assistance Class in a Lower Mainland British Columbia school district elementary school. Each of the students had a history of poor motor performance, poor handwriting and delayed reading ability. The three subjects were involved in an eight week intervention program which taught keyboarding and word processing techniques using the 'Write to Read' program, a systematic method of training the motor skills required. It was hypothesized that, if the learning disabled student with poor motor skills could use the word processor as an adjunct to handwriting, the improved legibility would facilitate consistent decoding by the student of his own work, reinforcing acquisition of early reading skills. Within the case history format, a theoretical frame of reference based on the simultaneous - successive information processing model was chosen and a limited time series design measured the effects of the intervention on successive processing as determined by a block sequencing task (Das, Kirby and Jarman, 1980). The data was collected for each student and graphed for visual inspection, graphic analysis and statistical analysis. One subject showed a stable and significant intervention effect, and no stable trends or significant results for successive processing were found in the other two subjects. Rates of word processing output increased over the course of the study and the number of errors declined. All subjects made progress in measures of decoding and word analysis. Implications for future research and professional practice were described. / Education, Faculty of / Educational and Counselling Psychology, and Special Education (ECPS), Department of / Graduate

Objects are entities which encapsulate data and those operations which manipulate the data. A distributed, object-based programming system (or DOBPS) is a distributed operating system which has been designed to support an object-based programming language and, in particular, an object abstraction. DOBPSs have the benefits of simplifying program construction and improving the performance of programs by providing efficient, system-level support for the abstractions used by the language. Many DOBPSs also permit hardware and software failures to be tolerated. This thesis introduces a definition for the term "distributed, object-based programming system" and identifies the features, that are related to objects, which are required by an operating system of a DOBPS. A classification scheme is presented that categorizes and characterizes these features to permit a number of implementation techniques to be easily examined, compared, and contrasted. / Science, Faculty of / Computer Science, Department of / Graduate

Unequaled improvements in processor and I/O speeds make many applications such as databases and operating systems to be increasingly I/O bound. Many schemes such as disk caching and disk mirroring have been proposed to address the problem. In this thesis we focus only on disk mirroring. In disk mirroring, a logical disk image is maintained on two physical disks allowing a single disk failure to be transparent to application programs. Although disk mirroring improves data availability and reliability, it has two major drawbacks. First, writes are expensive because both disks must be updated. Second, load balancing during failure mode operation is poor because all requests are serviced by the surviving disk. Distorted mirrors was proposed to address the write problem and interleaved declustering to address the load balancing problem. In this thesis we perform a comparative study of these two schemes under various operating modes. In addition we also study traditional mirroring to provide a common basis for comparison.

No description available.

Bibliography: p. 100-104. / Office automation systems have become an essential tool for the operation of the modern office. With the emphasis of a modern office being placed on efficiency and ease of communication, office automation systems have become the backbone of successful businesses. COSNET is a prototype office automation system designed and implemented at the Department of the University of Cape Town and runs on Personal Computers that are linked to a NCR UNIX TOWER, which acts as the host. This dissertation investigates the different facilities supported by some of the office automation systems compared in this thesis, and describes the COSNET features. This prototype office automation system supports many of the facilities that are supported by large office automation systems. COSNET allows the user to define any MS-DOS based editor or word processor, and uses a simple editor for the creation of mail. The electronic filing facility allows documents to be created, filed, retrieved and deleted, and thus provides the users with the necessary features for document exchange. A user may set access permissions to each of his documents and may grant other users either read or write access to a specific document. The mail facility lets the user read, file, forward, delete and print a message, and supports classification of mail. A calendar facility is used as an electronic diary and stores all the user's schedules. These schedules may be viewed in either daily, weekly and monthly display modes. Read and write access to the calendar can be set by the user, in order to allow other users to manipulate his schedules. Any MS-DOS based application software can be added to COSNET. This facility allows the COSNET user to configure the office automation system to simulate the office environment. COSNET thus supports most of the necessary features required by an office automation system.

Bibliography: pages 80-86. / The aim of the present study was to test a model of stress and to examine whether the theoretical construct of sense of coherence (SOC) moderated the relationship between stressors and health-related and work-related outcomes. This construct of SOC was identified by an Israeli medical sociologist, Antonovsky. He maintained that the current focus of research on stress is largely pathogenic in nature. He suggested that it would be of value to shift research more towards that which identifies the origins of health. He consequently developed the term "salutogenesis", which requires people to focus on those factors which promote well-being. He also argued that people are not either sick or well, but rather are located on a continuum between health-ease/dis-ease. With respect to their health, persons will find themselves somewhere along this continuum, where they may shift between the two positions. He then suggests that certain factors contribute to facilitating the movement along this continuum. These factors together form a construct which he calls the SOC. The SOC is comprised of core components. He hypothesizes that someone with a strong SOC is likely to make better sense of the world around him/her, thereby engendering resilience towards the impinging stressors. The person with a weak SOC is likely to capitulate to these stressors · more readily and by succumbing to them is going to increase the likelihood that (s)he will move to the dis-ease end of the continuum. This study attempted to investigate the following research questions, namely, whether (1) the stressors were related to the stress outcomes, (2) the SOC was related to the stressors and outcomes, and (3) the SOC moderated the relationships between stressors and outcomes. In the present study the subjects were drawn from all data processing professionals in a large financial organisation. The respondents (~ = 194) replied to a questionnaire which contained scales which measured a variety of job-related stressors, an SOC scale as well as job-related and health-related outcome variables. Intercorrelations between the stressor, moderator and outcome variables were calculated. Other statistical procedures that were utilized were subgroup analyses and the moderated multiple regression analyses. Partial support for all three research questions was obtained. Four of the six stressors were found to correlate significantly with somatic complaints, thereby suggesting that stressors result in persons feeling the results of stress and reporting them physically. The SOC was found to relate to some of the stressors and outcome variables. This would lend partial support to an interpretation of the SOC as having a main effect relationship to stressor and outcome variables. In the subgroup analyses the results showed that out of a possible 54 relationships, the SOC moderated in only seven of them that the moderated multiple regression (MMR) analyses showed out of 54 possible relationships, the SOC moderated in 12 of them health-related variables. Furthermore, the SOC moderated between six outcome variables and six work-related outcomes. These findings then partially support research question 3, which examined whether the SOC would moderate relationships between stressors and outcome variables. This study was concluded by a discussion of the findings, its implications, and the limitations of this research.