Describes a case study that applies the principles of “a proposed quantitative model for building repair and maintenance” developed in theory by Alani et al. The proposed model relies on the relationship between the number of defects and time where the collection of the number of defects data has been made using a condition based survey and specifically developed survey form. Using established declining balance depreciation methods a close correlation between the models cost/time and quality/time curves has been determined. Applies the proposed coded system of observing, assessing, recording and appraising the condition of the elements of a large residential building and calculates the cost of repairing and maintaining (at a pre‐determined condition) the elements of that specific building.

The Portsmouth smart homes project sought to identify appropriate technology to create energy efficient accommodation that supports the occupants in achieving a more independent life than would normally be possible due to their physical disabilities. A consultation process was initiated. This began with focus groups in three different parts of Hampshire. These were followed by in-depth interviews to ascertain how``smart'' technology might enhance independence, quality of life, security and affordability. The consultation process and a technology review informed a number of design exercises culminating in the design of six properties in Portsmouth due to be occupied in April 2002, and monitored through the following summer and winter. Three of these units will accommodate wheelchair users. Internal and external doors, locks and all windows will be motorised. The smart system will be programmed to operate heating, lighting, and household appliances, to a pre-determined plan, which can be overridden by the resident. The sensors and devices share data in order to act, react and interact in an appropriate way. Disabled people consulted report positive attitudes to smart homes. They want a home that can respond to emergencies and environmental changes. Potential occupants also express the wish to be able to counter automated controls. A majority also express preference for homes that will not stand out from neighbouring properties.

This paper arises from findings which emerged during the preparation of the Good Practice Guide to Maintenance Cost Forecasting. This Guide was commissioned by the Housing Corporation to assist Registered Social Landlords forecast the future maintenance costs of their stocks. It became apparent during the work that client experiences of stock condition surveys had been far from satisfactory. The surveys had often been carried out by professional surveyors and yet many had fundamental weaknesses in their specification, execution and use of data. The level of dissatisfaction expressed, coupled with recent evidence regarding the quality of domestic building surveys, should give cause for concern to the surveying profession. The paper identifies some causes of this dissatisfaction and suggests lessons that can be drawn from the experience. Further investigation is recommended into this crucial area of practice.

We describe the design and prototype implementation of a dialect of Java, XJ, that supports both closed and open nested transactions. As we have previously advocated, open nesting most naturally attaches to the class as the primary abstraction mechanism of Java. The resulting design allows natural expression of layered abstractions for concurrent data structures, while promoting improved concurrency for operations on those abstractions. Moreover, we describe our approach to constructing a prototype implementation of XJ that runs on standard Java virtual machines, by grafting support for transactions onto both application code and library code via load-time bytecode rewriting, for full execution coverage. We rely on extensions to the javac compiler, a JVMTI run-time agent to intercept and rewrite Java classes as they are loaded into the virtual machine, and a run-time library that tracks and manages all transaction meta-data. The resulting prototype will allow further exploration of implementation alternatives for open and closed nested transactions in Java. Our design also addresses the issue of internal deadlock caused by accessing the same data in both closed and open nesting fashion by carefully disallowing such access.

The importance of survey data accuracy is paramount if school maintenance programs are to be a true reflection of the maintenance needs of that school. Previous research has identified the issue of surveyor variability, i.e. the situation where two or more surveyors, surveying the same building, arrive at very different survey decisions. The research presented in this paper reports on social judgement theory ± a model of a surveyor's judgements where the varying values of surveyors, in terms of the``importance'' they give to building elements, can be elicited by using the regression formula. The results of the research can be used to normalise survey data in an attempt to make them more truly reflect the actual condition of a school. The results can also be used to assess training requirements for individual surveyors.