Older people living in residential and nursing care homes spend a large proportion of their time within the boundaries of the home, and may depend on the environment to compensate for their physical or cognitive frailties. Regulations and guidelines on the design of care buildings have accumulated over time with little knowledge of their impact on the quality of life of building users. The Design in Caring Environments Study (DICE) collected cross-sectional data on building design and quality of life in 38 care homes in and near Sheffield, Yorkshire. Quality of life was assessed using methods which included all residents regardless of their frailty, and staff morale was also assessed. The physical environment was measured on 11 user-related domains using a new tool, the Sheffield Care Environment Assessment Matrix (SCEAM). Significant positive associations were found between several aspects of the built environment and the residents' quality of life. There was evidence that a focus on safety and health requirements could be creating risk-averse environments which act against quality of life, particularly for the least frail residents. Staff morale was associated with attributes of a non-institutional environment for residents rather than with the facilities provided for the staff. The new tool for assessing building design has potential applications in further research and for care providers.

The concept of daylight coefficients is introduced, mathematical functions that relate the luminance distribution of the sky to the illuminance at a point in a room. They are illustrated first in the context of standard daylight calculations, and then in the form of finite element matrices. Graphs of the functions enable effects of sky luminance changes to be appreciated qualitatively; use of the coefficients in computer programs gives an efficient method for calculating daylight from a number of sky brightness distributions in succession.

A scanning pattern for sky photometry is described, in which the hemisphere is divided into 151 zones in bands parallel with the horizon.

There is less discomfort glare from a window with an interesting view than from either a neutral screen or a window with a poor view having the same daylight glare index. A view with patches of high luminance is more glaring than a more uniform view with the same mean luminance. In two experiments carried out during May and June 2005, subjects assessed glare from windows in a 20-storey building in Sheffield, UK. The rooms used were identical except that at different levels and orientations the prospects varied from close views of other buildings to extensive urban and natural scenes. In the first experiment, with 72 university students as subjects, the reported discomfort glare from a diffusing screen covering the window was compared with that from a view onto a nearby obstructing wall and from a view with trees and distant landscape. The weather was changeable so each scene was observed over a range of different luminances. In the second experiment, with 96 subjects, the glare from an urban view that included nearby elements, the middle distance and the far distance was compared with a restricted urban view, with an extensive natural view and with a restricted natural view. Independent groups of subjects rated the interest of all the views. It was found that subjects' reported discomfort from glare varied significantly between all three cases in the first experiment and all four cases of the second experiment: glare discomfort decreased as interest increased. The luminance range within a view also affected the experience of glare. The results are consistent with those of an earlier study of glare from small screens.

The CIE Standard General Sky consists of a family of luminance distributions which can be matched to measured sky brightness patterns. The daylight climate of a site can be described concisely and effectively by the statistical distribution of the General Sky types that best fit the sky luminance patterns that occur there. This paper gives a procedure for deriving the frequency of occurrence of General Sky types from International Daylight Measurement Programme sky scan data. Measured luminances are normalized with respect to horizontal illuminance and solar elevation, then the sky types giving the best least-squares fit are found. A data structure for daylight climate simulation is proposed.