Defining subsystems by using FD can lead to poor architectural concepts FD alone cannot be used to determine whether the subsystems will provide common functionalities that could be used across many requirements. Using functional decomposition /allocation to create architecture, can produce subsystems with duplicate functionality Designing subsystems with duplicate functionalities will create unnecessarily large and complex systems The logical functional decomposition enables – Realization that multiple system functions/requirements can be met by each component or subsystem – Formulation of alternate functional concepts; e. g. utilizing subsystems interactions to meet several system requirements Subsystems should, ideally, provide services in a "one- to-many, many-to-one" relationship: – One system requirement can be decompose into many subsystem requirements – One subsystem requirement may be derived from many system requirements – The approach to subsystems service definition is a form of functional decomposition (determining subsystems first, and then deriving their requirements)

We investigated 283 suicides in San Diego between November 1981 and June 1983. The first 204 cases were collected consecutively. The next 79 were consecutive cases under age 30. The sample consisted, then, of three separate but overlapping consecutive series: 133 cases under age 30, 150 cases age 30 and over, and 204 cases of all ages. Overall, substance abuse/dependence was diagnosed in about two thirds of the cases, both over and under age 30. Depressive disorders were most commonly diagnosed in association with substance abuse (SA) in both age groups. Many other disorders also accompanied SA. There were significantly more diagnoses of “drug” abuse among the younger cases than among the older suicides. The frequency of alcohol abuse was the same. There were no significant differences in the rates of SA between the men and women in the 204 consecutive cases. Of the 45 suicides with psychotic disorders in the consecutive series, more than half (62%) were also substance abusers. Of the 15 cases who had personality disorders, 14 also were substance abusers. Also, 12 of 13 gay males were known to be substance abusers. In all cases, the drugs most frequently abused were alcohol, sedatives, marijuana, and stimulants. Depressive symptoms were common in the cases with SA even if an affective disorder was not formally diagnosed. We conclude that SA is possibly far more important as a primary etiology of suicides than was previously believed.

This paper uses an assignment model to understand the cross section of house prices within a metro area. Movers' demand for housing is derived from a lifecycle problem with credit market frictions. Equilibrium house prices adjust to assign houses that differ by quality to movers who differ by age, income and wealth. To quantify the model, we measure distributions of house prices, house qualities and mover characteristics from micro data on San Diego County during the 2000s boom. The main result is that cheaper credit for poor households was a major driver of prices, especially at the low end of the market.