Abstract. Dynamic memory allocation has been a fundamental part of most computer systems since roughly 1960, and memory allocation is widely considered to be either a solved problem or an insoluble one. In this survey, w e describe a variety of memory allocator designs and point out issues relevant to their design and evaluation. We then chronologically survey most of the literature on allocators between 1961 and 1995. Scores of papers are discussed, in varying detail, and over 150 references are given. We argue that allocator designs have been unduly restricted by an emphasis on mechanism, rather than policy, while the latter is more important; higher-level strategic issues are still more important, but have not been given much attention. Most theoretical analyses and empirical allocator evaluations to date have relied on very strong assumptions of randomness and independence, but real program behavior exhibits important regularities that must be exploited if allocators are to perform well in practice. ? A slightly di erent v ersion of this paper appears in

No abstract

A full upper-and lowercase visual similarity matrix is presented for a standard set of computer characters, implemented on the Apple-Psych system. The 2,704 (52 x 52) letter pairs were rated by 12 subjects each. From the ratings, generation and veridical similarity values are derived, and they are tabled for use in research on mixed-case letter matching. In addition, the results of multidimensional scaling and cluster analyses are presented, which give complementary, simplified descriptions of the data. The impact of visual similarity on pure-ease letter comparisons (e.g., between "A" and "A") is well documented. Letters that are commonly confused with one another (e.g., "GQ") produce longer reaction times (RTs) in same-different matching tasks than do letters that are rarely confused (Bagnara, Boles, Simion, & Umilta, 1983). This effect is a source of evidence that pure-case letter pairs are matched physically (visually), and, indeed, the task of matching such pairs is known as the physical matching task (Posner, 1978). Mixed-ease letter comparisons (e.g., between "A" and "a") have come to be known as name matches, because introspection and early experimental results suggested that such matches are made by extracting and comparing phonetic representations of the letter names (Dainoff &

No abstract

Physically identical letter pairs are matched more quickly than are nominally identical or nonidentical pairs, which is an effect usually interpreted as resulting from the use of a visual memory code versus a phonetic or name code. However, prior manipulations of phonetic confusion and visual-field lateralization have provided little evidence consistent with this notion. Here, three reaction time experiments suggest through converging confusion and visual-field manipulations that a phonetic code is not used in either simultaneous or successive letter matching. Two additional experiments yield confusion evidence consistent with a rapid visual generation process underlying nominally identical and nonidentical matches, emphasizing the role of visual memory processes in all of the commonly used same-different letter-matching tasks. Implications for coding and hemispheric theories are discussed.