Narrating the archive and archiving narrative: the electronic book and the logic of the index

The creation of my hypermedia work Index of Love, which narrates a love story as an archive of moments, images and objects recollected, also articulated for me the potential of the book as electronic text. The book has always existed as both narrative and archive. Tables of contents and indexes allow the book to function simultaneously as linear narrative and non-linear, searchable database. The book therefore has more in common with the so-called 'new media' of the 21st century than it does with the dominant 20th century media of film, video and audiotape, whose logic and mode of distribution are resolutely linear. My thesis is that the non-linear logic of new media brings to the fore an aspect of the book - the index - whose potential for the production of narrative is only just beginning to be explored. When a reader/user accesses an electronic work, such as a website, via its menu, they simultaneously experience it as narrative and archive. The narrative journey taken is created through the menu choices made. Within the electronic book, therefore, the index (or menu) has the potential to function as more than just an analytical or navigational tool. It has the potential to become a creative, structuring device. This opens up new possibilities for the book, particularly as, in its paper based form, the book indexes factual work, but not fiction. In the electronic book, however, the index offers as rich a potential for fictional narratives as it does for factual volumes. [ABSTRACT FROM AUTHOR]

Three parallel programming paradigms: Comparisons on an archetypal PDE computation

Three paradigms for distributed-memory parallel computation that free the application programmer from the details of message passing are compared for an archetypal structured scientific computation -- a nonlinear, structured-grid partial differential equation boundary value problem -- using the same algorithm on the same hardware. All of the paradigms -- parallel languages represented by the Portland Group's HPF, (semi-)automated serial-to-parallel source-to-source translation represented by CAP-Tools from the University of Greenwich, and parallel libraries represented by Argonne's PETSc -- are found to be easy to use for this problem class, and all are reasonably effective in exploiting concurrency after a short learning curve. The level of involvement required by the application programmer under any paradigm includes specification of the data partitioning, corresponding to a geometrically simple decomposition of the domain of the PDE. Programming in SPMD style for the PETSc library requires writing only the routines that discretize the PDE and its Jacobian, managing subdomain-to-processor mappings (affine global-to-local index mappings), and interfacing to library solver routines. Programming for HPF requires a complete sequential implementation of the same algorithm as a starting point, introduction of concurrency through subdomain blocking (a task similar to the index mapping), and modest experimentation with rewriting loops to elucidate to the compiler the latent concurrency. Programming with CAPTools involves feeding the same sequential implementation to the CAPTools interactive parallelization system, and guiding the source-to-source code transformation by responding to various queries about quantities knowable only at runtime. Results representative of "the state of the practice" for a scaled sequence of structured grid problems are given on three of the most important contemporary high-performance platforms: the IBM SP, the SGI Origin 2000, and the CRAYY T3E.