|
Scientific Software Database
|
|
The Scientific Software Database (SSD) will be launched on January 1st 2010. The SSD currently contains computer programs in FORTRAN,, Maple, MATLAB, Java, python and Sage for a wide assortment of scientific computations in chemical physics, quantum physics, quantum computing, applied mathematics, pure mathematics, bioinformatics, computational genetics, chaos theory, game theory, fractal analysis and computational neuroscience. Purposes With the rise of the use of numerical computation as an experimental technique in science, many articles in the scientific literature have been published without confirmation of the veracity of their numerical results, and in many cases, false results have leaked into the literature - either because the authors used a program that had a bug about which they were unaware, or because knew they could cheat the current with false reports and that no one would care to spend the time rewriting a code simply to verify their results. In either case, the SSD community provides an open medium in which scientific programmers share their codes and therefore allows scientific results to be verifiable and in some cases, for codes to be enhanced. How it works Authors of programs used for any type of scientific computation can at any time upload their code to the database. Participating authors may choose to share their code for free, or to put a price for its use. In either case, downloads of codes are tracked, and anyone that publishes a paper in a scientific journal, in which the paper was based partially on use of that code, is expected to reference credit the authors appropraitely. Members of the SSD community monitor publications as they are put forth: If a paper is published by an author who has recently downloaded a particular code, and appears to have used that code for simulations leading to the publication without referencing it properly, their privilege to download more codes from the SSD may be severely restricted. The SSD keeps track of which codes appear most useful in the scientific literature based on citations, and authors of highly cited software can be awarded research grants. Premium members (members who have paid a fee) may have their codes reviewed by expert programmers for optimization and algorithm-enhancement or parallelization. Authors may also make downloading of their code restricted on a need for permission basis. Support and Controversy While open-source proponents like William Stein, and quantum computing researcher Michael Nielsen (see for example, the prequel to his newest book The Future of Science ) advocate the idea of a more open world of science , many people criticize the aims of the SSD. The two main arguments against the SSD are the fear of code pollution (the fear that source codes let lose can be "polluted" by someone customizing it, and likely changing its mathematical integrity, and then distributing it somewhere untrustworthy), and the lack of virtue in the market portion of the database. People arguing the latter point disagree that scientific code should be sold as a commodity in this manner. Other open source databases and scientific libraries www.sourceforge.com open source GNU LAPACK libraries 
|
| < Prev | Next > |
|---|
Main Menu
| Home |
| Articles |
| Search |
| Submit Article |
| Contact Us |
| Donate to Wikibin |
