132011-03-28 18:44:45snthomas99122010-05-11 21:34:52NealThomas112010-05-07 19:35:57NealThomas102010-04-29 21:46:31ncoulter92010-03-08 13:36:17NealThomas82010-01-21 03:56:18NealThomas72009-12-15 03:21:49NealThomas62009-12-15 00:04:29NealThomas52009-12-14 23:54:31NealThomas42009-12-14 23:53:09NealThomas32009-07-27 16:14:05NathanCoulter22008-12-09 15:36:59MikeKSmithChanged to use CamelCase12008-12-09 15:18:47MikeKSmith

The Chief Software Bug is Andrew Thomas. Other members of the original BUGS and WinBUGS teams were statisticians David Spiegelhalter, Nicky Best, Dave Lunn and Ken Rice. Dave Lunn has also made major contributions to the software development.

For technical coding details, see the Code page. This page includes instructions for building a new version of OpenBUGS. Instructions for maintaining the code repository and archiving a new release: repository guidelines If you want to try some development, give us some feedback (you will have to install Black Box).

The Windows installation includes a file script.txt in the main program directory that can be executed to confirm basic function of the program. The results can be compared the results in the 'seeds' example. The corresponding linux file is called testscript.txt and it is included in the 'doc' subdirectory of the primary installation. The R2OpenBUGS has a function called validateInstallOpenBUGS that executes 15 of the OpenBUGS examples and compares the results for agreement with the results produced on the developers' machine. There are currently two types of testing: 1. Comparison to results from WinBUGS 1.4.3 2. Cook-Gelman-Rubin validation testing

Comparisons of WinBUGS and OpenBUGS for each example are in execution times. The times are based on the same burnin and sample sizes, so they do not account for potential differences in convergence rates or autocorrelation.