JAGS is now able to run most of the classic BUGS examples. It is now
in beta.

Goals for version 1.0 are:
- Removing any bugs
- Stabilizing the library API
- Improving speed
- Reducing memory footprint

Currently working on:

- introduce loadable modules
- Gamerman's sampler for GLMs
- Multivariate slice sampler
- More efficient mixture model compilation
- Efficient likelihood calculations for discrete data
- Check parameter validity at initialization and update for stochastic nodes
- Default initialization for top-level stochastic nodes
- Use doubling method for Univariate slice sampler before burnin.
