Administration

The deveoper's meeting began with some talk of migrating the web server and subversion repository to a mac mini server. Some logistical issues regarding this issue were discussed. The next topic of conversation was designing an interface for the likelihood variables in a FIML objective function.

To create a new release, do the following:

• Update the TARGET in the Makefile
• Update the Version field in the DESCRIPTION file
• Update the Date field in the DESCRIPTION file (NEW)
• Rename the "trunk" in the CHANGES file to the new version number
• Update release and possibly version in docs/source/conf.py
  (only if Sphinx documentation has changed)

The files UnivariateTwinAnalysis20090925.R, Twin_independent_path_general_factor.R, TwinAnalysis_Multivariate_Matrix_Raw.R, and NTF_design.R are being moved to the models/failing directory in preparation for the next closed beta release. All files in the test suite must follow the following guidelines:

• Never use setwd(). It's not going to work on my machine.
• Never use source(). The point of a demo is that the entire example is in a single file.

I had an idea for shrinking the binary releases. At the moment, we ship the binary release for each platform with copies of NPSOL for all platforms, even though the binary release was built using only one of those copies. After release 0.1.4 has been made available, I'm going to play with the configure script such that only one copy of NPSOL is shipped.

It was noted that another release of the OpenMx closed beta library will be produced today.

There are a discussion of the next immediate implementation tasks. On the front-end, it was decided that square bracket notation should be implemented, and a tool from converting Mx 1.0 algebra expressions to OpenMx algebra expressions is needed. On the back-end, the most immediate goal is to implement some sort of checkpointing to disk.

The following question seems unusual, but if someone figures out how to do it then we can do a single mxEval() call for all the algebras in a model to test the algebras for conformability during mxRun before the optimizer is called, instead of one mxEval() call per algebra. This could speed up our test suite (not by much, expect something like 5 - 10% increase).

https://stat.ethz.ch/pipermail/r-help/2009-September/211731.html

As of revision 769, MxAlgebra are checked for correct dimensions before the optimizer is called.

> A <- mxMatrix("Full", nrow = 1, ncol = 1, name = 'A')
> B <- mxMatrix("Full", nrow = 1, ncol = 2, name = 'B')
> C <- mxAlgebra(A + B, name = 'C')
> model <- mxModel('model', A, B, C)
> mxRun(model)
Running model 
Error: The algebra 'model.C' in model 'model' generated the error message:
non-conformable arrays

A gentle reminder that scripts must live in either the models/passing, models/failing, or demo/ directories, and in exactly one of these locations. Duplicating the files causes "make test" to run longer than necessary (which is a nuisance), but also forces the developers to make corrections to an existing script in two locations (which they will forget to do). Use the "svn mv" command when moving files in subversion. Also, when adding a script to the /demo directory, remember to update demo/00INDEX, although people have been good about that lately.