One of the reasons I chose to plunge in with a conversion of ECJ in January was because I am implementing a comprehensive platform for EC in cloud environments. I wanted the EC engine to operate as efficiently and seamlessly as possible on Windows Server 2008 R2 and Windows7, because I'll be integrating with Microsoft's AppFabric, Azure, and a few other goodies.
The somewhat broader picture, however, is that I want to integrate with Project Trident for collaborative workflow. This, in turn, is related to myExperiment and myGrid. And those, in turn, are related to Taverna, an open source workflow management platform in java connecting global scientific research groups and projects.
All by way of saying, you should maybe take a look at integrating ECJ and MASON into such collaborative "watering holes". I think your systems are made-to-order for that kind of integration.
Just a thought!
On Wed, May 11, 2011 at 11:29 AM, Sean Luke <[log in to unmask]> wrote:
On May 11, 2011, at 10:34 AM, Ben Stabile wrote:

As you know, one of the reasons I'm discovering these little "quirks" is because in converting to .NET I'm building a lot of unit tests.

That's right, you're doing the .NET translation, I forgot.

Obviously for a conversion, the unit tests are mandatory. Do you have internal test batteries that you run against these projects?

Nope, and it's a big failing.  MASON and ECJ both badly need a testing regimen.  As they're academic projects, our major source of funding for such things are grants.  MASON's got a small NSF grant approved for exactly, but with the budget cut at NSF it's looking unlikely to be awarded.  Which means that they'll have to wait unless others can start pitching in.

Last year I built an extensive manual for ECJ, which required me to go through ECJ with a fine-toothed comb, kicking up a variety of (thankfully minor) bugs and misfeatures.  It was an enormously valuable exercise.  I'm now doing the same thing with MASON but it's taking quite a long time given my current academic schedule.  So anyway, it's high on the list.

I would volunteer to build such batteries myself were it not for severe constraints on my time. Surely there are comp science students there who could be enlisted to practice formal testing techniques on your projects? ;-)

Tests for complex 70K system are nontrivial!

In any case, I have a HUGE amount of work to do to translate the MASON code (Java3D, OpenGL, Swing, etc.). It is frankly quite daunting, and I'm still trying to find the best way to hook appropriately into WPF and Direct3D. Yikes!

As I've mentioned before, you are totally nuts.  :-)  Good luck!