[eDebate] Hanson-Cantrell STA-XL plus

Gary Larson Gary.N.Larson
Tue Sep 11 20:47:30 CDT 2007

Since I've received a couple of back-channel inquiries, let me explain my relationship to Jim's project.  From the early days when STA was embedded in SmartWare, I've never aggressively marketed my program.  A few brave souls have used it with as much support as I've been able to offer.  As a result, when Jim Hanson and Danny Cantrell approached me regarding their desire to adapt STA into a package that they believe will be more user friendly and one that Danny will fully support, I offered my enthusiastic endorsement.  While I continue to work on STA, particularly with respect to the judging algorithm, my role as an academic administrator prevents me from offering the ongoing support to a large cadre of end users who need troubleshooting, development of new features, and customization.  I anticipate that over time, Danny and Jim will gradually move away from the core STA code that still underlies the program.  This is entirely appropriate as I anticipate that a new generation of brighter programmers will advance the project.

In the meantime, it should be noted that I will continue to develop the algorithms on which STA has been based.  This summer, in response to the opportunity to test the ordinal data compiled at the NDT, I was able to develop a new MPJ algorithm that makes dramatic improvements over the best results I've previously obtained.  While Jim and Danny's program currently contains an older version of my algorithm, it is anticipated that as appropriate, they will incorporate newer versions of my algorithm into their project.  At the same time, I have discussed with Rich Edwards developing a collaboration that would permit my judge assignment algorithms to be incorporated into TRPC.

Finally, I will continue to use and support those others who currently use my own version of STA-XL.  While fully supportive of Jim and Danny's project, I have my own natural biases about user interfaces and programming strategies.  That said, since I don't have the time and resources to realistically support new users, evidenced by the fact that I don't have an up-to-date manual that keeps pace with the relatively rapid evolution of my program, I don't really recommend that folks use my version of the program at this time (as much as I personally like it).  To promote ongoing innovation, I want to endorse all efforts to move the state of the art forward - e.g. Jim and Danny's project, Rich's current TRPC as well his collaboration with Jon Bruschke to incorporate .NET programming.  I will continue to make my underlying code available to all projects that are willing to give rather than sell their software to the community.


More information about the Mailman mailing list