From: ituvidq-admin@its.bldrdoc.gov on behalf of H. Myler [hrm@engr.ucf.edu] Sent: Thursday, May 03, 2001 3:44 PM To: VQEG Subject: [VQEG] Burden on Independent Labs Fellow VQEGgers, In Section 4.3 (Results Analysis) of the Test Plan, the ILs are required to run a sample of the data through the submitted proponent software to validate results. Would it be conceivable to have the proponents only submit software to the ILs and only require a validation pending the outcome of the statistical analysis? The proponents would run their algorithms and then submit results to directly to VQEG. After analysis, some selection of proponents, presumably the best performers though not limited to them, would then be validated by an IL at the proponents expense on both the sequences seen by the proponent and sequences not seen. This approach would reduce the burden on the labs and if a proponent did not fare as well as expected, it would reduce the overall amount of testing. Can anyone tell the semester is over? -- Harley R. Myler, Ph.D., P.E. Professor of Electrical Engineering School of Electrical Engineering and Computer Science College of Engineering and Computer Science University of Central Florida Orlando, Florida 32816-2450 "A new idea is first condemned as ridiculous and then dismissed as trivial, until finally, it becomes what everybody knows" - William James _______________________________________________ ituvidq mailing list ituvidq@its.bldrdoc.gov http://www.its.bldrdoc.gov/mailman/listinfo/ituvidq