Dear all
As pointed out by Suzanna Widmer (and others) the consideration of when a new softwareis sufficiently ready for courses and "real projects". Experience tell me to say "never use the first release of new or adapted software" - since regardless of a lot of testing it always turns out that only in real projects will one find the problems and bugs. Test situations are always less demanding than real projects. On the other hand I tend to be optimistic with EpiData software since with my 30 year experience in teaching data quality and field epidemiology I can find a workaround and also I can assure that we will fix the bug to next class. Most other instructors have less experience and cannot directly promise bug fixes. Another rationale is that instructors are experienced in the old software, but not in the new.
As part of the EpiData-Apheo project which adds financing to the current rewrite (see http://www.apheo.ca/index.php?pid=47) there will be a teaching seminar in february in Canada. For this the planning group decided to base the seminar on the existing software, but also to include a shorter presentation where I explain and show the new software. This is a relevant way of tackling the transition. Had the same seminar been held in april or may I would have suggested to do the other way around. That is to use the new software and then show the older one, since there might be aspects which are still not functional.
Here in the office at the hospital where I work we will start to use the new versions for smaller projects. But not do so before we have the next test release ready - which includes correct updating of idnum, autotime and autodate fields and jumps.
In seminars and courses the most important is a precise principle and sound scientifically based concepts - and functional software which supports the principles without giving participants and instructors headache's due to malfunction. Many participants in courses cannot separate problems associated with lack of understanding principles and lack of practical experience in using the software (or computers).
Excuses for the long explanation. I hope the considerations can assist in the decisions on what to do in the transition period. I am very happy to see that the download counts of test versions implicate wide spread consideration of "what is coming".
regards Jens Lauritsen EpiData Association