On Wednesday 03 August 2005 13:54, Daniel John Debrunner wrote: > Michael J. Segel wrote: > > On Tuesday 02 August 2005 23:39, Daniel John Debrunner wrote: > >>Rick Hillegas wrote: > >>>Is this acceptable? If not, how would you like to see Derby handle the > >>>sunsetting of jdk1.3 support? > > [Sun,IBM,Blackdown dates & comments omitted] > > > Sigh, not to start a flame war.... > > Not a flame war, but the end-of-service dates of VMs are a piece of the > information for deciding on continuing 1.3 support. I work for IBM but > was not suggesting that Derby follow IBM service dates for 1.3 support, > only pointing out there there are other VMs aside from Sun's, with > different end-of-service dates. > True. To your point there are others. This is why I was suggesting a policy of n and n-1 version support, where n is the current GA release.
Lets say that version 1.5 of the JVM hits the streets prior to the next Derby release. I would suggest supporting 1.5 and 1.4. If the next release of Derby hits prior to 1.5, then using n-1 as our guide, version 1.3 should be supported. Also to your point, there are other JVMs that exist. Which ones should Derby support? You have to draw a line in the sand somewhere ... ;-) My suggestion would be to use Sun's version as a benchmark on which version levels to support. -- Michael Segel Principal MSCC (312) 952-8175
