Jerome has turned off the validation
machinery that made the
JWSDP 2.0 samples stop working on b32 of
In his blog
he describes the problem in detail: in a nutshell, he was trying to catch improper usage of annotation
at deployment time rather than at run-time,
but JWSDP 2.0 is very unusual so that backfired.
Jerome explains that he could special-case the JWSDP 2.0 situation, but he is not sure this is
a good idea and is soliciting feedback...