www-jcp-open mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Chris Gray <chris.g...@kiffer.be>
Subject Re: [DISCUSS} Alternate Proposed Changes to JCP Participation - Round 2
Date Sat, 07 Jul 2007 10:42:40 GMT
On Saturday 07 July 2007 04:24, Bill Barker wrote:
> I mostly agree with David here.  Also, I'd like to point out that requiring
> an OS TCK would be a legal minefield for the spec-lead.  It has to be
> frozen in time, otherwise if a "bug fix" suddenly causes an implementer to
> fail the TCK, where before they passed, lawsuit-city ;-).

Sorry but this is bogus. To be certified, a particular binary artefact has to 
pass a particular release version of the test suite. That certification 
remains valid even if a later release of the test suite contains tests which 
the already-certified artefact would fail. Note: I say "test suite" because 
I'm talking about certification as an abstract process, not the policy of a 
particular company.

I would argue that open source test suites are even more valuable than open 
source implementations. Would be great to have some public debate about what 
should be tested and how, no?

Chris Gray        /k/ Embedded Java Solutions      BE0503765045
Embedded & Mobile Java, OSGi    http://www.k-embedded-java.com/
chris.gray@kiffer.be                             +32 3 216 0369

View raw message