[standards-jig] Re: reference implementation? (was: Re: [rpc-jig] JEP-0009)

Iain Shigeoka iainshigeoka at yahoo.com
Wed Jan 23 18:27:39 UTC 2002


On 1/22/02 4:23 PM, "Peter Saint-Andre" <stpeter at jabber.org> wrote:

>>   No what you really want is at least 2 or 3 interoperable implementations
>> for every features in the protocol. That's the IETF rule. I would like
>> to also see a rule like one of them at least being opensource.
> 
> AFAIK the IETF requires 2 interoperable implementations. I too like the
> rule that one of them must be open-source (in conformance with the Open
> Source Definition: http://opensource.org/docs/definition_plain.html).

+1  Since the point of the Jabber standards is to improve Jabber
interoperability, we would be best served with at least this requirement.
Of course, saying we have 2 interoperable implementations also means that
Jabber standards must include specific, testable acceptance criteria to
judge compliance and "passing" interoperability...  This would probably need
to include some agreed upon test suite including test data...


_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com




More information about the Standards mailing list