> I'd like us to vote on releasing version 1.0 of the HTTP Client component.
I'm +1 on establishing a release plan for httpclient and moving quickly to
execute it, but I think I'm -0 for releasing it as it stands (with or
without log4j support/dependencies), for the following reasons:
a) Over the past couple of weeks there have been several significant fixes
implemented--either correcting the behavior or clarifying the proper
behavior, depending upon which way you look at it--including the behavior of
query-string versus input parameters, the use of SSL support, the behavior
of sessions, etc. I wonder what else might be lurking out there, and I
suspect we haven't cleanly and clearly specified the "contract" of
httpclient and how it should behave.
b) On a related note, there is literally no documentation for httpclient
outside of the java code itself, and what we see there is fairly sparse. I
would argue that httpclient is complicated enough to warrant some decent
documentation in order to maintain a standard of quality. I think this
applies to both end-user documentation and specifications of the proper
(expected) behavior for development.
Don't get me wrong, I think we could get to a 1.0 release in a week or so,
but I'm not sure that at this point the scope, use, contract or
freedom-from-defects of httpclient is well enough defined or documented to
warrant a 1.0 release. (I should point out that we use httpclient literally
continuously in a critical production application, so I think I'm as eager
as anyone to get to an official 1.0 release, I just think we should do it
right.)
- Rod
