Re: [boost] Re: Divergent Paths in Boost?

From: Rob Stewart (stewart_at_[hidden])
Date: 2003-11-03 12:50:29

From: David Abrahams <dave_at_[hidden]>
> Rob Stewart <stewart_at_[hidden]> writes:
> 
> > Every review of late seems harsher than the one before wrt
> > documentation, for example.  More and more discussion centers
> > around how the proposed idea will look when moved to namespace
> > std.  (That's always been in the back of many folks' minds around
> > Boost, but it is quite upfront now.)
> >
> > Many Boost libraries are well on their way to being incorporated
> > as is or with relatively minor changes into the standard.  I
> > think this has given many a sudden reality check that is being
> > passed along to others, even if inadvertantly.
> >
> > The review process has gotten to where the hapless first-time
> > library submitter is pummeled with demands for stellar
> > documentation and uncompromising implementation, even before the
> > submitter knows whether the library will be accepted.  Perhaps my
> > own perception is colored, but these things seem harsher of late.
> >
> > Reviews of later versions should, rightly, be demanding, but that
> > seems to apply to all reviews anymore.
> 
> Your characterization of the review process suggests that it is
> negative, confrontational, and more painful than helpful to the
> reviewee.  Boost reviewers, in my experience, neither pummel with
> demands, nor deliver harsh criticism.

I am being too dramatic, I suppose.  It's not that the comments
are without merit or are presented in antagonistic language, but
that they are many and are often about relatively small things
that an implementer really shouldn't have to worry about until
*after* the library has been accepted into Boost.

Going back to my earlier suggestions, if the documentation is
sufficient to describe the library, and reviewers understand it's
purpose and function, then the library could be accepted
provisionally.  Then, with that step behind him, the submitter
can focus on the technical issues raised during the review, bring
the documentation up to snuff, and then submit the library to a
final review.  Upon acceptance from that review, the library
could then be included in Boost distributions.

> One recent example comes to mind: Robert Ramey's serialization
> library.  While some of his reactions during the review seemed to
> support your point of view, I think his return with a new design that
> meets with nearly unanimous approval proves that the review process
> is working.

Wouldn't that review have gone better with the approach I've
suggested?

-- 
Rob Stewart                           stewart_at_[hidden]
Software Engineer                     http://www.sig.com
Susquehanna International Group, LLP  using std::disclaimer;