Re: release policy

[Ralf Wildenhues]

Hi Bob,

* Bob Friesenhahn wrote on Sun, Sep 18, 2005 at 07:49:38PM CEST:
> On Sun, 18 Sep 2005, Ralf Wildenhues wrote:
> >>
> >>Development libtool offers few user-visible features where a new
> >>feature is defined by new functionality.  Most new "features" are
> >>internal design changes.

> >- no stupid C++/Fortran 77 checks for C-only users (plus several
> > hundred KB smaller configure scripts)
> >- better support for linker inherited flags (-pthread etc)
> >- better support for necessary compiler flags in CFLAGS
> > (the dlsym file is actually compiled with them)
> >- basic support for Fortran 90/95
> >- somewhat better support for mingw/cygwin, AIX
> >- ltdl may be used without autotools
> >- (hopefully soon) msvc support
> 
> Some of these are "features" while others are "bug fixes".  A 
> "feature" represents new functionality while a "bug fix" represents a 
> fix necessary for an already defined feature to work correctly.  Most 
> products come with a feature list which describes the features of the 
> product (e.g. "four doors").  Laundry detergent does not follow this 
> model since each new version of the product is improved in some 
> intangible way from the previous version and is always "new and 
> improved".
> 
> For example, the first item you list is actually a "bug fix" since it 
> was not a problem in the 1.4.X releases.  The bug was added in 1.5.

Oops.  I did not know that, I never used 1.4 in practice.  :-/
I retract that then partly, I did not know 1.5 had regressions over 1.4.
I never meant to "sell" regression fixes as new features.

> >>To me "stable" means that there are very few changes needed.  It
> >>represents a point of equilibrium.  Rather than everyone needing to
> >>run to the other side of the raft, it should only be necessary for
> >>there to be a minor change in seating position to avoid capsizing.  We
> >>should be at that point after several years of development.
> >
> >I must admit that you speak in riddles to me -- I have not understood
> >this paragraph completely.
> 
> Probably due to the language barrier.

Yep.

> It is an analogy to survivors 
> in a life-raft which is not level unless everyone sits in the right 
> position to reach equilibrium.  If everyone were to move to the 
> high-point of the raft at once, the raft would capsize rather than 
> becoming level.  The successful approach is to move one person at a 
> time until the raft is level.  The same approach should be followed 
> while "stabilizing" a software product for release.

Ah, ok.  I agree then.

Cheers,
Ralf