Re: [Fwd: Re: [VOTE] [Discuss] P5EE without Repository and Widget]

James Duncan writes:
> But who is going to market P5EE?

No one, but it still has to have a name.  P5EEx::Blue is just too hard
to pronounce.

> is all valid stuff -- I don't think however, that it can become the
> signed, sealed and certified P5EE.

Who has the authority to sign, seal and deliver any Perl code except
the core?

> Ask yourself, would *you* risk your business on P5EEx::Blue right now? 
> If the answer to that is anything less than 100% yes then it is not
> ready.

Yes, but I won't, because we already rolled our own.  We have no
intention of switching.  This doesn't mean that P5EE is a bad idea,
rather it is an idea worth trying out for people who haven't already
rolled their own.

Currently most Perl modules I download from CPAN and trust my business
to (whatever that means) are:

    * buggy
    * specify arbitrary, inconsistent, and changing APIs
    * do not provide an "enterprise class" architecture

I still use them, because I can fix those parts which are really
broken, workaround (wrap) stuff I can't fix, and ignore the others.

> Sure, I'm not going to disagree with that either. So, IMHO, rather than
> blessing something prematurely as Sun did, lets try a little harder to
> get it right.

There is no "right".  Please name one "right" module in CPAN.
However, CPAN is being used for enterprise class applications all over
the world.  CPAN modules are "good enough" according to one's
definitions of "good" and "enough".  Where it isn't, people invent
something new and possibly share it.

I didn't put bOP up as a candidate for P5EE, because I don't want to
spend time satisfying other people's desires.  At the same time, we
released bOP as an option for people building enterprise class
systems.  If they "get" bOP, they'll use it.  If not, they won't and
maybe turn to P5EE which Stephan is willing to document better and
customize to people's needs (we have proof of this already).

That's why Stephan should be awarded the moniker P5EE.  It's pretty
much why J2EE is J2EE and RogueWave (among others) isn't.  You have to
be willing to put in the energy to evolve (and market) your system to
match divergent requirements.  I believe Stephan is willing to do this
(don't ask me why ;-) and that's why I voted yes.

Rob