Re: should GOOPS be built-in?

[Greg Harvey <Greg dot Harvey at thezone dot net>]

Mikael Djurfeldt <mdj@mdj.nada.kth.se> writes:

> Lalo Martins <lalo@webcom.com> writes:
> 
> > > Nope.  I think it is useful to merge it now.
> 
> Maybe I should be a bit clearer: The reason why it's useful is that,
> as soon as it has been merged (which basically amounts to 1. fixing
> copyright asigment with Erick, 2. copying the GOOPS file into the
> distribution, and 3. placing an init call at the appropriate place)
> work can commence on several things, one being developing a better
> representation than the structs, and, eventually, replacing them.

Wouldn't that be a bit heavyweight? Structs are heavy enough now, but
a goops class would be a bit heavier, as far as I can tell (which
isn't to say that structs couldn't use replacing). One thing that
really should happen before goops gets into the core is some good
goops documentation; getting documentation for what's already there is
going to be hard enough without adding yet another large, mostly
undocumented change (actually, this goes for anything large, not just
goops).

> With the current, separate distributions, things are a bit locked up
> due to compatibility issues.
> 
> I have no plans to do these things, however, but I will try to get
> time to do something about method compilation during Christmas.
> 
> > > > > Gregor Kiczales' "The art of the metaobject protocol" is excellent.
> > > > 
> > > > Where? Is this free? Can it be distributed with GOOPS?
> > > 
> > > This is a book which you will find at, e.g., Amazon.  Chapter 5, which
> > > contains the spec of the MOP for CLOS, can be downloaded via
> > > Kiczales' homepage at Xerox PARC.
> > 
> > So we need at least a brief _free_ intro to be distributed with
> > GOOPS. If possible, written by someone who read this book.
> > (Sorry to be playing RMS here, but we just agreed that
> > documentation is one of the major points to work on. And well,
> > I actually _like_ RMS, hehehe)
> 
> You're right.  Who will do it?

Someone who wants goops in the core enough to write it, I guess.

-- 
Greg