I’m very happy to see this. Thank you Ovid and TheDamian and stvn and everyone else. A fast implementation of meta-object programming in the Perl core? Way cool.
I think we can afford to focus on the positives right now; it’s too early to say that there will be long-term negatives. It’s early days, and whatever the first results of Corinna are, they’ll be built upon. I think Ovid & Co. have been very up-front about the plan to start somewhat smaller and limited and why it has to be that way. Think about where we might be in 5 years. If this project goes well, what will the active Perl world be doing for OO? I suspect it’ll be a mixture of:
Building software directly on what Corinna has become by then.
Building software on one or more new, extended frameworks that use Corinna under the hood (in somewhat the same way as our existing frameworks use “Classic” Perl OO under the hood).
Building new software and maintaining existing software on Moose/Moos/Moo/Mo, Class::Accessor and friends, Class::MethodMaker, Object::InsideOut, Object::Tiny, Class::Prototyped, Class::YourFavoriteHere, and whatever else people feel like using, because all that’s still going to be around.
I feel there’s no reason any Perl fan should be worried about Corrina. Corinna isn’t going to harm Moose or Moo and the many codebases that use them, any more than they harmed Class::Accessor and friends. Classic Perl OO will never go away.
This isn’t to say that I love the current MVP design. I don’t. But that’s OK. (For one thing, nobody asked me.) If Corrina is a success, it will naturally — if not in Core-inna then by way of CPAN extensions — come to somehow support most of what is in the mainstream of current development practice. That’s just the way of things, but also I suspect that only a minority of developers are quite as opinionated as the MVP design is. However, if Corinna proves to be a rock-solid base with speed and power, then other developers will build what they want on that base if Corinna itself doesn’t have what they want. Those extensions will still garner all the advantages that Corrina brings, besides what they (opinionatedly) reject. What’s not to like?
I’ll reply elsewhere with my armchair objections to bits of the MVP.
In reply to Re: The Corinna RFC for getting modern OO into the Perl core is taking shape
by Radiola
in thread The Corinna RFC for getting modern OO into the Perl core is taking shape
by Ovid
For: | Use: | ||
& | & | ||
< | < | ||
> | > | ||
[ | [ | ||
] | ] |