Re: no names allowed, we serve types only
From: Nilone <reaanb_at_gmail.com>
Date: Thu, 18 Feb 2010 02:12:58 -0800 (PST)
Message-ID: <fc8425f8-61ff-438b-8fb4-7c528423696d_at_g11g2000yqe.googlegroups.com>
Date: Thu, 18 Feb 2010 02:12:58 -0800 (PST)
Message-ID: <fc8425f8-61ff-438b-8fb4-7c528423696d_at_g11g2000yqe.googlegroups.com>
On Feb 18, 9:51 am, Ben Finney <bignose+hates-s..._at_benfinney.id.au>
wrote:
> Nilone <rea..._at_gmail.com> writes:
> > Your idea doesn't eliminate attribute names, it just delegates it to
> > the type namespace.
>
> Exactly. Given that, I don't see what is gained.
Nor do I, but I'll accept any definition which doesn't fundamentally break the model. I'm thinking "someone might implement it this way" - being flexible about the requirements could make it happen sooner. Received on Thu Feb 18 2010 - 11:12:58 CET