Re: [Dbound] Draft problem statement and IETF "bar BoF"

[Casey Deccio <[e-mail]>]

On Wed, Nov 5, 2014 at 11:36 AM, Andrew Sullivan <[e-mail]>
wrote:

> On Wed, Nov 05, 2014 at 10:58:39AM -0500, Casey Deccio wrote:
>
> > identification.  I don't think there is a question about whether or not
> > there so-called public/private boundaries currently exist in the domain
> > space.
>
> Actually, this is to me precisely the question.  I have been arguing
> that the public/private distinction is actually a miscategorization of
> the problem.  It's a good rule of thumb in lots of cases, but the
> corners where the distinction breaks down are instructive.
>

My feeling is that the problems with miscategorization are often because
this categorization is the *only* one that has been used predominantly
(outside of simple ancestral relationships), not because there aren't such
things as public/private boundaries.

Or... The misuse of a hammer does not mean that we need to reevaluate
whether striking it on nails is appropriate.  Rather, it means that we need
to reevaluate the uses of the hammer and the goals that we are trying to
accomplish with it.


>
> For instance, some universities have delegated name spaces to
> departments.  Is this a public registry, or a private one?  Well,
> neither, exactly.  In some sense, this is a delegation across
> organization boundaries, and the policies in one are most assuredly
> not the policies in another.  In some other sense, they're all part of
> the same legal entity.  Similarly, I really don't think it is a good
> thing if information leaks between different government departments,
> yet in some sense everything under gc.ca is in the same organization.
> Or anyway, Stephen Harper wishes it were.
>
> These cases show that the public/private distinction is not actually
> clean enough.
>

The draft problem statement document distributed to the list describes
public/private boundaries as only one property of domains.  It also
considers (for example) intra-scope relationships, and provides an example
nearly identical to the one you provided above (see section 1.1.4).

I am not arguing that public/private boundaries are *the* way to solve
things, nor am I suggesting that the definition of "public" is sufficiently
clear, as currently defined.  But to me there there is an apparent
distinction between a delegation from, say, "org" to "foo.org" and a
delegation from "foo.edu" to "cs.foo.edu".  In the proposal that would be
the difference between "public -> private" and "private -> private".
Inferring more than that is currently undefined and is referred to in the
third bullet of the problem statement (section 4).

Cheers,
Casey