Re: Alternatives (was Re: whois broke again?)

[Jeff Mcadams <jeffm () iglou com>]

Thus spake Rodney Joffe
> > Some private messages have said that NSI claims the whois contact
> > information is now their "property".

> Not only. NSI has said publicly that this is the case. As does the data
> they return with each whois result :-)

Of course...I think this is a crock...as, I suspect, most of the people
on this list thing...but that's not terribly relevant to my overall
response...just wanted to throw my $.02 in ehre.

> ...

> A totally new from scratch database needs to be created. And it should
> *not* be me, GeekTools, or CenterGate. Or any individual. It needs to be
> controlled by a body trusted by all. What ICANN should have and could
> have been.

> ...

> So, while this discussion is interesting to some, unless the majority of
> lurkers here care enough to open their mouths and comment, we're just
> creating noise. So sad.

I agree that a whole new database should ultimately be created.  As a
sorta-lurker, I give my $.02 here.  Whois sucks.  As a protocol...its
just pathetic IMO.  Although its serviceable in its current
incarnation...its *so* simplistic as to be a joke for the most part.

I figure protocols exist to enable stuff (technical term).  The whois
protocol...being so simplistic doesn't really enable anything....  All
of the current enabling is done in the back-end databasing and display
engines.

Let's face it...what we're trying to do here is directory services writ
large.  Something like NDS, or AD (shudder), or LDAP on a grand scale.
Maybe these specific implementations wouldn't scale to what we need...I
would guess LDAP would be the closest.  I'd say we probably need to
start from there...from a decent, enabling protocol...and build the
service on that.  Or perhaps, if the current protocols don't hack it, we
need to build a new protocol and build the service on that.  I really
just think the whois protocol...as an access protocol to the database
backend, just doesn't hack it.

The addition of structure on the protocol will help do the referral, and
delegation stuff that whois lacks and *DESPERATELY* needs.  The lack of
referral capability is what really prevents that setup from being
distributed.

The other suggestion I saw here...distributing stuff via DNS...seems
like a hack (no offense to anyone...its a rather elegant hack...but
still a hack).  I tend to be of the opinion that overloading DNS with
more and more functionality is causing more problems than it solves.
Again...I think its being implemented fairly well...but the core
solution to the problem is not optimal IMHO.


Well...its early and my coffee hasn't made it into my bloodstream yet,
so I'm not following through and explaining my ideas very well...I do
think this discussion is a good one and will continue...hopefully I'll
be able to flesh out my ideas and present them better later on.  :)
-- 
Jeff McAdams                            Email: jeffm () iglou com
Head Network Administrator              Voice: (502) 966-3848
IgLou Internet Services                        (800) 436-4456