RFC 1480 - does it generalize (was What Problem; was ICANN/Pandora)

Jay R. Ashworth jra at baylink.com
Tue Jul 1 09:05:37 CDT 2008


On Mon, Jun 30, 2008 at 08:40:15PM -0500, Joe Greco wrote:
> > On Sat, Jun 28, 2008 at 08:46:33AM -0500, Joe Greco wrote:
> > > Yes.  It completely marginalizes the remaining positive qualities of the
> > > Domain Name System as a way to find things, in the name of giving people
> > > "more options."
> > 
> > The Domain Name System is not now, and never has been, away to *find*
> > things, anymore than 123 Elm St, Worcester MA is a way to *find* a
> > house.
> >
> > It's a way to *denote* things, uniquely.
> > 
> > You *find* an address by looking in a map directory, and then on the
> > map.  You find things on the Internet using a search engine, and the
> > second-order derivatives.
> 
> It seems clear that the authors of 1480 did not agree with this.  This may
> have been because there were no "search engines", as we know them today, at
> that time.

Possibly.  

But the inference I draw from the fact that they only specify really
deep structure for things which are municipal and related such items,
is that they were trying to provide for the delegation of reputation
that I allude to later.

> > > In others, it's ridiculous - why the heck do I get someplace in California
> > > when I go to "martyspizza.com", rather than our local very excellent pizza
> > > place?  (sadly this example is less effective now, they managed to get
> > > "martyspizza.net" a few years back).
> > 
> > Sure.  Local collisions are inevitable.  Blocker Transfer, a local
> > moving company client of mine, wanted to register a domain back in
> > 1997... when the company was 99 years old.  blocker.com was taken.
> > 
> > They took blocker100.com, and promoted it.
> 
> That doesn't seem to be a "local collision," except insofar as the Internet
> can be collectively considered "local."  In any case, company's solution is
> only useful if you already know the domain name (or have a way to find it).

Which is true of *anyone's* domain name in the DNS as we currently know
it.

But it could just as easily have been true here: if I was Westshore
Pizza (a local chain with 26 stores in about 15 cities), would people
look for -- well, I'll expand it to the top level that's pertinent -- 

westshorepizza.pinellas.fl.us or
westshorepizza.hillsborough.fl.us

or both?

Trying to shoehorn geography into the DNS *too deeply* works only for
things which are unique and not ever moving, like city and county
governments and agencies and schools.

I don't think it's necessarily reasonable to assume that the authors
of 1480 intended that scheme to be applicable generally to all domains.

We can't ask Jon, but I have carboned Ann Cooper; perhaps she'll have
some interesting input to put in.  :-)

> > > We never had any business allowing small, local businesses to
> > > register in .com, or non-networking companies to register in .net,
> > > or non-organizations in .org... but a whole generation of Internet
> > > "professionals" "knew better" and the end result at the end of
> > > the road is that DNS will end up being almost as useless as IPv4
> > > numbers for identifying the more obscure bits of the Internet.

> > Correct; this is exactly the problem.  But a lot of it stems, Joe, from
> > the misconception you led with.
> 
> Not a misconception.  Simply a different view of how the system could have
> evolved, had it followed the lead of RFC1480, instead of having everyone
> and their brother load on in to .com...
> 
> It could be said that search engines doomed any hope of trying to make the
> DNS sensible.

Perhaps.  But I've been on the net almost as long as DNS: got my first
Usenet account in 1983 and still remember the day I stopped being able
to read the entire feed.  (Well, the entire feed SPJC got over our 1200
baud modem that ran continuously.  :-)

I think that DNS started to break well before Alta Vista showed up.

> > > I look forward to many more years of having to remember that Marty's
> > > Pizza is "martyspizza.net" instead of "martyspizza.brookfield.wi.us", 
> > > that Milwaukee's Department of Public Works is at "mpw.net" instead of
> > > "dpw.ci.milwaukee.wi.us", etc.
> > 
> > I am, in turn, very pleased with a lot of my local municipalities.
> > 
> > Some of them, admittedly, *have* silly pinellascounty.org or
> > pinellas-park.com names, but they also answer to the long-form .fl.us
> > names you would prefer.  Sometimes they redirect one way, sometimes the
> > other; sometimes each domain merely overlays the other.
> 
> That's fine, I have been wishing for DNAME support for years.  Locally, in
> the pre-1480 days, we were "mil.wi.us", then became "milwaukee.wi.us", but
> only through the magic of duplicating everything between the zones, and in
> the more recent "delegate elsewhere" model, that means that some registrants
> invariably only configure one or the other, so "foo.mil.wi.us" works while
> "foo.milwaukee.wi.us" doesn't...

In fact, the Hillsborough County Sheriff's office has
hcso.hillsborough.fl.us on the back of the cars, bless them.  That's a
really *strong-minded* IT director; I ought to find him and shake his
hand.  

Do we have RFC 1480-Champion badges somewhere?

> > But at least they are, as you say, deterministic.
> > 
> > I don't think it's fixable anymore, either.  But I remain determined to
> > spit into the wind, Jim notwithstanding.
> 
> :-)

I decline to mess with the old Lone Ranger, though.

Cheers,
-- jra
-- 
Jay R. Ashworth                   Baylink                      jra at baylink.com
Designer                     The Things I Think                       RFC 2100
Ashworth & Associates     http://baylink.pitas.com                     '87 e24
St Petersburg FL USA      http://photo.imageinc.us             +1 727 647 1274

	     Those who cast the vote decide nothing.
	     Those who count the vote decide everything.
	       -- (Joseph Stalin)




More information about the NANOG mailing list