Date: Mon, 27 Jan 2003 20:31:14 +0000 (GMT) From: William Palfreman <william@palfreman.com> To: Mark.Andrews@isc.org Cc: Barney Wolff <barney@pit.databus.com>, "" <freebsd-stable@FreeBSD.ORG> Subject: Re: 4.7-R-p3: j.root-servers.net Message-ID: <20030127044456.U23956@aqua.lan.palfreman.com> In-Reply-To: <200301270358.h0R3wsEN057216@drugs.dv.isc.org> References: <200301270358.h0R3wsEN057216@drugs.dv.isc.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 27 Jan 2003 Mark.Andrews@isc.org wrote: > Did you ever here the term "natural monopoly". The DNS root is a > example of such. Yup, I'm a trained economist, so I know about natural monopolies. Or rather I don't know, because in real life they never exist. What exist are government supported monopolies, where an organisation and a government discover a mutual interest in preventing rivals from operating. I see no technical reason why this need happen to DNS. > When you try to change it all you > do is reproduce it with additional unnecessary baggage like > have to find all the "roots" to register the new TLD in. Isn't that the whole point of IRON / www.root-dns.org and alt.roots? As usual the problem has been solved according to the principle of co-operation and agreement before anyone "in authority" could "prove" this wasn't possible. > And why do you think that you should have the right to create arbitrary > new TLDs? If I had the bandwidth and the server power in redundant locations, and I had convinced the operators of the applicable roots of my technical competence, I don't see any reason why I should not operate my own TLD. I already do for the lan on this site, so really it is just a matter of scaling it up. > Hell I would like the "prestige" of my own TLD but I know > that it is *not* in the best interests of the world as a whole to have > lots to TLDs which is the natural consequence to allowing anyone to > set one up. In the end you end up with a massive root zone (similar > to the COM zone) that requires very large machines to serve it or > requires many smaller machines with a fancy front end. As I read that, you are saying that unlimited TLDs are a bad idea because that would make . as crowded as COM. Doesn't the popularity of COM go to show that in use, people want domain names to be TLDs and absolute? - So what they get is a COM, because COM has become a shorthand for a domain. Using TLDs simply for primitive load-balancing seems like the wrong approach to me, and is already being undermined by reality. > Well "za" should have known from the start that the South > African government would possible want control at some point > in time. It was quite evident from the start that governments > would eventually wake up to the fact that the Internet > existed and the naming structure included country codes. If you look at the terms of RSA proposed takeover of the za domain you can see that it would have been an absolute disaster. But it didn't come to that because *DNS need not be a function of government*. The beauty of the domain name system is that, not being a natural monopoly, and being dependent on peer support (i.e. other DNS admins), there is a natural method of preventing abuses from occurring. As far as I can see, I've yet to hear a good technical argument why there need be only one root. Possible problems with multiple roots don't seem to actually occur in real life, whereas problems with the mainstream root occur all the time - abusive behaviour by some government wannabees, abusive interference by courts in domain name ownership, ICANN, DDoS attacks, simple volume of traffic problems and so on. All of these are alleviated by multiple roots. -- W. Palfreman. Tel: +44 (0)771 355 0354 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030127044456.U23956>