Re: NetSol screwing the pooch?
NSI's been screwing the pooch so long, it's surprising we're not all knee deep in puppies!
I find it amusing that large ISPs tolerate NSI for so long despite the enormous costs its incompetence imposes on their businesses. Now the bunch of self-proclaimed rule-setters called "ICANN" is going to screw them even more. I would think that these large ISPs should form the consortium and finally build a normal registry database; they already have infrastructure for customer service, and offer their customers assistance with domain registration. There is no need for independent registries. Customers wishing to register a domain should simply talk to their ISPs whose personnel can then go and update the consortium-owned database. Oh, and BTW, they _already_ have the contact info for their customers, so they do not depend on the NSI-claimed "ownership" of that information. And, unlike NSI or other registries, they can actually verify that information - and they have a pretty simple method of dealing with abusive customers. This approach scales, it can improve domain-related customer service by an order of magnitude, and it does not require any political games. Despite the cut-throat competition in the market, ISPs managed to maintain a coherent and functioning global routing infrastructure. The DNS is as essential for their customers as the actual packet transport; so i think it is time for competent people to take over it, too. Now, the question is when the ISP managers will wake up and get a clue. --vadim
On Thu, 13 Apr 2000, Vadim Antonov wrote:
There is no need for independent registries. Customers wishing to register a domain should simply talk to their ISPs whose personnel can then go and update the consortium-owned database. Oh, and BTW, they _already_ have the contact info for their customers, so they do not depend on the NSI-claimed "ownership" of that information. And, unlike NSI or other registries, they can actually verify that information - and they have a pretty simple method of dealing with abusive customers.
This approach scales, it can improve domain-related customer service by an order of magnitude, and it does not require any political games. Despite the cut-throat competition in the market, ISPs managed to maintain a coherent and functioning global routing infrastructure. The DNS is as essential for their customers as the actual packet transport; so i think it is time for competent people to take over it, too.
Problems: 1. You'll have two databases, one maintained by consortium, and one maintained by NSI/other registrars. I suppose you can say 'consortium data takes priority' on consortium's root servers, however, what will happen when data begins to diverge? Example: Domain expires in NSI db, gets deleted, someone else takes it in NSI database. You have then to mirror NSI changes...Its gets very ugly very fast. 2. Last attempts to do things like these (by edns, alternic) were fraught with personality clashes among its founders, and ended badly (servers brought down, and ISPs who were persuaded to point their roots at these servers had to back out the changes). Of course, one can say that they failed because they tried to _extend_ the namespace, not just manage it. Who knows... 3. The other thing is that anything involving DNS governance is a dirty business, and everyone who attempted to get involved in it doesn't want to touch it again :) (I'm sure Randy will make comment here ;) -alex
[ On Thursday, April 13, 2000 at 23:05:33 (-0400), Alex Pilosov wrote: ]
Subject: Re: NetSol screwing the pooch?
Problems: 1. You'll have two databases, one maintained by consortium, and one maintained by NSI/other registrars. I suppose you can say 'consortium data takes priority' on consortium's root servers, however, what will happen when data begins to diverge? Example: Domain expires in NSI db, gets deleted, someone else takes it in NSI database. You have then to mirror NSI changes...Its gets very ugly very fast.
"Divergence"? "Priority?" The only people it gets "ugly" for are those who would try to still follow NSI and NSI's root and tld servers. Everyone in the consortium, and presumably everyone who points their nameservers at the new root servers, would be quite happy I think. This of course assumes that most everyone who provides connectivity of any kind joins the consortium, and of course that everyone who buys connectivity turns their DNS over to the consortium. A large and truly non-profit consortium could work if almost everyone were to simply turn their backs on NSI (and presumably that part of IANA which gives NSI and the current root servers their authority). Whether this can happen politically or not is highly questionable. It would require truly global co-operation of a very large number of commercial entities. Of course the process is relatively simple so if the ball got to rolling there's no telling how far it might go. If too many profit and power seekers start staking their ground without there first being an independent overseer then it'll certainly stop as fast as if it were a cube and not a ball though.
2. Last attempts to do things like these (by edns, alternic) were fraught with personality clashes among its founders, and ended badly (servers brought down, and ISPs who were persuaded to point their roots at these servers had to back out the changes).
Yup, when you pull down from underneath lots of things tend to fall on you. However if a large bunch of organised people were to try to pull up from the top we might just lift ourselves by our bootstraps! ;-)
Of course, one can say that they failed because they tried to _extend_ the namespace, not just manage it. Who knows...
Well, there's that little issue too.... (hinted to by my comment above about crazed DNS prospectors)
3. The other thing is that anything involving DNS governance is a dirty business, and everyone who attempted to get involved in it doesn't want to touch it again :)
And that one as well! ;-) -- Greg A. Woods +1 416 218-0098 VE3TCP <gwoods@acm.org> <robohack!woods> Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>
The problem isn't ICANN, the problem is the USG giving NetSol legally binding (in the US) exclusive rights. Recently, another 4 years.... Of course, the ISPs could host a revolution. You have some experience with that. Vadim Antonov wrote:
I find it amusing that large ISPs tolerate NSI for so long despite the enormous costs its incompetence imposes on their businesses. Now the bunch of self-proclaimed rule-setters called "ICANN" is going to screw them even more. ... Despite the cut-throat competition in the market, ISPs managed to maintain a coherent and functioning global routing infrastructure. The DNS is as essential for their customers as the actual packet transport; so i think it is time for competent people to take over it, too.
Now, the question is when the ISP managers will wake up and get a clue.
WSimpson@UMich.edu Key fingerprint = 17 40 5E 67 15 6F 31 26 DD 0D B9 9B 6A 15 2C 32
On Thu, 13 Apr 2000, William Allen Simpson wrote:
The problem isn't ICANN,
I suggest you look more closely at what exactly they have produced. /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Patrick Greenwell Earth is a single point of failure. \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
Patrick Greenwell wrote:
On Thu, 13 Apr 2000, William Allen Simpson wrote:
The problem isn't ICANN,
I suggest you look more closely at what exactly they have produced.
With regard to operational issues, I have followed the general proceedings of ICANN. The results seem to be well considered. For example, the resolutions of .pn and .ps are reasonable and documented. Other than the USG agreements with NSI, which ICANN is required to follow for operational continuity, I have found no operational decisions by ICANN. Perhaps you could be more explicit? The issue raised here is that NSI is operationally incompetent. Wresting the whois database and name server operations from NSI will be difficult, even with concerted action by ISPs. As an example in a related communications industry, I offer MediaOne cable service. MediaOne and predecessors have been in violation of their franchise agreement for over 20 years. They don't deliver service in locations where it is expensive to install, including downtown Ann Arbor, despite explicit contractual requirements. They have never met federal or local requirements for customer support, both for timeliness of resolution and reporting of results. Yet, it has proven impossible to deny them franchise renewal. So far, only two such denials have been upheld by courts, both in Kentucky. Attorneys for the City of Ann Arbor advised the cable commission that denying franchise renewal would be prohibitively expensive and ultimately fruitless given the existing legal climate. I conclude that, if the major ISPs desire a change, they will need to work together to move the DNS registry service outside of the US, and be willing to defy unreasonable and irrational US court orders. How do you propose to make it work? WSimpson@UMich.edu Key fingerprint = 17 40 5E 67 15 6F 31 26 DD 0D B9 9B 6A 15 2C 32
On Fri, 14 Apr 2000, William Allen Simpson wrote:
Patrick Greenwell wrote:
On Thu, 13 Apr 2000, William Allen Simpson wrote:
The problem isn't ICANN,
I suggest you look more closely at what exactly they have produced.
With regard to operational issues, I have followed the general proceedings of ICANN. The results seem to be well considered. For example, the resolutions of .pn and .ps are reasonable and documented.
Other than the USG agreements with NSI, which ICANN is required to follow for operational continuity, I have found no operational decisions by ICANN. Perhaps you could be more explicit?
The UDRP(http://www.icann.org/udrp/udrp-policy-24oct99.htm).
The issue raised here is that NSI is operationally incompetent.
The "cure"(ICANN) is far worse than the illness.
Wresting the whois database and name server operations from NSI will be difficult, even with concerted action by ISPs.
There is no "wresting" to be done. ISPs have a choice where they point their servers, period. That they do not avail themselves of this capability serves to illustrate that it is not important enough to ISPs as an industry.
I conclude that, if the major ISPs desire a change, they will need to work together to move the DNS registry service outside of the US, and be willing to defy unreasonable and irrational US court orders.
See above.
How do you propose to make it work?
There have been several proofs-of-concepts(alternic, eDNS, etc.) that demonstrate the technical feasbility of such a change. Unfortunately, these previous attempts have been conducted by (mostly)well-intentioned technical folks who lacked the business and/or political acumen necessary to be successful. /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Patrick Greenwell Earth is a single point of failure. \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
On Fri, 14 Apr 2000 bmanning@vacation.karoshi.com wrote:
On Thu, 13 Apr 2000, William Allen Simpson wrote:
The problem isn't ICANN,
I suggest you look more closely at what exactly they have produced.
Patrick Greenwell
and what do you think ICANN is supposed to produce?
Please refer to the USG White Paper. /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Patrick Greenwell Earth is a single point of failure. \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/
participants (6)
-
Alex Pilosov -
bmanning＠vacation.karoshi.com -
Patrick Greenwell -
Vadim Antonov -
William Allen Simpson -
woods＠most.weird.com