This comment came in response to my little survey; I thought it was worth reading. --MM Frank O'Connor wrote: > G'day Milton, > As a lon time member of ISOC and a past contributor to various IETF working > parties I've got a couple of comments to make about the current ICANN > debate. Much of the debate seems to be focusing on how ICANN is > administered and its structures and powers. > > From a technical perspectibe the following seems obvious to me: > > 1. What we are talking about here is network administration ... not a > political exercise, not an exercise in national influence or whatever. > > 2. Networks rely on agreed to standards to function. Such standards must be > technically correct if they are to function efficiently and effectively. > > 3. Networks require some sort of centralised control to function. Without > this centralisation, networks can fragment into a morass of incompatible > standards, sub-networks and mis-placed ideals that have little to do with > the efficient running of the medium. In other words ... without some form > of centralised network standards and resource control you risk throwing out > the baby with the bathwater. > > 4. The Internet functioned quite well in the past under the ambit of > various organisations each with well defined roles to play ... ISOC and the > NSF, IANA, the IETF, the NAB etc. > > 5. Broadly speaking ICANN in its current format is a continuation of the > IANA with added responsibilities for IP number allocation and domain name > controls. It is non-profit, it is responsible for efficient network > administration, and it is centralised. There is some provision for outside > representation and non-technical representation, but the overally thrust is > for a continuation of the roles of the bodies which operated previously ... > and which made the Internet such an efficient medium for the backbone of a > world wide WAN. > > 6. Generally speaking the NSF has been removed from the ambit of the > equation (given that the US government no longer funds the Net), but the > other bodies mentioned above survive and have active and vital input into > the development of the Net. One thing that does depress me about ICANN is > that those roles weren't defined in its Articles. > > 7. That said, although ICANN should accept input from other bodies, ICANN > can't be run on a consensus (political) basis, in that a number of bodies > which have now appeared on the Net have their own interests to push. > Standards must remain open. Standards must be enforced (with respect to the > Internet) and enforceable by a single body. What one uses on their own > intranets and extranets (read VPN's) is their own business ... but for > those expecting to use the worlds WAN they should adopt the standards > agreed to by the technicians (and users). Adding another level of > decisionmaking to this inherently rational process (eg. political, > national, proprietary or whatever interests) essentially defeats this > purpose. > > In this regard I feel that one of the reasons the ISO and ITU failed with > OSI (TCP/IP's competitor in the past) is that they allowed politics and > pressure to dominate their decisionmaking. Hence OS became a bloated, dated > and fundamentally incompatible setup of various subnetworks which relied on > (and tolerated) closed proprietary standards and extensions that were > pushed by various members of the OSI standards making bodies (read big > metal computer manufacturers, software proprietors and telecommunications > vendors). In the TCP/IP world we already have the beginning of these > political and proprietary blocs ... the World Wide Web Conference for > example ... which could destroy the homogenous and universal nature of the > Net as it stands. > > Personally I would have preferred the continuation of the old IANA scheme > with some redefinition of the role of Network Associates (given that there > was an inherent conflict of interest inherent there), but the idea of ICANN > seems to be a reasonable compromise to a situation that was generated > substantially by political concerns (read Ira Magaziner and the lads, the > US and EC governments, the UN and the need for the ISO and ITU to remain > relevant and survive in a TCP/IP world) rather than the needs of the > network itself. > > So after that rather longwinded response ... I suppose I'd have to come > down on the side of your option 3. :-) > > Regards, > ***************** > Upgrade to Earth 2.0.2. > *****************
