Global council of registries???

[Philip J. Nesser II]

Jim Fleming supposedly said:
> 
> 
> All of the allocations would not come from
> the MIT registry. I guess that was not obvious.

Your statement of:
> @ > Would MIT be able to carve those allocations out of
> @ > the /8 space it has and set up a system to help educate
> @ > ISPs and to make the allocations to the ISPs ?
> @ 

was what confused me.

> 
> If there are 50 InterNIC clones created then each
> clone would only have to make 60 of the /18 allocations
> to the ISPs. That should not be a difficult side-line
> to add to one of the operational TLD registries that
> are starting to appear.
> 

But why?  This is a function that can easily be handled by a central
entity.  I don't believe in making busy work to help launch a new
industry.  The Internet has naturally created numerous business
opportunities, why artficially create another one?

> @ > 
> @ > Would the Internet be better served by having better
> @ > educated ISPs ?
> @ 
> @ Why should anyone pay to educate ISP's?  They are businesses.  They need to
> @ educate themselves and pay for their training.  Hey, I would prefer that my
> @ auto mechanic was better trained and I would be better served by them
> @ becuase of it, but should the government pay for it?  No.  Should tax
> @ dollors pay for it?  No.
> @ 
> 
> OK...thanks for your viewpoint...
> I wonder if the reverse is true ???

I can't figure out the reverse of my statement?  Of course I will benefit
if my ISP is well informed and trained, but that is a business decision of
the ISP.  If one ISP hires more competent engineers who design a better
infrastructure then they probably pay for the better service, but
conversely have higher reliablity and hopefully generate better profits
because of it.  I don't think the government should try and subsidize
ISP's.  Hiring cheaper help is a business decision and may allow ISP's to
have higher profits, but it may not.  Thats a free market and thats what
the ISP market should be.


> 
> @ > 
> @ > The entire IPv4 address space must be evaluated
> @ > as one space. The same rules should apply to the
> @ > entire space as much as possible.
> @ 
> @ I disagree.  I think any allocations made within a given time frame should
> @ be exactly similar but we can't change history nor can we forsee the
> @ future.  IPv6 may catch on quickly and all of the scrimping of IPv4 space
> @ will have been useless becuase it isn't needed except for legacy systems.
> @ We need to do the best we can go forward.
> @ 
> 
> I think that we need to agree to disagree.
> IPv6 does not solve the routing problem
> and IPv4 is going to be here a long time.
> There are solutions that recognize this.

IPv6 will not solve the routing problem, but will solve the address
conservation problem.  The problem we face now is a doubly constrained
situation (scarcity of addresses and routing prefixes) both of which
optimize at two different extremes(For assignment efficiency we would only
have /32 routes and every IPv4 address would be used, while for routing
efficieny we want to topologically group addresses on bit boundaries into a
31 level tree.).  In IPv6 we can effectively ignore address scarcity which
makes the problem easier to solve.

> @ > 
> @ > BTW, have you ever computed what a small percentage
> @ > of the space that ISPs actually have ?
> @ 
> @ Your point?  
> @ 
> 
> The point is that ISPs seem to be accused of
> exhausting the IPv4 address space. This is just
> not the case.
> 

True, ISP's only have a small percentage of the total space, but they are
the fastest growing users of address space so it is certainly an area to
watch.  

> @ > Have you computed the costs to ISPs (businesses)
> @ > for all of the InterNIC run-arounds they have endured ?
> @ > Who is going to pay for those costs ?
> @ 
> @ Have you computed the costs to businesses for all of the government
> @ run-arounds they have endured?  Who is going to pay for those costs?  
> @ 
> @ Have you computed the costs to businesses for all of the parts supply
> @ run-arounds they have endured?  Who is going to pay for those costs?
> @ 
> @ The point being, there is a cost for doing business.  If you can't handle
> @ it then you go out of business.  Once again I don't believe that the
> @ government should fund the inadequacy of an ISP.
> @ 
> 
> I wonder if you feel the reverse is true...?

Once again I don't understand what you mean by the reverse.  I run a
private consulting company and bill by the hour.  I spend countless hours
that I could be billing waiting in Doctors offices when I go in for a
checkup, in the Division of Motor Vehicles when I renew my drivers license,
in the state offices when I get my business license, filling out government
paper work etc.  I can't bill any of those hours to anyone, but they are a
price I pay for living(in some cases) or choosing to do business(in other
cases).  

> 
> Thanks for your comments. To summarize,
> I suspect you do not mind if all of the businesses
> that are trying to survive, just move forward and
> get on with the tasks in front of them.

Of course.  I work hard to make my business survive.

> 
> If this includes the creation of many new companies
> in the Registry Industry to handle all aspects of
> Internet Resource allocation, I assume that you
> will not mind. After all, you more or less said they
> are on their own and will get no help from you.

This is where we once again disagree.  I do not believe that an industry is
needed for these services and since I believe that such an industry will
only do two things:  endanger the stability of the Interent infrastructure
and drive up costs, I will oppose the creation of such a false industry
since it does not create any gain except for a few while delivering no
better service to the public.  In my mind it is no more than an attempt to
tax the Internet to line the pockets of a few.

--->  Phil