A proposal about hostcount and DNS

>>>>> "PB" == Piet Beertema <Piet.Beertema at cwi.nl> writes:

 PB>     In the days before HTTP1.1 and virtual interfaces, it would be
 PB>     easier to count the number of physical nodes on the Internet
 PB> Not really: long before that machines could have multiple
 PB> interfaces and thus multiple A records, and based on that
 PB> counted as more than 1 physical node.

 PB>     As for private machines, running through a proxy, do we count
 PB>     these as Internet hosts, since they are not directly connected
 PB>     to the Internet and merely request proxies to contact internet
 PB>     services and relay information.
 PB> Something similar goes for hosts on private internets:
 PB> I wouldn't count those as Internet hosts.
    
 PB>     When we start talking dialup machines, it becomes more obvious
 PB>     that the count should try and focus on physical nodes.  If an
 PB>     ISP has 2,000 customers and 30 modems, only 30 hosts will ever
 PB>     be using the internet at any one time, and only 30 IP addresses
 PB>     are being used to accomodate all 2,000 people.  Therefore I would
 PB>     suggest that the ISP has 30 nodes.
 PB> Again, this depends on the purpose of the hostcount. If
 PB> you want to count the number of hosts that can be active
 PB> at any time on the Internet, then this ISP should indeed
 PB> be counted as 30 nodes/hosts. If however you want to count
 PB> the number of hosts that comprise all of the Internet, then
 PB> hosts that *can access* the Internet should be counted; in
 PB> that case this ISP should be counted as 2000 nodes/hosts.

Surely must a discution and definition of various types of numbers be
done and one or more of those be coarsly measured. If the usefullness
of the statistics can be found by changing measurement method and
definitions by which it is done, then could we go further and really
measure it that way.

 PB> But I still don't see the purpose of any attempt to give
 PB> an "exact" hostcount: what difference does it make if a
 PB> count would give 15.000.000 or 15.478.329 hosts/nodes?

I am not out to find an "exact" host count, that would be a way too
overambigous goal. But surely would diffrences like 15.000.0000
compared to say 17.000.000 or so be worth discussing. 

But then, it may very well be that there is no interest in better
numbers and then we can just forget about it.

Cheers,
Magnus