A Network Wizards' 1998 report was finally published on their site today:
It is an extremely interesting and useful document by its own merit, with many new parameters screened, so I'm sure it is worth your readers' attention.
On the other hand, I must note, that at least two entries in this report look seriously flawed.
According to data published, the total number of hosts worldwide has grown from July's 19,540,000 to 29,670,000 in January 1998 (duplicate hostnames excluded). Which means quite a substantial growth for the global network as a whole.
On the other hand, when it comes to RU and SU (Russian and former Soviet) hosts, the numbers show some sort of stagnation and freezing: in July 1997 there were 119467 hosts counted in these two zones, and in January only 121133 were spotted, including duplicates.
With all available Russian sources indicating an avalanche growth rate for the RU zone hosts, this makes one assume there's something wrong with NW's assessment of Russian sites. This assumption is further confirmed by NW's estimate about the number of Level 2 Domains, registered under .RU: 2790 by last count. While another source
http://www.neystadt.org/russia/sites/sites.htmlists 4182 Level 2 Domains in the .RU zone - and this source only reports those servers that respond to HTTP requests on the 80th port (i.e. are reachable, and host websites).
Moreover, when it comes to Israeli figures, Network Wizards show 64233 hosts in January 1998 vs. 61140 in July 1997.
This again leads us to check the domain stats for Israel (.IL). Network Wizards report 2139 domains for both second and third level. The Israeli Internet Guide
http://www.neystadt.org/israel/begs to differ, listing twice as much:
Again, the Israeli Internet Guide deals only with WWW-hosting domains, and these are far from 100%, since many domains host mail exchangers, provider subnets and corporate intranets, that are unreachable by their name from the WWW.
There must be some sort of explanation for the domain count controversies, but one thing is quite evident: the "stagnation" in both Russian and Israeli hostcount has more to do with Network Wizards' methodology, than with the effective growth of the Net in Russia and Israel. A pity, since many international researchers use this survey to base their estimates on regional Net growth.
With best regards, - Dr. Anton Nossik, Editor
Evening Internet Daily in Russian (since 1996)
PS. The NW story will run in Russian in
http://www.cityline.ru/vi/06feb1998.htm#nwto be published this Sunday (i.e. today).