[11635] in Commercialization & Privatization of the Internet
Re: Internet fact data needed
daemon@ATHENA.MIT.EDU (Bruce Gingery)
Tue Apr 12 11:12:53 1994
Date: Tue, 12 Apr 1994 00:57:38 -0600 (MDT)
From: Bruce Gingery <lcbginge@antelope.wcc.edu>
To: wantan@sejsun1.nri.co.jp
Cc: com-priv@psi.com
In-Reply-To: <9404120224.AA04668@sejsun1.nri.co.jp>
Any answer you receive will be quite approximate, but nevertheless, I do
hope you will summarize replies back to the list. With the two distinct
addresses below, am I two users?
As for hosts, again it is most likely impossible to actually count
them. First, "other" networks which are international in their own right
such as FidoNet(sm) are "part" of the InterNet, because traffic may pass,
directly, from your address at sejsun1.nri.co.jp to
f5.n310.z1.fidonet.org, for example, which is a local BBS, from which I
had a "point" connection until fairly recently, and as such also was
directly addressable as @p1.f5.n310.z1.fidonet.org, yet that connection
was by dial-up, from once-per-week to several times per day. Would you
count my system in that configuration as "connected"? Since MOST FidoNet
systems are public-access, any person with a computer and a modem may be
considered to be "connected".
The account from which I am posting, will, when I press send, attempt
to directly pass this message both to your address and to the list server
via SMTP. Is the computer on which I am typing "connected" or only the
one which will attempt to E-Mail you the message, with the SMTP
connection. I could also have posted this reply directly into the mail
system on my own machine, which would have routed it via uucp using a
TCP/IP connection over a SLIP link to another machine, which would later
have forwarded the message via uucp over a direct uucp serial connection
to a gateway system which would route the message back out via SMTP to
your sejsun1.nri.co.jp system.
I am not trying to confuse, rather to explain that "connected" is quite
a relative term. Speaking of which, "term" (a play on words), I could
start up the Australian package "term" which would allow me to originate
IP traffic on my machine across this login link, into a Unix socket on
this antelope.wcc.edu host, and then directly across other interconnected
hosts. Using that same utility, I could publish access to my host at this
IP address with a given port number and others could telnet or ftp
directly into my workstation here on the desk.
At the same time, there are users all over the world behind firewall
systems which can prevent their "access" or your "access" to them, yet in
a very real way, they are more "connected" at the moment than I am.
In a very real sense, you would not be underestimating to say hundreds,
and not be overestimating to say millions. (I have two other computers in
this same room which COULD be connected (one of which IS directly
connected to my workstation) at the same time, and use the same
"connection" pathways onto the net. It happens that one of them is not at
the moment powered on, and the other has noone logged into this
workstation, so they certainly are not "connected", yet one is??? The
MS-DOS PC has uupc installed and a locally recognized host address under our
TotSysSoft.com domain, so it is also directly addressable from your
sejsun1.nri.co.jp system, although it has been several months since any
traffic actually passed over that link.
When a flashlight or portable radio has a cord which allows it to be
plugged into the wall socket, is it "connected" ot the electric utility when
plugged in? -- when plugged in and turned on? -- when plugged in, turned
off, but recharging its internal batteries? -- when plugged in, turned
off, but the batteries have attained a full charge and any power drain is
due to mismatch between the charging circuit and the battery capacity?
If "connected" means potentially reachable in some way, from another
computer, in the midst of which, some traffic may pass through another
host which is ALWAYS reachable, then nearly every computer in the world is
in some way "connected" already (a few of the older ones still in use lack
a telephone modem, or other means of "plugging into" the InterNet, but
many sitting in the back of a closet, or on a repair bench, or an
office desk, have a built-in modem which has never been attached to a
telephone phone line). Many portable radios which have a "power cord"
available have never been plugged into the electric utility, also. Also,
very like any other kind of population statistics on a global scale, it
changes from second to second.
Bruce Gingery
---
InterNet NeXT-Mail: bruce@TotSysSoft.com
InterNet MIME-Mail: lcbginge@antelope.wcc.edu
On Tue, 12 Apr 1994 wantan@sejsun1.nri.co.jp wrote:
> I'm Itaru Watanabe from Nomura Research Inst. Japan.
> I will have meeting about the Internet for my customers
> in next week and one of the customers tells me he wants
> to know how large internet is,how many hosts it is connected
> and so on.
>
> Does anyone know where I can get internet fact data ?
> Thank you in advance.
>
>
> Itaru WATANABE
> Nomura Research Institute,Ltd.
> YOKOHAMA, JAPAN
> telephone:+81-45-336-8360
>
