Subj : Re: computers
To   : Malvinas
From : tenser
Date : Fri Jul 19 2024 06:56:43

On 18 Jul 2024 at 03:23p, Malvinas pondered and said...
 
 Ma>  Ni> (...)  I didn't even know about the
 Ma>  Ni> internet in 1993, and I'm not entirely sure how many people did.
 Ma>  Ni> 
 Ma>  Ni> Nightfox
 Ma> 
 Ma> IIRC, 'the internet', as such actually started out in 1995, replacing
 Ma> Compuserve.

Oh goodness.  Nope, the Internet has a much longer history.

The basic ideas started percolating in a few brains in the
early 1960s; the idea of packet-switched, store-and-forward
networks arose from the desire to make networking more
robust against failure.  The basic idea is to break communication
up into small, individually addressed units of data that are
independently routed from a source to a destination.  Then if
some switching hub failed (or got hit obliterated by a nuclear
explosion), data could route _around_ the failed node, and
communications would be only minimally disrupted.

The first large-scale test of this idea was the ARPANET, which
became operational in 1969.  ARPANET was a smashing success,
ultimately connecting several hundred research sites, but it
was limited (for example, host addresses were limited to 8
bits).  Early success gave rise to additional research in
_inter_networking --- that is, connecting networks, not just
individual machines.  That, in turn, led to the layered design
of the IP/TCP (as it was originally called) networking suite;
a few successive designs led to IPv4, which was first widely
available IP protocol, and that was introduced for real in
1983, when it officially replaced the old ARPANET NCP protocol.
Note that the early Internet actually used ARPANET as an
underlying link-layer protocol.

Of course, TelCos and others wanted to get in on the packet
switching goodness, so in the late 1970s, they came up with
X.25, which was a packet switching protocol designed for signaling
in the PSTN.  It's...ok, I guess, but kinda gnarly.  Eventually
people started to get Serious(TM) about packet switched
networking, and the ISO convened a working group to come up
with a Standard(TM) internetworking protocol, resulting in the
OSI (Open Systems Interconnect) specification.  The idea was to
take over from the researchers and deliver an offering that was
industrial production-ready.  CompuServe, and some of the other
relatively early timesharing service providers, I think did end
up using X.25, but no one ever really adopted OSI.

In 1978, Digital Equipment Corporation released the VAX-11/780
"super" minicomputer.  This was a 32-bit machine that immediately
attracted the attention of much of the industry including, crucially,
the researchers at Bell Labs who had written Unix a decade prior
and targeted it to DEC's 16-bit PDP-11 series of minis.  They quickly
ported Unix to the VAX, where it escaped to the University of
California, Berkeley, which started rolling its own Unix distributions:
the Berkeley Software Distributions, or BSD.  In 3BSD, support for
virtual memory on the VAX was added, and BSD became a serious
competitor to VMS, the DEC-supported operating system for the VAX.
Around this time DARPA, which had sponsored the ARPANET, was looking
for a way to standardize computing across its vendors and researchers.
They'd pretty much decided to use the VAX, but the question of what
OS they'd use was still up in the air: Unix (as in BSD) or VMS?  The
researchers at Berkeley, in coordination with Bolt, Bernanek, and
Newman in Boston, collaborated (kinda...it was fractious) in providing
TCP/IP for BSD on the VAX.  DARPA chose BSD for its software
standard, and that spread like wildfire across computer manufacturers.

Around the time this was all happening, the Internet was undergoing
exponential growth.  Bear in mind that the Internet had been done
under the aegis of the US Government as a research project, but in
1992-ish the Clinton administration opened the Internet up for
commercial development.  It was widely thought that OSI would be the
future, but the Internet was there.  Also, in 1991, some guy named
Tim Brenners-Lee at CERN in Switzerland invented this little thing
he called the World Wide Web.

So that's the basic idea: by the mid-90s, commercialization of the
Internet, based ont he web, running on pretty much every Unix available
at the  time, was in full swing.  Within a couple of years, OSI was
a distant memory as the Internet swept all that came before it.

CompuServe, Prodigy, etc, were all pretty small compared to the
Internet.  They dated from an earlier age of commercial timesharing
services that were pretty obsolete by the start of the 90s.  But
that's a story for another time.

--- Mystic BBS v1.12 A48 (Linux/64)
 * Origin: Agency BBS | Dunedin, New Zealand | agency.bbs.nz (21:1/101)

.