Couple that with the early search engines such as Lycos and WebCrawler and there's a story to tell.
I was a volunteer at a local non-profit fledgling ISP way back in the day. Sparc 10, bank of Practical Peripherals modems, Portmaster 2e, blistering fast frame-relay T-1 to the net, typical state-of-the-art setup for its time.
I was doing a lot of the early network stuff, and another guy was the system admin. He put together a quick personal page that included a picture of a potted plant. No real reason for it, he threw it together primarily so he could test the new Mosaic browser.
Somewhat as a joke, he included an ALT tag on the photo of the plant, "This is not a picture of a naked woman." This was for the benefit of the majority of the visitors using Lynx.
About a week later, the blistering fast T-1 line became quite saturated with visitors to his personal test page. Search engines picked up the keywords and people on the Internet did what people on the Internet are still doing today.
I don’t remember hearing about IP for VAX/VMS 2.4, but I was part of a group at Intel in 1981 looking at ARPAnet for moving designer tools and design files as an alternate to leased bandwidth from $TELCOs using DECnet and BiSync HASP. The costs of switching from 56 Kbps to ARPAnet’s 50 Kbps convinced us to wait. Clearly, private demand drove the subsequent transition as the TCP/IP stack became effectively free.
I'm not sure how we heard and got a copy of the CMU IP stack, but it was probably Mark Reinhold who now owns Java. It was definitely after 1981 and definitely before 1985, probably somewhere in the middle. Just the fact that we could get such a thing was sort of remarkable in those early days, and especially for VMS which was, I won't say hostile, but had their own ideas. I don't know when early routing came about but DEC charged extra for routing for DECNet, so that was yet another reason IP was interesting is that it took little investment to check it all out.
I miss DECUS, but not DELNIs.
Yeah, I miss DECUS too. I remember one plenary when somebody asked when the VAX would support the full 4G address space to laughs and guffaws from panel.
>
> Before 1983, the ARPANET wasn’t an internet, let alone The Internet. Each
> ARPANET connection required a host-specific interface (the “IMP”) and simplex
> Network Control Protocol (NCP). NCP used users' email addresses, and routing
> had to be specified in advance within each NCP message.
Then again there were IMPs fitted to various systems like TOPS-10,
ITS, Vax/BSD Unix, IBM370, etc.
So was that really all that different from ethernet vs, oh, wi-fi or
fiber today, you needed an adapter?
>
> Even so, the Internet as a platform open to anyone didn’t start until 1992. I
> know you joined late, in 1999, so you probably missed out on this history.
Well, we certainly tried in 1989 We had customers from all over
The World, um, the big round one you see when you look down.
For several years we had UCSB’s IMP control panel hanging in our office as a wall decoration (it belonged to Larry Green, one of the UCSB IMPlementors). I still have the manuals. The actual IMP with 56Kbps modem was in a huge rack with lifting eyes for a fork lift, and weighed about 500 lbs. Every IMP had a unique customized host interface, which packetized bit-serial data from the host over the host’s usually proprietary I/O bus.
While this was part of computers internetworking with each other, it was not the capital-I Internet.
I really envy folk that love DC for networking gear :-).
Work in 2007 was an all-DC network. I rebuilt it into AC, considering the ISP also owned the data centre (most of whose customers bought AC). The space we freed up and the ease of deployment was night & day.
Currently, we obviously need DC for the terrestrial Transport and wet plants (because that's just how classic telco rolls), but I also switched all IP/MPLS gear to AC soon as I arrived. Heck, even the Arbor (now Netscout) gear, as well as the HP server rack, was loaded with DC power supplies. Those things just had to go.
There is an avenue of pleasure in not having to spend inordinate amounts of time adding major electrical planning to deploying/decommissioning a router, switch or server.
But yeah, I know the AC vs. DC discussion can become a rat hole.
I'm aware of data centre operators now providing DC as an option for their expansion projects, when they previously had it as the norm, FWIW.
LOL! Yeah, and in 1838 Samuel Morse’s telegraph system used electric impulses to transmit encoded messages over a wire to Speedwell Iron Works in Morristown, New Jersey. Was/ that /the Internet?
No, but you are ignoring the point of my message…
The TCP/IP internet existed _BEFORE_ the flag day you mentioned. The flag day was the end of NCP, not the beginning of TCP/IP. IIRC, at the time,
it was IP version 2, but IP versions 2, 3, and 4 came in relatively rapid succession of each other and 4 was the first version with (relatively) clean
layer separation between 2, 3, and 4.
According to Final report on TCP/IP migration in 1983 - Internet Society , TCP/IP was developed starting in 1975 and
declared the official future standard of the ARPANET in March, 1982, with a transition plan supporting both protocols (NCP and TCP/IP)
until January 1, 1983.
January 1, 1983 is more analogous to the future happy day we finally turn off IPv4 at the majority of peering points and PNIs than it is to the
past days when IPv6 began being deployed.
True, the initial deployment of TCP/IP and the flag day were much closer together for the implementation of IPv4 and deprecation of NCP
than has been the case for IPv6 deployment and IPv4 deprecation, but nonetheless, it is still true that there were at least several months
of TCP/IP deployment, testing, and use at multiple sites and on multiple systems prior to the deprecation of NCP on January 1, 1983.
Not only do I not miss DELNIs, I don’t miss any of the trappings of coaxial-based ethernet, including, but not
limited to:
AUI cables
AUIs
Vampire taps
BNC T-Connectors
BNC Terminators
SO-239/PL-259 terminators
TDRs that cost $10k+
Co-ax pinning
Thin-net cables getting cut by furniture legs
Following a long poorly documented thickness cable looking for the shorted vampire tap
Does anyone miss any of these things?
Twisted pair ethernet was probably the best thing to come to networking since TCP/IP.
Before 1983, the ARPANET wasn’t an internet, let alone The Internet. Each
ARPANET connection required a host-specific interface (the “IMP”) and simplex
Network Control Protocol (NCP). NCP used users' email addresses, and routing
had to be specified in advance within each NCP message.
I think you mean before 1982.
TCP/IP was deployed starting in 1982. NCP was deprecated (removed from the
ARPANET) January 1, 1983, but TCP/IP was implemented (and deployed) prior to that.
Then again there were IMPs fitted to various systems like TOPS-10,
ITS, Vax/BSD Unix, IBM370, etc.
So was that really all that different from ethernet vs, oh, wi-fi or
fiber today, you needed an adapter?
It really wasn’t, but even if you just want to count from TCP/IP forward, 1983
isn’t the correct date. 1983 was when we turned off NCP. It wasn’t when we
turned on TCP/IP. The turn on of TCP/IP occurred over several months, so there’s
no particular date that can be assigned to it.
The TCP/IP internet existed BEFORE the flag day you mentioned. The flag day was the end of NCP, not the beginning of TCP/IP. IIRC, at the time,
Owen,
But we’re not talking about the birth of TCP/IP. We’re talking about the birth of the capital-I Internet, which by definition runs exclusively on TCP/IP, and that didn’t start until Jan 1, 1983. Although there was experimentation using IP during 1982, that was still ARPANET. It was the guaranteed exclusive availability of IP that made 1983 the Internet’s birth date.
And no, it’s not analogous to the eventual IPv6 transition, because both IPv5 and IPv4 are Capital-I Internet standard protocols.
The TCP/IP internet existed BEFORE the flag day you mentioned. The flag day was the end of NCP, not the beginning of TCP/IP. IIRC, at the time,
Owen,
But we’re not talking about the birth of TCP/IP. We’re talking about the birth of the capital-I Internet, which by definition runs exclusively on TCP/IP, and that didn’t start until Jan 1, 1983. Although there was experimentation using IP during 1982, that was still ARPANET. It was the guaranteed exclusive availability of IP that made 1983 the Internet’s birth date.
IMHO, that’s an absurd definition. It was still ARPANET after January 1, 1983 too. Prior to 1982, it was ARPANET on NCP. During 1982, it was ARPANET running on NCP+TCP/IP, much like the Internet runs dual stack today on IPv4 and IPv6.
In 1983, NCP was removed from most of the backbone, as I hope will happen with IPv4 in the next few years.
And no, it’s not analogous to the eventual IPv6 transition, because both IPv5 and IPv4 are Capital-I Internet standard protocols.
You’re picking arbitrary definitions of Capital-I Internet standards. NCP was every bit as standardized as TCP/IP in 1982.
Both were documented in the same IEN series of documents.
IEN later (well after TCP/IP) evoked to become RFC.
Don’t believe me? Look at the hosts.txt file from IPv4 days which still referenced IEN116.
