An early social un-network
Feb 16, 2025
reminiscence
POWCON
Vax
With one thing and another I’ve recently been feeling quite nostalgic for the technology landscape of my early university days (around 1990) and so I wanted to write a little bit about that.
Background
To set the scene for you, I was at university - or rather at polytechnic - in South Wales. Polytechnics were the very much less prestigious more vocational institutions. While I was taking a Batchelor of Science (BSc) course many of my contemporaries there were taking Higher National Diplomas (HND) or similar non-Batchelors degrees. In fact I think my course had some kind of option to convert it to a Batchelor of Engineering (BEng) after the industry year component but I never looked into that. A year or two later the Polytechnics were all standardised as Universities and so I entered a Polytechnic but graduated from a University.
So anyway, a not-so-prestigious institution, but with some decent courses and lecturers, out in the valleys of South Wales in the UK; at that time still quite a deprived area in the aftermath of the destruction/collapse of the local coal mining industry.
Also I was a huge nerd. A friend once saw a picture of me in those days and declared “Wow! You were a hipster back then” and I had to gently explain that we didn’t have hipsters back then. I was just a massive dork. It didn’t have even the faintest whiff of cool about it.
Networking
Not the social kind. The electrical/physical sort.
Where you measure the beginning of “The Internet” does depend a little on perspective. To be 100% clear I’m talking about the internet here, not the Web and HTTP. ARPANET was its progenitor so did the internet start in 1969? But there are a series of steps after that taking things closer to the TCP/IP based network of networks that we now think of. Whenever that threshold was crossed, in 1990 the Polytechnic of Wales was far more firmly on the Joint Academic NETwork (JANET) in the UK and that was an X.25 based network. We could send email to people over that network, and for the students that was about it - but… I knew perhaps half a dozen people who had email anywhere else at all anyway and they were all computer students like myself. So on the broader national and international scale the network was just not something we thought about very much.
I did know about Usenet (I think) and was mildly interested in it but as we had no usenet feed at our institution that was about as far as the thought went. The local academic landscape was where it was at for me.
I should mention, incidentally, that Bulletin Board Systems (BBSs) did exist in the UK, but because our phone calls were fairly expensive and because we had to pay for local calls as well as for long-distance calls, they were not something generally available to someone with a student’s financial resources and didn’t have the same kind of impact there as in the US.
Vaxen
For my first year in 1990 the majority of the hands-on courses were taught in the computer labs. Although this was well into the microcomputer era, our primary compute resource was a small cluster of Digital Equipment Corporation (DEC) Vax minicomputers running VMS (now known as OpenVMS¹).
Cabinets resembling those of our V8800 and 8650 Vaxes
These were multiuser systems - at any given time as many as forty (or more) students would be logged into these beasts. Usually they were sitting in front of Terminals connected via serial connections (and some kind of multiplexer device) into the Vax. At this time most of the terminals were DEC VT220s. Imagine a room filled with these and slightly damp students (it rains a LOT in South Wales).
A fairly typical VT220 with the usual green screen
While the Vax minicomputers were physically large, and these machines had been cutting edge in their day, 1990 was a little after their day - presumably the Polytechnic’s computer department obtained them second hand from a prior industrial owner - and while they weren’t exactly slow (lots of I/O support) they did suffer a bit when heavily loaded. Enough students compiling at the same time and the Vax might crash and take a good 20 minutes or so to come back up. A nuisance, but on the other hand you immediately had a room full of people with nothing to do except have a chat while we waited for it to come back.
At this same time I had a cheap clone IBM PC/XT compatible (i.e. an Intel 8088 based machine) microcomputer of my own and the perceived performance was in the same kind of ballpark. I didn’t feel that the Vax was particularly slow, although it had some notable deficiencies even then - particularly each student was allocated a fixed quota of disk space of 100kb or so, whereas my PC had a 32mb disk (quite large for the time) and both a 5¼" and a 3½" floppy disk drive that would have supported 360k and 720k respectively at that time (I don’t think I ever upgraded the 5¼" drive and it would have been a year or two before I had a 1.44mb 3½" floppy drive).
The social dimension
So why do I refer to all this as a social un-network? There were those moments when everyone in a room was thrown out due to a crash (and there were several labs full of terminals, not just the one), but actually I’m thinking far more of the software environment available to us on those machines.
One of them that I alluded to earlier in this piece, happily, is still with us. If not thriving then it is at
least hanging in there: email. You could definitely send email to any user of the Vax cluster. You would only
need to know their account name, no @ symbol necessary (much like emailing your colleagues from the typical
corporate account today). After a few years of that, despite almost never using my middle initial in other
contexts I still tend to think of myself as (all-caps) DCMINTER for computer-related purposes. I think all
computer students at this time will have had a similar experience with email and perhaps a similar fondness
for it.
Twitter, eh? Ok, no, fair, we didn’t have that. Or the original Facebook “wall” (anyone here remember when that was just simple text statements one after another?)
But we did (kind of) have something similar. Those of you with a Unix background are probably thinking of
.plan files and the finger command, but VMS didn’t
have those by default. But you could issue a command which I think was SHOW PROCESS and see a list
of all interactive processes - logins - on the system and you could also run a command after login to customise
one of the fields displayed by SHOW PROCESS - and thus say something in about 10 characters of space. Virtually
everyone customised this in some minor way and it allowed you to declare yourself an enthusiast and not
just a user of the system.
Instant Messenger
No really, we had this. On the Vax it was DEC$PHONE and allowed up to four (or was it five?) simultaneous
participants in a real-time text chat channel. If you “phoned” someone using their account name on the
system then a “ringing” text alert would appear in the middle of whatever else they were doing on their
terminal. Awfully like getting @here alerts in Slack in the modern era.
18-year old nerdy students being what they are, we’d look through the list of processes to see who looked like an enthusiast, “ring” them, and make their acquaintance for a chat. Here we’re really seeing something that was definitely social and had a lot of the characteristics we see in modern social networks.
The DEC$PHONE system is almost certainly the reason for my pretty unnerving typing speed - a year or two of “chatting” to friends on that thing (with the rather stiff but gloriously tactile LK201 keyboards for the VT220 terminals) and I can rattle off a phrase “slightly slower than I can talk but slightly faster than I can think” as I sometimes like to say.
What’s sort of astonishing is that 35 years after my cohort was messing around on this sort of primitive system, Google and Microsoft still seem to (somehow) be screwing up their instant messaging strategy, and even Facebook (fine, Meta) are somehow squandering the advantage they had with Facebook by making the Facebook experience so dire that they’re obliged to splinter between their acquired alternatives. Ugh. But I didn’t mean to get onto that; let’s look at another favourite.
Forums
The Vax had forums. They were really neat! A lot of the software names on the Vax contained a dollar symbol for filesystem related (I think) reasons, so this was DEC$NOTES and while we’d think of it today as a forum they called it a “conferencing” system. In fact they called it POWCON (for Polytechnic of Wales CONference) and it was - well, it was exactly like a modern forum system. Reddit on the small scale if you like. It just wasn’t on the internet.
From this paper, “Computer conferencing: a versatile new undergraduate learning process”, you can see that it was introduced with high minded academic use in mind in 1988, but by 1990 it was at least as popular for science fiction book threads and an interactive story thread (each contributor takes it in turn to extend the story) and similar lowbrow topics as for any academic pursuits that it might enable.
As it happens I captured the interactive story thread to disk and so I still have a narrow window into that perspective 35 years later. Most of it’s in-jokes about lecturers, science fiction books, nerdy pop culture generally, and our location in the rainy Welsh hillside, but I can still feel a spot of cameraderie through the juvenile humour.
I still enjoy forums - at one time Slashdot, and in parallel a lot of postings on the Sun Java forums (now, alas, gutted and wrung free of fun by Oracle), then Hacker News of course - that latter somehow proving remarkably resilient but doubtless to be toppled one of these days (I never got into Digg or Reddit in a big way for no particular reason).
Sneakernet
Of course my nostalgia is based on a lot of things. One of course is just that we were all physically present in roughly the same place, often doing related courses and thus knowing the same (occasionally oddball) lecturers. We could easily meet and often did. Also I was younger, more energetic, in some ways nicer in other ways not, definitely much more naiive and gullible (and literal minded and awkward). I don’t miss it all. But the human factors of sharing a small minicomputer system that was as much a communications hub as computational resource will not, I think, return again and I lament that a little.
Damn it, I had hair back then too.
Unix people
Our weird little offshoot of a Vax/VMS system on a tendril of the great tree of JANET was possibly not typical even for UK computing. This was the time of the great Unix systems - and those of you who had the benefit of that quite possibly had a superior resource! Not least because you probably had much easier access to the greater internet, but even within your Unix systems the software was a little more varied and a little more network and socially oriented.
I rather envy the finger command that allowed each user on a Unix system to obtain the basic details of their colleagues and the contents of a “.plan” file - very like a little blog posting that could be updated as often or as little as they liked.
Like DEC$PHONE you had talk I believe? Instant messenger but
far more network friendly and so closer to what we think of today.
Usenet - the great peer-to-peer forum system still exists (though I haven’t dared peek into the spam pools recently) and what is that if not a greater Facebook than Facebook and a cooler Reddit than Reddit ever aspired to? We still haven’t solved the spam problem even all these years after that fucking Green Card agency pissed in the well of truth.
Then most of the really good FTP sites were hosted on Unix systems of one sort or another. For some reason several of the really good ones were in Sweden so it’s kind of appropraite that I live there now. Though I think sunet.se is up in Umeå which is quite a long train ride from Stockholm!
I feel like the Unix kids got a headstart on the internet stuff over us Vax/VMS people - reading the amazing The Cuckoo’s Egg at that time gave me serious Unix envy. But maybe our little system was cosier.
Later…
Our Vax cluster did get connected to the internet eventually. There was a very narrow window where gopher and http hadn’t really caught on when we had access to the “PAD” (Packet Assembler/Disassembler) which was some kind of routing device or software or … something (honestly I’m hazy) … that allowed us to connect via our X.25 connection (NOT TCP/IP) to external systems and then from those we might be able to connect further onwards.
One of the X.25 addresses is still burned into my memory: 00004001018057 … because that was a relay system
(NFSNet Relay) from which one could access the real internet! I think I set up an account on Nyx
and it’s kind of amazing that still exists! At least I think it’s the same thing in some ship-of-theseus sense.
But contemporaneously with that the Linux Boot/Root disks (distribution if you can call a two-floppy system a distro) was doing the rounds, and the year after that I was a full Linux convert and installing … Soft Landing System I think? Though it might have been Slackware. I spent an industry year working for International Computers Limited² (ICL) who had their own minicomputer systems that, like the Vax but even more thoroughly, have disappeared. ICL were on X.25 as well and I remember using an email remailer system to pull the Slackware disk images (uuencoded) down from one of the Swedish FTP sites in 64k chunks and manually stitching them back together! Totally worth it. Even though it was 50 disks and they were definitely 1.44mb by then!
Returning to college Gopher appeared and was rapidly pushed aside by
http and the World Wide Web - though the web wasn’t quite a household word that year. A lab of Viglen 486 PCs had pushed
aside most uses of the Vax. I do recall managing to figure out the appropriate Vax incantations to launch one of the
DecWindows GUI client applications on the Vax 8800 minicomputer but with its display running on one of the Viglens -
it sounds simple, but remember, TCP/IP was a foreign land to Vaxes - you couldn’t just set your DISPLAY environment
variable and trust the network!
A million years later, here we are, in a foreign land. I like it, but sometimes I miss that smaller network with a more parochial user base and even the Welsh rain.
Footnotes
¹ It's not (and never was) very open though. There was a bit of a fad at that time of calling things "Open" which I think generally implied some kind of committee owning the technology or at least openly publishing some of the standards or protocols (even if they were patented, copyrighted, or otherwise locked down). For example the X/Open Group owned the XWindows system for Unix GUIs. Apparently VMS became OpenVMS because of its support for Posix but that does seem a bit of a stretch to me, not to mention a bit pointless because I can't imagine any potential customers failed to notice the rather large license fees involved.
² ICL is so fucking weird. It started out as "Lyons Tea House" a huuuge chain of cafés in the UK in the early part of the 20th Century. Sort of a British Starbucks if you will. Anyway, their accounting department grew too big for its boots, eventually becoming so large that they needed a new fangled computer system to keep track of it all, and eventually that computer department was split out as its own company. I think there were connections to the Manchester Baby and Turing's stuff as well. From such an acorn a British computer giant grew and then got swallowed by Fujitsu some time in the late 90s. I didn't find the management very impressive during my time there - they definitely seemed to be missing the boat on the internet but I couldn't say if that's what doomed them in the end. Most minicomputer makers did badly and while they had a nice range of PCs they were probably too expensive in an era of rapid commodification.