As many of you may know, I have a somewhat large and extensive computer museum, a lot of which can be seen on I didn’t start collecting these because it was trendy; in fact it was the complete opposite of trendy when I started. Back in the late 80s and early 90s I seemed to be on a one-man mission to try and convince people NOT to throw this stuff away. I would try and convince universities and companies that we needed to keep machines, peripherals, data and manuals for their historical importance and that pretty soon; we’d end up regretting chucking it all away.

I would have the heartbreak of going to places to rescue an old piece of equipment they were scrapping and being shown a warehouse full of stuff that was about to be dumped. I didn’t have the transport or the storage space for it so I had to be very selective in what I could take.

The majority of my first collection was actually mostly stolen when a truck of mine was broken into. I assume all the contents were just trashed because financially, none of it was worth anything at all; people were still effectively paying to have this stuff crushed or at best, having it taken away free for the price of the gold on the circuit boards. I lost a lot of stuff that is completely irreplaceable, and just about all of the early MUD and BB history I had on various disks, tapes and paper tape.

These days of course, things have changed – Retro is in fashion and it seems to be pretty trendy to collect ancient computers – This has the advantage that I don’t need to any more, since I didn’t collect most of this junk because I liked it, I collected it because somebody had to and nobody else seemed to be stepping up to offer. Quite a lot of my stuff has gone to proper museums now but I still keep hold of my core collection though, mostly out of petulance and spite…

And so, to the subject of this post!  I read an article once that claimed that mankind lost more data in the 1970s and 1980s than at any other time in history and from my experience, this causes some unusual problems. I thought I would give two somewhat ridiculous examples that I have come across lately.

The first relates to the title of this article: “Primary Sources” – Wikipedia’s aim is to become an encyclopaedia of just about everything and because of its position of being the major encyclopaedia on the Internet it is generally a very good source for documenting the history of computing. A couple of years ago I decided to update an article about something of which I am one of the primary experts, having written it. It was an article on some obscure Multi User Game history. I made a few changes to the Wikipedia article and corrected some things. Apparently I was not meant to do this. Shortly afterwards I got into a discussion with an editor who was complaining that I hadn’t referenced any proper sources. I explained that I was the primary source on this matter, but apparently that didn’t matter. Had I ever given an interview on this, or written a book, it would have been fine, I could have referenced that; but it seems that I can’t just reference myself. Wikipedia’s rules say “Do not base articles and material entirely on primary sources. Do not add unsourced material from your personal experience, because that would make Wikipedia a primary source of that material.”

The problem is that all of this stuff happened in the late 80s and in the late 80s there was a LOT of reference material that would have been a gold-mine for Wikipedia; servers full of documentation, academic (and non-academic) papers, and the ever present and ever busy Bulletin Boards. When the World-Wide-Web came along these things didn’t migrate and as the old systems were decommissioned, the data was simply discarded and lost forever. There was no “way back machine” or Google Cache in those days and at best, some people may have their own backups on floppy disks or paper printouts. I used to have lots of these, but most are gone now. I still have some of the actual machines which probably have the data on them – but there seems little point me pulling it off if the data itself becomes an unusable Primary Source.

And so the problem is that there is a distinct lack of primary source material from the 1970s and 1980s. If Wikipedia really does want to document this era in which a lot of ground-breaking, fun and interesting history was actually made then they really should consider allowing Primary Sources to contribute. A lot of people who did a lot of good stuff back then, developing, using, and researching things that weren’t “invented” until decades later simply aren’t self-publicists. They don’t give interviews (even if anybody had a clue that they should be interviewing them!) and they don’t write books or appear in them unless they accidentally happen to cross paths with the likes of Tracy Kidder or Katie Hafner at just the right moment. They are distinctly absent from history and the way things are, along with the data we have lost, the people will be lost too.

Another completely different problem happened a few years ago when I was trying to resurrect the first Multi User Games, Essex MUD and MIST, to run at Bletchley Park’s Computer Museum. Essex MUD ran on a DEC PDP-10 running the TOPS-10 Operating System. Being a good Systems Manager all those years ago, I backed-up everything I could think of before I finally turned the off-switch on the Essex Games. I certainly had enough so that one day, I could recover it – At least, I thought I had; and now I had to put this theory into practice.

The first challenge was to get hold of a PDP-10 – We thought we had one at Bletchley but it turned out to be an obscure (but very pretty) PDP-11. As far as I know there are no complete and working PDP-10s left anywhere in the world but this was less of a problem than it may have first seemed. One thing that was relatively easy to get hold of was a PDP-10 TOPS-10 simulator. I could run a completely realistic (if not slightly too speedy) simulator on another system, in this case a MicroVAX. I could have used something else, but I wanted to at least keep the whole thing on DEC machines. I got the simulator running, I loaded all the data from my various backups and I went to get the BCPL compiler to compile the source code for MUD; and this is where everything started to go wrong.

It seems that nobody had ever thought to keep a copy of the BCPL compiler. Why would they? It’s Systems Software. It’s not like it’s just going to vanish one day, is it? Well yes. It is, and yes. It had. I contacted Richard Bartle, who along with Roy Trubshaw originally wrote MUD1 back in the late 70s – I thought he may have a copy of the BCPL compiler somewhere and he confirmed that he may have one on an old half inch tape from 1981 ish. Half inch magnetic tape isn’t really meant to last 25 years but it was worth a try, so he sent it to me and I popped it in my tape drive to read it. And it wouldn’t. Of course, PDP-10s used a different physical 7-Track tape format than my more modern 9-Track reader could cope with. If it had been a different logical format I could have fixed that, but physical was completely out of my control.

Never being one to give up, I decided to put out a call to see if anyone, anywhere in the world, had a working old PDP-10 7-Track tape drive. To date, nobody has. If (and that is a big “if”) the BCPL compiler is on that tape, and it may be the only copy left anywhere in the world, then I can’t get it off. Without it, I can’t get anything else to work at all and never ever will be able to.

Even when I was sitting, doing a backup in 1992, knowing that one day I would probably want to recreate this stuff, I never dreamed that I would not be able to get hold of a vital compiler just 15 years later – And bear in mind that I knew more about this stuff back then than just about anybody. I was one of the only people collecting old technology and preaching the need to keep things for the future; a major part of my job was recovering data from ancient and obsolete National Health Service tapes and disks that would otherwise have been lost and I was, and still am, completely obsessive about archiving against accidental loss.

In a few years we may well realise that we have lost so much of the 70s and 80s that it is verging on the unbelievable. In other areas of history and modern archaeology we have finally understood the need to keep first-hand personal stories from sources such as the mill-workers and miners and the soldiers in the First and Second World Wars. One day maybe I should expect somebody to turn up at my house with a tape-recorder in an attempt to force me to try and remember stuff. With luck, they will bring tea and cakes and forgive the fact that as a somewhat senile Primary Source, I will probably be quite useless by then.

It’s easier to learn from history when we bother to preserve it. The arrogance of not doing so seems quite incredible to me. But what would I know? I am just another unreliable primary source.

** Update: It seems that a BCPL compiler was finally found in 2020!