Malgond May 7, 2019 8:00 AM

Back in the 90s I was an admin at my university. Terminals were a scarcity. We’ve arranged a room full of old PCs assembled from what parts we could find or buy cheaply, no HDDs, but most had FDDs for data transfer. They were configured to boot from network to a basic image of DOS plus Win 3.x and some browser (was it Mosaic still? or Netscape? can’t remember) and basic utilities (anyone remember NCSA telnet?).

The problem? Mice were stolen regularly. We’ve made a contraption: got out one unused backplate from each case, drilled two holes through it and ran mouse cable through them (plus some rubber padding). Then we screwed the backplates back in and closed the cases. The cable run from the serial port plug through the drilled backplate (in and out) and then to the mouse itself. You couldn’t steal the mouse now without cutting the cable or opening the locked case. Problem solved – for some time.

Some time later someone – probably out of frustration that he can’t steal mice – stole all the balls from them. That was over the top. We’ve closed the venue and posted a message “Closed until all balls are returned”. Some patrons must have got really angry at the thief and had a few pleasant words with him – the balls were found in a bag near the door next morning…

Gunter Königsmann May 7, 2019 8:10 AM

I have seen cables run from the connector through the case of an PC to an mouse- but on a private PC whose owner had “learnt this at work”. He believed that to be a kind of an strain relief, though…

AnonCoward May 7, 2019 8:46 AM

RAM theft wasn’t common, but did happen. We had the locks to physically prevent the case from opening. Trying to keep track of what key, goes with what computer, in which lab wasn’t trivial, especially since the lock was located at the back of the case and the computer desk had an out-of-the-way location for it.

Kristian May 7, 2019 10:08 AM

The first time I messed with the inside of a computer was removing the pin the lock was connected to from the motherboard because my mother wanted to punish either me or my kid brother by locking it. (It was a commodore pc 40)
She found out by catching my kid brother playing on the same computer but thought we have found the key and I heard her scramble around to find a new hiding place.

Sed Contra May 7, 2019 12:30 PM

Clearly the only way to have real computer security of this type is to put your program on a deck of Hollerith cards – I think no one ever stole those big heavy metal key punch machines – and then stand in line in the room at 2:30AM (with a cold winter night outside) to hand your deck through a locked window that is briefly opened by a surly beard shadowed guy who takes it, relocks the window and submits it to the 7094 in the locked room behind the locked window. You’re on your own to secure the lineprinter paper output; better stay and keep watch.

Tatütata May 7, 2019 1:19 PM

I can only very faintly remember seeing locks on computers or maybe terminals in those days, but I don’t think I was ever inconvenienced by them, even in corporate environments. I did learn coding on PDP/11s locked away in cages (they did have a semblance of a front panel, and you could reconfigure the disks just by swapping the plastic caps!), and big iron (IBM and clones, Control Data, CDC) in glass aquariums that were attended by high priests. VAXes were already somewhat more democratic.

RAM theft on a name brand IBM AT would have been pointless, unless the target machine was also an IBM AT. Big Blue created early 1x128kb modules by piggy-backing two 4164-ish DIPs, with the wires of one chip soldered on the other one, I suppose that one of the chips had an reversed Chip Select input. An IBM part number was slapped on the result, these were about as useless as Hewlett-Packard’s.

It wasn’t widely known back then that tubular locks were worthless, as they could be picked by fumbling with the pieces of a Bic pen!

There was a big recall action in the early noughties of Kryptonite bike locks, I had one exchanged that was quite worn at that point. A good deal. I tried the pen, I can’t say it was easy, but I did manage to open a lock once.

Jon May 7, 2019 2:01 PM

As a technician repairing and upgrading computers in the early 1990s, insider RAM theft became such a problem that the whole procedure got revised to include documented handoff after handoff for every memory module. J.

Tatütata May 7, 2019 2:35 PM

Re: PDP/11 emulator. Tnx! The Russian help files can be deciphered if you set the charset to KOI8.

It doesn’t have TECO, which, amazingly, I entirely forgot. (I won’t dare touch vi, which is apparently a son-of-a-TECO).

I managed to edit, compile and link mostly without (?) error a trivial Fortran program (“lala.for”, the files remain from one session to the other), but it gives me the infamous “trap four” when I attempt to run it. (4=illegal instruction, 10=memory fault, or was it the other way around?)

Basic also works. I didn’t check out Pascal or the Macro assembler. I also almost totally forgot the PDP 11 syntax [ something like MOV R1,@(R2++) made a lot of sense], which seemed to me at the time the 8th world wonder. Crap like “STM 14,12,12(13); BALR 14,15” is more deeply engraved in my neurons…


For theft, now that I think of it, floating point coprocessors (8087 or 80287) were rather more valuable than memory in those days. I remember breaking the piggy bank ordering one in the US which had to clear customs. I agonized for hours until I could summon the courage to flip the power switch back on after plugging the ceramic thing onto the motherboard.

SpaceLifeForm May 7, 2019 4:31 PM

In the 90’s, i worked on a telco project to be deployed in the country of [redacted].

It was a sparc machine.

One day, the telco in [redacted] called to say the system was running real slow.

My response: Put that ram stick back in.

They did. System response time back to normal.

The candidates of [redacted] is not long.

You likely can guess the city.

Rj Brown May 7, 2019 5:34 PM

@Tatütata: 012727 MOV -(PC),-(PC)

That was lot of fun! The 1 instruction program. It copies itself backwards thru all of memory. Plant that word at some random location in someone else’s computer (via the switch register) and wait for the loud scream. HAHA

Key Lock wise, I have my laptop configured with the boot drive totally encrypted using LUKS, no partition table, I use LVM. I carry a USB stick on my keychain along with my house and car keys. I must have this boot stick to boot mny laptop, then a pass phrase for LUKS do decrypt the boot drive, then my usual login password. If anyone steals the machine, they get the hardware, but not the data. A second drive boots windoze so I can easlit clear airport security etc. If the boot stick is not present, it boots into windoze.

Anon Y. Mouse May 7, 2019 5:51 PM

I worked at the main campus of “World-Famous-High-Tech Company”
in the mid-1990s — several connected multi-story buildings, a few
thousand employees — RAM and CPU theft were a problem. This
was when Pentiums were the latest chip. The company had a dog’s
breakfast of older PCs, but was constantly deploying newer machines.
These were the usual targets for theft; despite security checks at
every exit, it was relatively easy to smuggle to chips out. At one
point several machines per night were being cannibalized. I never
heard who was doing it, whether it was cleaning staff or employees,
but eventually the thefts tapered off, probably the combined result
of catching some of the thieves and declining prices for the chips.

Dave May 7, 2019 7:19 PM

We had that problem at Uni at the time, someone was stealing physical memory from Macs and hiding it by winding up the virtual memory setting to match the amount of stolen physical memory. Took awhile before it was discovered.

TRX May 7, 2019 10:05 PM

RAM was expensive enough that thieves targeted it in the 1980s.

Infoworld and PC Week both covered the story of a break-in as a (fab? warehouse? it was long ago) where thieves killed someone while stealing RAM from inventory.

Adrian May 8, 2019 12:50 AM

Working at a university through the nineties we had entire PCs & Macs stolen, plenty of times cases were opened and RAM or CPUs stolen, a few times where the PC was locked down but someone managed to punch in a 5.25″ blanking plate and squeeze their hand inside and still steal the RAM.

Also had an entire lab of mouse balls taken one time, but that was probably not intended for profit, more as a joke or to inconvenience everyone, same as the endless sticking of a thumbtack through the ethernet coax and breaking it off, a great way enjoyed by some to take out a lab for a few days

Mike May 8, 2019 5:17 AM


Your college must have had very lax security. Physical security camera and limited access can do a lot to deter theft in a university.

Back then, I’ve seen a lot of “dumb terminals” deplored around my campus and quite a few Next terminals or whatever they were called running an odd Steve Jobs operating system. Nobody steals them because they would have no use for such equipment in a home.

Anders May 8, 2019 6:07 AM

In ex-Soviet bloc the object of stealing was actually EPROM chips
(2764, 27128, 27256 etc). Original IBM PC/XT/AT and some clones
such as MAZOVIA (XT clone) had BASIC in their ROM chips, but at
that time nobody used it anymore (MAZOVIA came with HDD). So people
steal the ROM chips (computer runs just fine without them so noone
noticed) and used them to build Sinclair Spectrum home computers.

Malgond May 8, 2019 8:10 AM

@Mike: back then CCTV systems were expensive (at least in my part of the world), only major bank branches and big industrial complexes had them. Not even police stations AFAIR.

As for limited access, it was deliberately a publicly accessible computer room, to enable students from all branches and faculties to have some computer time. Other faculties offered very limited capabilities if at all. During my student time situation was still worse, you had to almost beg for a VMS account and there were very few terminals and lines of students waiting… This is why, when my friends and I got to work at the university’s computer facility, our first initiative (besides slowly cleaning up that Augeas stables, I could tell stories for days on end) was to set up this room.

Theoretically only students and staff were allowed, but there was no strict entry control, no door warden etc. It wouldn’t make much sense anyway. There was always some of us sitting in the adjacent “op room” to supervise and offer help etc. but the person on “op duty” had also other tasks: answering phones and emails, small system admin tasks (resetting passwords, receiving and verifying new accounts applications, hundreds of them, fixing up small malfunctions, etc.) so there was only half an eye on the computer room. The act of stealing a whole computer could be possibly noted, but a mouse? no way.

The whole area including the server room did have an alarm system with door sensors and motion detectors albeit quite an old one. It was only armed for the night etc. when nobody was inside. No electronic access control, either, too expensive then.

Yes, it was in the early 90s, and you may wonder at it all, but take into account that my country was well behind the USA and other parts of the world then, only just started to catch up.

kronos May 8, 2019 8:39 AM

I recall an accounting manager who got a ‘high end’ PC AT clone with a 40MB, yes mega-byte, hard drive. To prevent theft he ordered a heavy steel cabinet that sat on his desk. A few weeks later he went on vacation, and as luck would have it some other manager just had to get a critical file off that computer.

What to do? The cabinet was locked and there was no networking in those days. I was called to see if I could get in it (as a lowly PC tech at the time), but I said my boss would have to approve it. He was called and signed off on the adventure. I found that a simple finger-nail file, borrowed from a secretary, would open the lock if it was wiggled while turning. Voila!!

Everybody was happy except the accounting manager, who was furious at me when he returned from vacation. He made me promise I would never do it again, and I said I would comply except when my boss approved such action. The latter did not make him happy.

MikeA May 8, 2019 11:31 AM

@ Tatütata — IIRC, VI is descended from ED (via EX?), while EMACS (EDiting with MACros)is descended from TECO. I still sometimes use a version of micro emacs, with key bindings to mimic DEC TPU, set up to mimic VMS EDT, which gets most of its “visual mode” UI from PDP-11 KED.

Slightly more on-topic, the first “lock on a computer” I ever saw was on an IBM 1311 disk drive, where it physically controlled the ability to do a format command. This combined nicely with the feature of having a bit in the sector header determine if the data for that sector could be written. When you have exactly one disk drive, and are using the machine to (among other things) teach assembly language programming to beginning students, it’s nice to not have to re-build the system files from cards after each student’s run.

Of course, many of us knew where the key was kept, but this was more intended to prevent accidental damage, not “enemy action”.

Frank Rizzo May 9, 2019 9:33 AM

In the 90s there were two forms of ‘Ram Raiding’. The first involved a stolen Ford being rammed into a shop door and high value contents (jewelry, perfume, footwear) taken.

The second was the theft of server RAM modules. It was not so common but I did have first hand experience of it.

I was working for a government agency in Newport in 1995. We got called in one evening because a couple of servers had gone offline. They were in a 13 story tower block, which you would think would be secure but it was not.

Someone had climbed the stairwell, kicked down the server room door, carefully removed the panels from a couple of Proliants and took all the SIMMS.

IIRC these were 16MB modules and were worth around £500 each? So that would have made them very valuable. But I could never understand where they would end up. How could you fence specific server RAM back in the days?

vas pup May 10, 2019 2:42 PM

In US signature is kind of stamp of identity.
This machine could easily do this with signature. Yes, forensic guy could verify thereafter that signature was not made by human. Same applies for staging murder as suicide with hand-written pre-suicide statement. Now it is possible to frame somebody for something person never did, person image is destroyed directly/by media, It could be restored later, but damage already done. Particular for election purpose.
Generate fake letter as of 20 years ago with racial/sexual you name it content, and your opponent is kicked out of the race.

For long letters is important to incorporate style (set of language tools usually used by the person you want to frame) utilizing AI not a problem, BUT you need paper, ink matching date of ‘framing’ letter as well.

counterkleptocracy May 10, 2019 6:18 PM

Wow, thanks for this content. Despite my own recurrent spite, instead, this was inspirational.

I really might get into a hobby of home-designing experimental chassis locks and similar stuff.

I feel this has been missing for several years and I often think about stuff like this as I’m tinkering with my own gear trying to modestly customize stuff just a little bit more. I never get very far, but I could if I put my mind to it.

Most stuff these days is just NOT designed for effective ergonomics nor effective electrical safety nor effective real-life security. And just as implied in the video, regardless of all the fussy inbuilt accessories, most people just use their own computer and really do NOT want nor need vast user systems and multi profiles or even any passphrasings at all.

This is common knowledge in some specific applied computing circles, where we routinely TURN ALL THE B.S. OFF. (B.S. in this context does NOT stand for “Basic Scripting” nor “Basic Syntax” nor “Base Systems”).

Thanks again. This puts me back on track.

2019 May Continental USA; land of the theif, home of the slave.

Leave a comment


Allowed HTML <a href="URL"> • <em> <cite> <i> • <strong> <b> • <sub> <sup> • <ul> <ol> <li> • <blockquote> <pre> Markdown Extra syntax via

Sidebar photo of Bruce Schneier by Joe MacInnis.