Remembering an ugly font and a forgotten monster

IBM has always loved selling stuff to banks… and banks have sometimes loved to buy that stuff. In my opinion IBMs most iconic banking machines were their cheque sorters. For many years processing cheques was the time consuming bane of every bank branch (not any more, but thats another story). IBM got involved in automating cheque proofing (the back office processing of a cheque) way back in the 1930s with their famous IBM 801 Proof machine which looked like a truly giant ugly cash register.

IBM 801 Proof machine – not a cash register (from IBM website)

They were so proud of it they even sung about it in their company song book (sing along with me!)

What do the banks consider great? Proof Machines. 
What is it they all advocate? Proof Machines.
A job that once was very hard now this machine with ease regards. 
We sell them all our Proof Machines. 

They don’t write songs like that anymore! (thankfully, now please stop singing it).

But did you know that in 1959 IBM nearly lost the interest of the banking industry. The reason was simple:

They invented an ugly and expensive font.  

The banking industry needed to automate check handling and were evaluating pre-printed machine readable data entry. But IBMs proposed barcode font was condemned by the American Banking Association as both ‘expensive and ugly’ (and they were not talking about IBM machines in general).

A rather modernistic font invented by a Bank of America academic partnership completely outfoxed IBM with a massive deal going to GE/NCR. Their font which became known as E-13B revolutionised the processing of cheques and would go on to change the banking industry. And IBM had both not invented it, but had also lost a massive deal allowing GE to eat some of their breakfast.

IBM needed to pivot and pivot hard they did. Within 3 years they had built a range of cheque reader/sorters to work with E13B that often played the lead role in their banking sales campaigns. The Banks wanted cheque sorters… and they had to buy the computer that ran it. If you want to see some amazing retro images, check out this archive from Martins Bank in the UK.

IBM cheque samples with E-13B MICR line used for testing and demos (supplied to me by an old IBMer)

IBMs first sorter could read 900 cheques a minute, but with a speed boost they managed to raise that to 1400 with their IBM 1419, which was their premier check sorter into the 1970s, by which point GE had abandoned the scene. Unable to compete with the S/360, they ended up selling their Computer division to Honeywell in 1970.

But banks wanted more.  With only 13 pockets the same cheque had to go through the 1419 sorter multiple times…. Each time exposing it to the risk of being mangled in a jam.   And the number of cheques to be processed was booming year on year.

IBM 1419 at the Australian Reserve Bank (supplied to me by an old IBMer)

IBM began work in the 1960s on a new much larger and faster sorter, but it was running very late and the market was again threatening to move on to Burroughs and NCR. They needed a stop gap, but when they did produce one, it would become yet another machine that IBM forgot: the IBM 2956-5

As a student of IBM history, this machine was a total mystery to me. I learned about it while researching Wikipedia updates when I started hearing stories from ‘old-timers’ of a mysterious 2956, a device which was apparently two 1419s bolted together. This was a fairly astonishing claim that I needed to document, but searches of IBMs own sites, Google, archive.org and bitsavers revealed nothing. Literally the only solid thing I could find was for sale listings in Computerworld magazine:

IBM 2956 Advertisment in ComputerWorld magazine

I was at a dead end. I could find no further concrete documented proof this product existed and updating Wikipedia on the basis of a for-sale notice was not a winning strategy. Despite claims to the contrary, Wikipedia editing standards are VERY high.^{(Citation Needed)}

So I asked a very special person for help: IBM’s own archivist.

Now I say “person” because this is a “real” individual. IBM not only has an archivist who responds to emails, but a person who is prepared to do leg work. Now I assumed that IBM Archives look something like this:

Indiana Jones closing scene (property of the US Government)

And that the IBM Archivist looked something like this:

Pieter Claesz Soutman (1601-1657), En skaegget olding, 1595-1657

I could not have been more wrong.

He replied a few days later and confirmed they had no electronic records of this product at all.
Nothing. Nada. Nix. Not an electronic sausage.

But he said he would take a look in the hard copy library.
I was excited that I had got a response from an actual person… but pessimistic about any further progress.

So imagine my surprise when I got a follow up email a few days later. The hard copy search had unearthed a single document which he had scanned just for me:
The IBM 2956-5 Multi-Pocket Reader Sorter Installation Manual-Physical Planning Manual.

And this document included a single image:

IBM 2956-5 (IBM Doc supplied by IBM Achives)

I finally had documented proof that the product not only existed but was exactly what was claimed.

But here is the kicker…. This hard copy document is literally all that is left. It is the only record of this magnificent Frankenstein created by IBM. They have nothing else despite two extensive searches of their own archives.

And the archivist? Here is that young whipper-snapper working on a S/360 exhibit. Is this the famed IBM archives behind him? I am unsure but it looks impressive.

From twitter

So yet another mystery of IBM history explored, documented and committed to wikipedia.

IBM may try and forget…. but I won’t.

Remembering the time IBM went to sea and then quite literally abandoned ship

About two years ago I was searching for Creative Commons (CC4) images to add to Wikipedia pages, when I came across a dump of old IBM marketing images hosted by a Norwegian technical museum – and yes this is what I do for fun. 

It was a gold mine of old pictures which I happily harvested to update various pages on IBM products.

However I was stumped when I came across this rather remarkable image:

IBM 5090 Bridge Console (Wikipedia)

What was I looking at? It looked like an IBM radar set mounted on a ship. I had so many questions. Was this real? Did IBM really sell this? A radar? On a ship?

Initial Google searches revealed nothing, so I started to dig, and the story I unearthed was amazing.

The story begins in 1970 when IBM released an industrial control computer system called the System/7. This was not a IT product to run batch operations or perform payroll. This was designed to collect real time data from analog devices like temperature sensors or door card readers, convert these into digital signals and process them. What I love about this product is that it was the domain of the creative sales rep. Someone who could think outside the dots, listen to client requirements and solve them with technology. Some of the real world uses for the product that I uncovered included:

• Detection of out-of-spec emissions in car manufacturing at AMC (this was the launch customer)
• Preventing meal theft from University of Pennsylvania student dining halls
• Taking remote telephone orders for Pfizer sales reps on the road
• Monitoring slot machines at Harold’s Club Casino in Lake Tahoe
• Controlling carpet weaving machines at Deering Milliken, managing selection of thread colours in real-time rug production
• Controlled the induction ovens at Wagner castings in Decatur Illinois which melted iron ingots
• As a network concentrator at many IBM sites.
• As a telco tolling billing system for long distance calls

It was most certainly used by people in hard hats:

Man in hard hat pretending to use a System/7. (Wikipedia)

I even found an IBM Patent application on archive.org for the “System/7 continuous dough making process control”.   Did someone at IBM bake a lot of bread? 

In case you didn’t believe me about the bread. (archive.org)

What I then learned was that a team of IBM developers in Boca Raton used the System/7 to design a maritime control system for ships like VLCCs (Very Large Crude Carriers), that was launched (see what I did there?) on October 15, 1973 as the IBM System/7 Maritime Application/Bridge System.

This is the red edition. Please destroy the blue edition (IBM Brochure)

It was designed to provide a form of auto-pilot for ships, along with collision assessment and detection, route planning and route tracking.    IBM tested the solution off Port Everglades, even installing a radar tower on the top of building 032-3 at Boca Raton, to test communications from ground to sea.

Until I found this, IBM radar was never on my radar (IBM Brochure)

I did try and find a photo of this mystical radar tower, but it eluded me. Still you can clearly see the sea in the distance in this promotional photo:

IBM Boca Raton (IBM Promotional image)

IBM setup a demonstration centre in Milan Italy to sell to European shipping magnates and Lloyds Register were so impressed they added it to their list of Approved Control and Electrical Equipment, which was a first for an onboard navigational computer.

The Maersk-Line reportedly had the System 7 installed on several of their ships, which was unsurprising given the Maersk CEO at the time, Mærsk McKinney Møller was also on the board of IBM (from 1970 to 1984).

Set the controls for the heart of the sea (IBM Brochure)

Total installations (ships) reached around 100 by 1979, at which point IBM Chairman Frank Cary had a dramatic crisis of confidence. What if one of these ships ran amuck and caused an environmental catastrophe? He got the lawyers involved who confirmed his fears. Their conclusion was that IBM should never have gotten into this business. His sudden change of heart (given he had been involved in the product launch!) was dramatic. He literally not only chose to withdraw the product, but also to remove it from existence. He assigned an executive named Robert Hood (sometimes referred to as Robin Hood) to oversee a program where all customers were not only compensated for the withdrawal, but agreed to remove the equipment altogether.

In hindsight, maybe Frank was right. Exon Valdez showed us in 1989 what can happen when a large oil tanker runs aground (which to be clear had nothing to do with IBM).

And as for the history books, ironically the IBM System/7 exhibit on the IBM website is very detailed, but there is literally not a single word about this remarkable side project. Clearly IBM sent the IBM System/7 Maritime Application/Bridge System to the bottom of the ocean, never to be heard of again.

That is until I found that image and updated Wikipedia.

Remembering how IBM had a best selling hand held productivity device in 1965!

In 1965 IBM released a truly epic hand held productivity device, so popular it was used by American Presidents and the famed actor Yul Brynner. And the story behind it is wild!

Yul Brynner wondering how he ended up in my blog (from Wikipedia)

The story begins in the late 1950s when Thomas Watson Jr ordered all divisions of IBM to double their business size in the next five years.

The Electric Typewriter business worked up some left field ideas to achieve this – deciding among other things that they would buy a dictation machine company.

They did an extensive search for an American manufacturer (since it didn’t occur to them that there might be better products overseas – which there were) and found the Pierce Wire Machine company, who manufactured magnetic belt dictation machines. They bought the company in 1959.

Now you might think buying companies was normal to IBM back then, but the gap to the previous business acquisition IBM had made was 26 years earlier (1933). Compare this to the year 2021 when IBM acquired 7 companies in a single year.

The key thing IBM liked was that this technology was transistorised (IBM had only recently switched to all transistor products) and used a magnetic belt (they liked magnetic things), which could hold 7 minutes (later 14 and then later 20 minutes) of dictated text. The belt was portable, durable and reusable and could be easily handed to a secretary who would listen to it using a similar looking product called a transcriber to type up the dictated text. The belt could be easily erased and could even fit in an envelope to be posted back to the office by road warriors (that’s a 1960s form of email for the kids out there).

Their first order of the day? Sack everyone! They had the previous owner fire the entire company (take that Elon Musk!) because they wanted the technology, not the people. Although they later re-hired a single sales guy called Sam Kalow, mainly because it turned out IBMs famed reps had no clue how to sell a dictation machine (hiring Sam turned out to be a good decision as he went on to have a legendary career at IBM).

They then discovered that the portable dictation machine they had been promised was not actually finished and they had to invest considerably to produce a klunking unit that weighed 3 kg and was not particularly portable (although the desk top units they got as part of the sale were a reasonable sales success). In the picture below you can see the magnetic belt on right and an IBM branded microphone.

IBM Executary Model 214 (Image from curiosmarc.com)

Given how hard the hand-held was to sell, IBM chose to design a completely new machine styled by the legendary Elliot Noyes and his associates (the Johnny Ives of the day who had designed the covers for the famed IBM Selectric Typewriter) and this product was remarkable.

It was handheld:

Small recorder or giant hand? You be the judge. (IBM Print Advertisment)

It was light!

Model cheating by holding the thing she is weighing (IBM Print Advertisement)

It was endorsed by Santa!

Santa dictating his naughty and nice list for his elves to transcribe (IBM Print Advertisment)

And it was beautifully packaged!

Very slick packaging from IBM! (Image from Wikipedia)

For those curious about what is in the package above, the IBM branded white rectangular object in upper centre is the magnetic erasure wand and the IBM branded cylinder in lower centre is the mercury battery (which is not legal any more). There are printed usage instructions in upper right and a soft bag to store spare recording belts in lower right. Finally the recorder in leather case and a microphone.

Once IBM had trained enough sales reps and got the incentive plan right, sales for their range of dictation machines boomed, driven by this amazing little portable device and by 1969 IBM was #1 in the US dictation machine market selling 98,000 units in a single year and driving at least one competitor out of business. The IBM 224 (and its follow up the 274) was so popular it was used by President Johnston and later President Richard Nixon who kept one by his bed (and we all know how his recording habits turned out).

Presidents Nixons and Johnston discussing their favourite recording technology (from archive.gov)

During the 1970s IBM went on to sell a variety of dictation machines using magnetic belt technology and when that started to lose traction in the marketplace they tried to reproduce their old days of glory with a home grown magnetic disk based product called the IBM 6:5. Sadly it was not a sales success although it was somewhat popular in the medical and legal professions, for example being used by Police detectives in the City of Raleigh to record their case notes.

IBM 6:5 Portable Recorder with hand-strap (from Wikipedia)

However the humble cassette recorder, launched into the mainstream by products like the Sony Walkman changed the cost expectations of the market and IBM had no choice but to try and offer a lower cost option. Their last portable dictation product was the embarrassing Executary 294, an inferior micro-cassette player manufactured by a third party (and proof that yes, IBM did at one point sell a personal cassette recorder).

IBM 294 Portable Recorder

When John Opel became IBM CEO in 1981 he ordered a review of the business and it was obvious that they had lost the initiative, so he withdrew IBM from the dictation machine marketplace. IBM would not offer another portable hand-hand productivity device until they began reselling the Palm Pilot in the late 1990s.

But in places like YouTube, the legend of the IBM Dictation Machine lives on – a truly legendary product in the history of IBM.

I have a day in court in Delaware – just like Elon Musk

So you may have heard about Elon Musks recent misadventures in the world of finance. In between repopulating the world with his many children, he made a bid of $44B USD for Twitter but suddenly decided he didn’t want to own a social media company. So now he will have his day in the Delaware Court of Chancery.

He seems to be taking this seriously

But would it surprise you to learn that I could also have a day in court in Delaware?

My tale of misadventure begins in 1989 when I started buying IBM stock using their employee purchase program. They would take a small % of your salary and buy IBM shares for you at a small discount with no brokerage. It was a good program (for the time), IBM paid regular dividends and the stock price kept going up.

The problem is the dividends came as a US dollar cheque mailed to Australia. And as any Australian who has ever tried to deposit a foreign cheque has found, this is not a welcome form of payment (the local banks want to charge a big handling fee). So I switched to the dividend reinvestment option and let IBM use my dividends to buy more stock. Problem solved.

Jump forward to 2012 and I decided enough was enough and left the warm embrace of IBM for a startup called Actifio (one of the best decisions I ever made). I happily sold my IBM portfolio and moved on (well I kind of moved on, for some reason I keep writing blog posts about IBM).

But then something weird happened.

Three months after leaving IBM a letter arrived. It seems I still owned some stock: Around $0.05 cents of stock to be precise. A micro share.

I presumed they would soon charge me transaction fees or some other such thing and this would go away.

I was wrong.

Every quarter for the next 10 years IBM (well Morgan Stanley) wrote me a letter:

We have your micro share! It even earned its own micro-dividends and attracted micro transaction costs.

My nest egg slowly grew. And regular as clockwork those letters would come, mailed from England.

I estimate IBM (or Morgan Stanley) spent at least $200 keeping me informed about the status of my financial nest egg.

But this gravy train has now suddenly and dramatically run off the rails. I recently received this letter declaring that I lost their loving feeling, and they felt I was lost to them. They declared my riches as lost property.

My precious money was going to the state of Delaware where I could retrieve it by going to court!

Now they did give me 30 days to stop this process, but this gracious period didn’t account for the fact that their letter printed on June 28 did not arrive at my house till July 31, the day after the 30 day period ended (even accounting for timezone differences).

Well played Morgan Stanley and IBM!

And so like Elon, I now have a choice. Do I take my day in Court in Delaware? Or do I accept that my financial fate is now in the control of others?

Decisions decisions….

Remembering when a camera company took IBM to the cleaners.

In 1981 IBM Office Products Division (popularly known as OPD) made one of the most spectacularly bad deals in IBM history. A deal that spawned a product that IBM could barely sell and was forced to withdrew in less than 2 years.

The background to this disastrous tale started so well, because during the 1970s IBM OPD  was wildly successful. Almost every product they made, from the beloved selectric type writer and type setters to dictation machines was a sales success; a market leader; loved by clients and hugely profitable.

Thats a dictation machine, not the worlds largest shaver.

But by 1982 the golden geese of OPD had mainly faded to unhappy turkeys.  Like a desperate record producer, they needed a hit and the solution they settled on was to release a desktop photocopier.

Only one problem: they didn’t have one.

To make matters worse their main copier product, the Series III had proven to be a reliability disaster.

So what to do??   

OPD Executives looked around and found that Minolta was willing to sell them a copier (the Minolta EP310) at what looked like a competitive price. Project Falcon ensued – an OEM photocopier product that would turn the tide and get IBM back into Americas offices.

They proudly named it to the Executive 102. This was an early promo poster:

I don’t know if it came with a gum ball machine.

The deal was duly signed and 10,000 IBM branded Executive 102s were loaded onto a container ship in Japan to set sail for the Americas. Great optimism was felt from the highest executive to the lowest sales rep.

But just like the men of middle earth who accept Saurons magic rings,  they were deceived.

IBM Executives after realising what they had just done

Because while the ship was sailing to San Fransisco,  Minolta was dropping the price of their own identical Minolta branded copier by 30% and telling their partners they should freely compete with IBM. Astonishingly IBM had failed to get price compete or non-compete guarantees from their OEM partner. 

It seemed there were no limits to how they could be out-competed.

So when the ship docked, IBM suddenly found that it was very difficult to sell the copier at anywhere near list price, something IBM OPD sales reps could (and would) famously do.

In fact it was nearly impossible to sell the damn things at all.  Senior Managers demanded solutions and settled on a simple one: Shame.

Boxes were literally stacked on sales reps desks. Every rep was ordered to sell at least one. Managers would stalk the halls, demanding to know who had not sold their allocated boat anchor.

To add insult to injury, IBM had a 90 day satisfaction guarantee.   So even if a marketing rep sold one, a  Minolta partner could appear at the customers door step like some angry avenging angel and simply swap the copier out at a lower price.   The torture seemed never ending.

And so in 1982 IBM withdrew the Executive 102 and drew to a close one of the least successful products in IBM history.

And IBM Office Products?  I will leave the rest of their riches to rags story for another day.

It’s Halloween, so let’s visit a graveyard…. an IBM Graveyard… filled with typewriters

Hey kids, did you to go trick or treating this Halloween?

Yes? No?

Ok… how about we visit a spooky graveyard scattered with the bones of products long since dead?

Thats right kids… its time to visit an IBM Graveyard.

In this series of articles I am going to take you to some of the more interesting graveyards and recount tales of great success and of miserable failure from my beloved IBM. Oh yes.. let’s go to one of IBMs graveyards!

But where to start?? What’s that sound? Is it the whiring of an IBM Selectric Golfball?
Why yes… its 1973 and the IBM Typewriter business is king! No not just king but Emperor! By some accounts IBM had over 75% of the electric business typewriter market at that time. In 1973 you did NOT get fired for buying IBM Typewriters…

But actually our story starts forty years earlier in 1933 when IBM bought a struggling typewriter company called Electromatic Typewriters Inc. It invested heavily in its new typewriter division and in 1935 brought out the first IBM Typewriter: The Model 01, which was an immediate sales success.

• Quiz question – did IBM bring out the first commercially successful typewriter?
• No! That was Remington – the same Remington who made guns!

However a true first was the IBM Executive Typewriter, announced in 1941 and first shipped in 1944. It literally changed the world of typing, since it introduced proportional spacing, just like a book printer could do. In this example of typed output you can see non-proportional font, every letter is the same width:

However the new IBM Typewriter used a proportional font meaning narrow letters were narrow and wide letters were wide. Suddenly typed letters looked like they had been sent to the printers.

• Quiz question – did IBM bring out the first commercially successful electric typewriter?
• No! That was the company they bought, Electromatic, although many companies made electric typewriters before that.

However while IBM continued to innovate bringing out more reliable and quieter models, the big change happened in 1961 with the introduction of the Selectric golfball typewriter. The ideas that IBM explored to create the Selectric were not new. The difference was that IBM was the first company to create a commercially and mechanically successful product using the concepts they contained. Ironically they also kept designing and manufacturing typebar style typewriters (with letters on individual arms), principally because a true speed demon typist actually could type faster on a typebar than a golfball. Still, the removable golfball meant one typewriter could produce over 40 different fonts (you would need 40 typebar typewriters to do that!).

• Quiz question – Did IBM introduce the first typewriter with a replaceable font?
• NO! That was Blickensderfer in the 1880s (yes you read that right, nearly 70 years earlier)

However it was the introduction of the Correcting Selectric II in 1973, one that could literally remove typing errors straight off the page, that literally blew away the remaining competition. At one time the delivery time on new orders went from 4-6 weeks to 6-9 months. In Lexington Kentucky, there were 5 manufacturing lines building them, each working 24 by 7. One of the five was a ‘clean’ line designed for tours. The other four dedicated to just cranking them out. In 1974 the Lexington plant made over 1 million correcting Selectrics. This feels like a peak time for American manufacturing. And there were also IBM typewriter factories all over the world (including Wangaratta in Australia!)

I personally feel the Correcting Selectric is something like IBM’s iPhone. Now if you think that’s a stretch, consider my reasons:

• It was truly pervasive, probably the most visible product IBM have ever released. It was also sold into the most clients, since it worked in every industry vertical and any business size as well as for personal use. More than any other product they made, IBM and their partners could sell into nearly any type or size of business.
• The Correcting Selectric generated a mania that reminds me of iPhone fever. It was full on. Once a secretary in the office got one, the clamouring from all the other ones began…
• There was a real focus on design, a focus that I have never seen on any other IBM product. They truly wanted it to look good. Compare the Selectric to this later 1967 Model D Executive (which is a TypeBar, not a Golfball). The difference is chalk and cheese.

The Selectric become a real symbol of a 1960s office. So much so that the TV show Mad Men chose to use the Selectric II in many episodes, even though the episodes set in the 1960s was showing a model introduced in 1971.

The colour choices were also outstanding, including, but not limited to:

• Raven Black
• Garnet Rose
• Emerald Green
• Pearl White
• Topaz Bronze
• Classic Blue
• Sandstone Beige

And it was common for clients to request specific colours, like University of Texas orange or Kansas State University purple. The sales reps for Law firms would send fabric samples from their desk chairs to ensure their typewriters blended in (unlike this cat).

Garnet Rose

Emerald Green

Classic Blue (http://ibmselectric.blogspot.com/)

Raven Black. (http://ibmselectric.blogspot.com/)

Topaz Bronze http://ibmselectric.blogspot.com/2016/03/topaz-bronze-selectric-ii-with.html

Marlin Blue Selectric III (http://ibmselectric.blogspot.com)

Pearl White Selectric III

Now checkout this video. It is short and perfectly explores how the GolfBall worked. What strikes me in this video is just how good the print quality looks even today.

Then watch this video and marvel at how it worked at all:

And shipping millions of typewriters meant 1,000s of service engineers. A CE (Customer Engineer) as they were called would routinely get a patch of 1000-1500 machines, often doing 10-20 service calls a day while trying to fit in regular Preventative Maintenance (PM) on each machine in their patch. Sometimes a client literally had over 1,000 typewriters in a single campus or building, so the CE never left the site. The HQ for JC Penney on 6th Avenue in New York City had 1100 typewriters – all in one building. The IBM CE became a familiar face, a part of the family (sometimes literally meeting their future partners while taking service calls), walking into schools and office buildings across the country.

These were the IBM Territories for the City of Melbourne in 1975. Territory 7 looked like way more walking then territory 10!

The Selectric also appeared in other products, like the IBM 7030 Stretch, which led to some amusing requests from the Large Systems engineers for Office Products Engineers to come and help them fix their mainframe typewriter.

Are they waiting for the typewriter repairer?

Of course they could also have asked their local KGB spy for help, since the KGB turned out to be expert at bugging US Consulate Selectrics. Below are Soviet era bugs planted in US Embassy typewriters in the USSR. Literally one of the first ever keystroke loggers:

Bugged Selectric Power Switches (found by the NSA)

Apart from being an easy device to spy on, it was also an easy device to lend out. These photos were shared to me by Henry Mydlarz, an IBMer who volunteered at the Melbourne Royal Children’s Hospital Good Friday Appeal in 1975. IBM lent the television station several dozen Golfball typewriters to be used in the donations room by the station staff. IBM also provided volunteer typewriter technicians to stand by round the clock and attend to any problems working in shifts. Note the colour of that gorgeous Raven Black unit that the child is playing with.

Under lights and camera… the operators are typing up those donations!

The fascinating thing about the Selectric is how slow the release program looks and yet how enormously successful it was. The Selectric came out in 1961. The Selectric II came out in 1971. The Selectric III did not come out till 1980! By the 1980s the enthusiasm to innovate in typewriters seemed to be gone. IBM instead was focused on things like producing one of the worlds first hybrid photocopier/laser printers (the IBM 6670) and one of the worlds first office inkjet printers (the IBM 6640).

With the release of the IBM PC in 1981, the focus was then truly elsewhere. If the goal in the past was to get a typewriter onto every desk, now it was to get a PC onto every desk alongside that typewriter. That launch hysteria of 1972 was a distant memory and companies like Olivetti and Xerox were competing with better and cheaper offerings. IBM released an electronic version of the Selectric III, internally called the E-tron, a term they had to ban when they realised Étron in French means excrement (as Audi found in 2019.) The nickname was appropriate though, as the machines extensive use of plastic parts and dodgy power supplies resulted in major client dissatisfaction and a much higher call rate than it’s all-mechanical predecessors. All the hard earned lessons about how to make a mechanical machine truly reliable seemed to be forgotten and replaced with a plastic carriage, not built for heavy duty work.

The Wheelwriter follow-on used a daisywheel, which was not an IBM invention and one that IBMs competitors (like Xerox) already had. Wang and other competitors meanwhile brought out high quality desk top word processors and the magic spell was truly broken. Although to be clear, the first company to claim they had developed a Word Processor was IBM themselves with their 1964 release of the Magnetic Tape/Selectric Typewriter or MT/ST, which could record typed material on a tape and allow the operator to ‘play it back’, having the typewriter type the recorded material automatically.

In 1991 the mighty Lexington plant, a site that produced among other things millions of typewriters and the legendary IBM model M keyboard, was sold to what became Lexmark. By 1995 the electric typewriter market was effectively dead, killed by the word processor and the PC.

And the Selectric? It became the favourite of typewriter enthusiasts and collectors and the goto story by IBM marketing. The typewriter, once a core part of “Machine” in International Business “Machines” went to the IBM Graveyard.

Which product should I cover next? Have you got a favourite IBM Graveyard you would like to visit? There are so many to choose from….

Remembering the time that IBM said all tape cartridges could have a good lie down

I recently stumbled across this fascinating study about what happens if an LTO tape drive is mounted vertically.

So rather than mounting it horizontally in a rack like this:

They tried mounting it vertically like this:

Now if they had rung me and asked if this was a good idea, I could have saved them a mountain of time by sharing my mildly qualified opinion:

“Hell no”

But apparently they didn’t have my number.

So instead they ran a whole battery of backup tests and worked this out scientifically. And yet they came to the exact same (but slightly longer) conclusion:

Based on the data from testing the new IBM LTO8 drive, when the drive is mounted in the vertical direction, the drive delivered significantly worse performance than in the horizontal position. In this calibrated test, LTO Tape Driver Doctor, during repeated tests, measured as much as 3X slower performance when the drive was mounted in the vertical direction as compared the horizontal direction. Furthermore, the drive reported an order of magnitude increase in write servo skips in this drive. Large servo skip count indicates the potential for large losses of cartridge capacity that will reduce the total useable capacity of the LTO8 tape drive. Based on this test, we recommend mounting IBM LTO8 drives in the horizontal direction.

Resulting in this brilliant and informative graphic:

The reality is that all cartridge drives starting with IBMs 3480 and marching on from there through the 3490, 3590, 3592 and LTO drives are all mechanically designed for the cartridge to be inserted horizontally. The mechanics of moving tape reliably are hard enough without complicating things by making the orientation of the drive optional.

So it is settled then: tape cartridges are horizontal creatures.

But here is something curious. Look through the square viewing window of this STK ACS:

Or this picture of IBM’s classic 3495:

What do you notice about the tapes?

THEY ARE ALL MOUNTED VERTICALLY!

And yet the carts need to be horizontal to go into the drives, making the robots job much more complex than it needs to be.

Compare this to the IBM TS4500 where the carts are all horizontal, making the robots life much easier:

So what gives? Why were early tape libraries designed to make things harder than they needed to be?

The answer lies in these reel to reel tapes (thanks Columbia University for this great wide-angle picture):

For any sort of reel to reel tape (including audio, film as well as data) it is a well established fact that flat storage even during transport could cause wraps of tape to slip and push against the flanges, resulting in edge damage and in some cases erratic tape motion. This requires all reel tapes to be hung vertically so that the weight is supported by the tape hub.

So when IBM brought out the 3480 cartridge, the same logic was applied. There was even an official manual shipped with every single cartridge drive:

 Care and Handling of the IBM Magnetic Tape Cartridge, GA32-0047

This declared you could only lay your cartridges flat temporarily:

Although cartridges are shipped and stored on their sides, you can lay the cartridges flat temporarily while moving them. The bottom of each cartridge has two raised areas that fit into the indented label area on the top of another cartridge. This construction helps prevent the cartridges from sliding while moving them.

Do not stack more than six cartridges.

However when IBM began work to add auto-loaders to their 3480 cart drives, they realised that tapes could end up sitting horizontally in the loaders for long periods of time (while waiting to be loaded or unloaded). So IBM began to study what if any impact was being experienced by this, literally running 1000s of cycle tests along with bench tests and scientific modelling. There was no tape servo positioning data in these early drives, meaning everything had to be written and read back and temporary errors and retries carefully monitored. It became clear that the cartridge leader block (which earlier reel to reel tapes did not have) acted like an anchor to the outside of the wound tape. So when properly rewound with the leader block properly in place, the cart could be safely stored horizontally.

So the requirement to store cartridges on their side was removed and because this happened while they were developing the 3584 (aka the TS4500) it allowed a much simpler robot to be designed. It also meant the above statement literally changed to just this:

Do not stack more than six cartridges.

IBM Distinguished Engineer Ric Bradshaw tells me that each time IBM has released new cartridge types (such as 3490 to 3590 to 3592 and LTO) plus extended length and density carts, similar testing and modelling continues to be done, with increasing scientific accuracy due to improved servo positioning data. This ensure new carts can always be safely stored horizontally.

And so allowing all tape cartridges going forward, to have a good lie down.

Sleep well sweet cartridge…

Remembering that time when IBM brought a beehive into the datacenter

I love the idea of biomimicry: the art of solving engineering problems by observing nature. If you have ever used a piece of velcro, then you have used something invented through biomimicry.

Japanese bullet trains (or Shinkansen) are also famous for using biomimicry, copying features of Owls, Adelie Penguins and KingFishers to dramatically reduce noise and vibration and resulting in their ability to travel at remarkable speeds.

Japanese Bullet Trains

But do you know that IBM also once embraced biomimicry? Thats right…. in 1974 the B in IBM also stood for BEEHIVE!

This all started in the late 1960s when IBM began developing something they had never tried before: a genuine robotic tape library. Developed at IBM’s Lab in Boulder, Colorado, the developers concluded the most efficient shape to store data cartridges was in a honeycomb array. The product they developed was code named Comanche and was announced as the IBM 3850 Mass Storage System (MSS) in October 1974.

It contained a huge number of firsts for IBM. For one thing it used a whole new circular tape cartridge that looked like this:

Each of these bullet shaped cartridges were two inches in diameter and four inches long, holding a spool with 770 inches of tape. The cartridges could hold 50 MB and were read diagonally. They were designed to be robotically fed into a whole new tape drive known as a DRD or “Data Recording Device” – pronounced “DiRD”. These drives were going to be called a “Tape Recording Device”, but that was changed when the developers starting shortening that to “TiRD”.

The carts were stored in a honeycomb library known as an IBM 3851 with two accessors (robots). Given the time period, it is very hard to find decent photos (or videos) but here are a couple of the accessor and the honeycombs:

This product was packed with firsts for IBM:

• First robotic tape library (with not one but two accessors!)
• First circular tape cartridge
• Brand new tape drive design
• Brand new Control unit (the IBM 3830-003)
• Disk virtualisation using automated tape. Thats right, this thing could take two 50 MB tape carts and use them to emulate a 100 MB removable disk pack. This meant the tape library could appear to the mainframe as a mountain of removable disk packs, with up 16 online at any time. This feature was phenomenal and users loved it because it gave them what seemed like unlimited storage for huge batch jobs and testing new apps with copied production data.

And in case you are unimpressed by 100 MB of disk space, compare the relative sizes (for the time), remembering those little carts were just four inches long….

Now at this point you may be thinking, wow this sounds fantastic – how come I have never heard of it?

The problem was that list of ‘firsts’: there was simply too much new technology all in one product. The MSS was a master class in complex and demanding microcode (firmware) and there was way too much dependence on exact mechanical performance. Unless the IBM Engineers who installed and maintained it were truly expert at their job, way too many things could go wrong. The pickers had to be very well aligned to get the cartridges in and out of their honeycomb positions. Different sections of honeycomb could be imperfectly aligned when assembled together, or worse the honeycomb itself could be distorted (or badly manufactured). If a cartridge was forced into an undersized honeycomb position it could cause severe damage to the machine and destroy the data in the cartridge. The accessors were also notorious for occasionally crashing into badly inserted carts or even dropping and losing carts.

War stories about the MSS abound, such as the attempt to detect badly sized cells using a slightly over-sized test cart. A library microcode engineer brought one to a troubled install at the Johnson Space Center in Texas. They began the test routine and within minutes it found a badly shaped cell. The accessor strained and strained to insert the test cart into the bad cell, trying so hard it tripped not only its circuit breaker, but a whole series of upstream circuit breakers, powering off the whole NASA data centre. A classic ‘what have you done’ moment.

Over-sized test cartridge that turned NASAs computer centre off

Another amusing story concerns the photoshoot for when the MSS was featured on the cover of the 1974 IBM Annual Report. The position of the accessors for the cartridges was based on light-based indexing. If anything went wrong they would park themselves in garages at either end of the aisle. The photographer was shooting at an actual customer location (Montgomery Wards Data Center in Lenexa, Kansas) and needed more light to capture the honeycomb, so he instructed the account rep to get inside the machine with an auxiliary flash. As soon as the flash went off, the accessor concluded it was broken and headed for it’s garage, moving at five mile per hour. The salesman’s comment:

“I never realized how fast 5 mph could seem until I saw that thing headed for my belly-button.”

He was not injured, but the unexpected accessor crash literally caused a system outage (oops).

An example of IBMs struggle to master the microcode was that while there were two accessors, only one could be active at a time. IBM could never get the two to work simultaneously, as they would sometimes violently crash into each other. IBM only managed to finally achieve this feat more than 20 years later in a different product, the IBM 3584, with a feature we called dancing robots, or officially “dual active accessors”.

Another problem with the product was IBMs obsession with moving things and people around (IBM at the time truly stood for “I’ve been moved”). Having developed and initially manufactured the MSS in Boulder, in 1976 they moved MSS manufacturing to San Jose, California to make more space in Boulder to manufacture IBMs Copiers. And having moved the MSS to San Jose, in 1979 they then moved it to Tucson Arizona to make more space in San Jose for disk manufacturing. And then having moved it to Tucson, they then decided to cease manufacturing completely at that site and moved MSS manufacturing back to San Jose. Each move lost experienced staff and resulted in an overall loss of “tribal knowledge”.

So things did not go well for the MSS. In fact the big hint that things were not going well was when in 1984, IBM came out with a whole new 200MB cartridge (for the new IBM 3480), that was…. square! It clearly could not be used in the MSS.

At its peak there were only 350 MSS subsystems installed world wide (peaking at 274 in North America). Rapid advances in disk sizes and disinterest inside IBM, all meant the end for the humble honeycomb robot. It was withdrawn in August 1986 (without replacement) and maintenance was stopped in 1991 (which was quite fast for those days).

The MSS is now best remembered by IBM Collectors, who love to show off (and sometimes also sell) their souvenir MSS parts. You can see from left to right a data cartridge that has been opened, a green customer engineer test cartridge and a yellow system engineer test cart, as well as a honeycomb segment.

In terms of classic IBM recycling, the cartridges came in plastic trays, which were perfect to put in your drawers to sort out small bits of stationary. It is quite possible people are still using these, with no clue where they came from or what they were originally for:

Want to read more? Check these out:

https://www.ibm.com/ibm/history/exhibits/storage/storage_3850.html

https://en.wikipedia.org/wiki/IBM_3850

http://www.columbia.edu/cu/computinghistory/mss.html

Remembering the time a giant yellow robot saved IBMs Tape Business.

In 1987 IBM’s Executives in Armonk hated Robots.

THEY JUST HATED THEM.

Now to be clear I am not talking about this kind of robot:

Lost in Space Robot and Dr Smith in 1965

Or even this kind of robot:

Arnold as a robot in 1984

No… the kind that IBM hated was this kind… the kind that moved tapes for a living:

Photo from 1992, ok I cheated a little with the timing

The problem was that while IBM led the world in developing automated and virtual tape libraries, none of these products had ever been a commercial success.

Here is a short list of just some of IBMs robot products:

• In 1961 IBM developed the Tractor tape library for the NSA, which turned out to be a bit of a one off.
• In the early 60s IBM released the slightly insane Walnut photostorage library developed for the CIA. Total sales: 7
• In 1974 IBM released the Mass Storage System (the MSS, known as the 3850). Which was in hindsight an overly ambitious product, years and years ahead of its time. It was one of the first IBM products to be totally dependent on microcode, had a disk virtualisation layer (tape being used to virtualise disk in 1974!) and was inspired by a bee-hive! It was also by all accounts unreliable and not economical, which led IBM to finally withdraw it in 1986.
• In 1982 IBM announced the RS 1 robotic Manufacturing System followed by the RS 2. However despite several updates and releases, by 1986 it was also clearly not a commercial success. Pictured is the IBM 7565 Manufacturing System, introduced in 1983 (I think the robot should have made the operator a taller chair).

The IBM 7565 Manufacturing System controlled by an IBM Series/1

So by the late 1980s the last thing any IBM executive wanted to hear was a proposal for another robot. It may surprise you to also hear that there were IBM executives at the time who believed tape was dead and further investment in tape was wasted money.

This is what one IBM Vice President publicly said around this time:

The market for a robotic tape library is negligible, since 90 percent of the 300 million computer tapes in the world do not need to be accessed at the speed of a robot.

Now for those of you who have read The art of possibility, guess what management at StorageTek heard him say?

IBM just said there are thirty million tapes that need robotic access! HELL YES, SHOW ME THE MONEY!

Now contrary to what some people think, StorageTek (STK) was not founded to make tape libraries. It was started way earlier (in 1969) by four ex-IBMers who planned to make IBM compatible I/O (like tape drives and printers). They actually then went on to drive STK to bankruptcy trying to build an IBM compatible Mainframe. Tape Libraries started as a skunk works project (probably by ex-MSS IBMers) and were not in the companies plans until well into the 1980s, just as they were coming out of bankruptcy. They released their first Tape Library, the StorageTek Automated Cartridge System (ACS) 4400 tape library in 1987 and its followup, the PowderHorn, in 1992.

In short STK rapidly found they were on a winner with a product that customers loved and by 1993 they had sold over 5000 ACSs and were selling into pretty well every IBM Mainframe shop in the world. Later their libraries (bizarrely and impossibly modified to make the robot more visible) even starred in movies like Clear and Present Danger and Eraser (staring Arnold, not playing a robot):

Scene from Eraser with Vanessa Williams in a chair

IBM was now in serious trouble in the tape business. Selling IBM tape drives into STK silos was VERY hard work… it was like a walled garden (literally). The company that basically created the magnetic media market was being driven from it. Something had to be done.

IBMs only automation solution was the addition of auto-loaders to their tape drives. Taking this (a string of IBM 3480 Tape Cartridge Drives):

And turning them into this by adding auto-loaders. The benefit was that you could pre-stack tapes in the loaders at the start of batch and empty them at the end. So an IBM Marketing Rep could argue: “Look, you can load up to 40 tapes at once!”

This attempt at automation was so meagre that Thomas Henkel of the Boston based Yankee Group thought it was part of some secret plan. He was quoted in Computerworld magazine in June 9, 1986 as saying: “I think IBM has some tape library on the way”.

He could not have been more wrong.

The follow up product, the IBM 3490, announced in 1991, offered really cool display panels, but only slightly larger autoloaders.

Meanwhile the STK reps counterargument to these auto-loaders was, “Hold my beer and keep your 40 cartridges, because my ACS can store up to 6000 tapes!”.

Which meant STK was selling 1000s of these:

So the real solution was obvious: IBM also needed a robot! But clearly to outcompete STK they needed a big robot… no… bigger… NO.. I SAID BIGGER…. AND BRIGHT FREAKING YELLOW!!!!!

And thus in 1993 the world got the IBM 3495.

The 3495 remains IBM’s longest ever product. A model L50 (the largest model) was 28 meters long (that’s 92 feet)! That’s crazy huge for a datacenter. The internal IBM Code name was Caballero, but in the field it was sometimes referred to as Conan (as in Conan the Librarian). The robot itself was a repurposed General Motors Fanuc automotive industry robot. And it was yellow… really really yellow.

The thing I love about this product is that the robot was designed to make cars in a production line. The robot was bolted to the floor and the cars came to the robot. But now the robot needed to come to the tapes and so IBM made it a buggy to ride in. Yes that’s right, IBM made a car, for a robot that makes cars. And not just any car, but a car that could move a 408 Kg (900 pound) robot at 9 kmh.

Now the 3495 was not a huge seller, I believe total sales were not much over 100. In my home country of Australia, IBM sold precisely 8 of them. One issue was finding the level floor space to install one. But it freaked STK management out (they would threaten to kick you out of the STK Users Group if you bought one), it generated huge buzz and held the line while a far more sensible product was rapidly being finished (as well as Virtual Tape Library technology).

There are two videos on YouTube that are worth watching. The first is a slightly amazing marketing video made in 1992 that I was given when I started blogging at IBM. Given IBM had no official portal for videos, I posted this on my personal YouTube channel, where it has amassed over 40,000 views, seriously out competing one of my most popular personal videos: my Uncle starting a Lanz Bulldog with a blow torch.

This second video is equally amusing, shot I believe at a Kraft datacenter, it also has some footage of other old kit at the end. But there is a scene that shows the robot in its little car! It’s about 58 seconds into the video, don’t miss it!

The 3495 was withdrawn in 1998, but by then IBM had the very successful 3494 which IBM are still shipping (with a great many improvements) as the TS4500. The IBM TS7700 Virtual Tape Solution (VTS) that attached to it was also hugely successful. In Melbourne we used the VTS to totally evict STK from several Mainframe clients, tearing out dozens and dozens of STK silos.

STK meanwhile was bought by Sun in 2005, who were then bought by Oracle in 2010, who still ship tape products and libraries under the StorageTek brand.

And where did the big yellow robots go? Well they did not go onto to become our new robot overlords in their IBM designed cars. However I know at least one of them is still doing loyal service… guarding a letter box in Canberra. Possibly the finest recycling effort of an IBM product that I have ever seen.

Stand strong big yellow robot, you helped keep IBM in the tape business.

In 1957 the CIA asked IBM to invent Google Images, and IBM said yes.

This story starts with a scene straight out of a spy movie and I swear nearly every word is true.

In the mid 1950s, a CIA spook comes to IBM with a request: “Hey Blue suited dudes, make us a machine that can store millions of images that are searchable by keywords. In other words a data lake of images – just like Google Images, except Google is still 40 years in the future.”

The IBMers didn’t even blink (or ask what a Google was). They replied, Hell yeah!” (or something similar), and the resulting product (code named Walnut) was delivered to the CIA in 1961.  

It kinda worked like this:

• The Walnut has 100 large cylindrical carousels called document stores.
• Each document store held it’s data in in 200 small boxes that IBM referred to as cells.
• Each cell contained 50 strips of photographic film
• Each strip of film contained 99 photographs arranged in a 3 by 33 grid.

In total, each document store contained images of 990,000 documents, and up to 100 document stores could be used in a single Walnut system, for a total storage of 99,000,000 pages

Users would look up keywords stored on a separate system that used an IBM 1405, identifying individual documents to be retrieved. The index produced punched cards that were inserted into the Walnut. The Walnut system then retrieved the documents, copied them onto a blank film strip and developed them. The output could be viewed using a photo-negative viewer, or developed for full-sized printouts.

So to get this right, our CIA Operative searches for photos of villains using keywords and gets a list of images on punched cards. They then feed them into the Walnut, which then spits out a negative they can use to develop photos.

Imagine the James Bond scene: “Walnut, show me all images of Blofeld!” (followed by a 10 minute wait involving punched cards and photo development).

This is just like Siri, but with punched cards! (now hang on, which one is the real Blofeld?):

IBM tried to sell a commercialised version of this but got no takers. It seemed no one wanted to buy a product called Walnut (although it was now known as the IBM 1350).

However IBM realised that if they stored actual data on those negatives, rather than images, they might have a truly giant online read-only file system.

So they then productised this idea in response to demand from the Atomic Energy Commission to store 1 terabit of information to support super computer simulations. Yes you read that right – 1 whole terabit… that’s 128 GB! At that time no single system could address that much data.

The resulting product however was mildly insane and bizarrely huge and was an incredible mix of pneumatics and robotics combined with an electron gun and automatic photographic development. The product looked nothing like this:

It actually looked like this (a little bit larger, but how could you not love those blue covers?):

The Lawrence Livermore National Laboratory also loved it, using it for over 10 years as a (for that time) enormous online archive of read-only data (albeit with a response time in minutes). But despite the enormous amount of data you could potentially store in such a beast, it proved as equally impossible to sell commercially as it’s photo storing cousin.

The Computer History Museum in Mountain View has the only remaining example of an IBM 1360 (from Lawrence Livermore National Lab) shown in the following video. Note the tour guide claims it was retired because all the Service Reps who knew how to fix it had retired, but the truth is IBM no longer wanted to make spare parts for it and the client could not find any one to take over.

You can learn more here:

https://www.computer-history.info/Page4.dir/pages/Photostore.dir/index.html

and here: https://en.wikipedia.org/wiki/IBM_1360

And if you want to read about another story about IBM and the NSA, check out this one:

https://en.wikipedia.org/wiki/IBM_7950_Harvest

and here: https://www.ibm.com/ibm/history/exhibits/specialprod2/specialprod2_2.html