 Hello everyone, I'm Sam Lemley, curator of special collections at Carnegie Mellon University Libraries. Welcome to this third episode of Coffee with a Curator, a video series in which I share some of the many fascinating objects, artifacts, and books held in CMU's special collections. So today's episode is actually pre-recorded and broadcast as part of an event that was sponsored by the CMU Alumni Association and it was filmed on the fifth floor of Hunt Library in the Hunt Institute for Botanical Documentation and I'd like to go ahead and just open by thanking the institute and its director in particular Terry Jacobson for hosting us for filming this event. I'll link to the full event which features more objects in the video script description below and I think in our next episode, episode four, I'll be back sharing something directly into the camera with you. So thanks for joining me, I hope you enjoy and more soon. Thanks. The first thing that I want to share with you tonight is this book. It's a single volume of a scholarly journal that was published by the Berlin Academy of Sciences in 1710 and for those of us that are familiar with the way that publishing works in academia, the idea of a scholarly journal will be somewhat familiar but at the time in 1710, this was an incredibly new and groundbreaking genre and I think the real advantage of this form is that it passed the cost of publication from the author, from the printer, off to annual subscribers to the particular journal and what that meant is that scientists and particular mathematicians but also archaeologists, linguists, really anyone working in sort of niche fields were able to publish very short form discoveries and quickly. These were sometimes issued monthly so I think you really can't overestimate the impact that this genre had on the history of science and its advancement, its rapid advancement particularly in the 17th and 18th centuries. So in this particular volume, there's a very short article by the German mathematician Gottfried Wilhelm Leibniz and this article is titled Brevis Descriptio Machini Arithmeticae or a brief description of an arithmetic machine and you can see that this is something that Leibniz is describing and announcing in the first line. He says he invented it in his adolescence or rather modestly and in 1673 but that's interesting in itself right so it's only published in 1710 decades later that's an indication of the amount of refining that Leibniz had to do on this machine that he's describing. This is the first published account with an illustration of a mechanical computer and Leibniz was sort of his design was groundbreaking in a number of ways which I'll get to in a minute. So what is this machine? Well if we turn to an accompanying illustration at the back of the book there it is right so it's kind of an odd looking thing it's in a wooden box with a handle some hinges there are these small windows at the top dials here and then this sort of hand crank so you probably can see that we have something right next to me on the table here resembling this illustration and in fact this is a replica of Leibniz's 17th century calculating machine. This particular model was built by an Italian model maker Roberto Guattelli in the middle of last century and it came to CMU with the Trauma Cordic Collection just a couple years ago. So you can see a little bit in a little bit more detail what's going on with this machine unfortunately we can't take the casing off to show the the sort of intricacies of its mechanics but you can kind of get a sense of how it would be operated so there again on the upper edge there are these 16 windows and that's where the solution would appear so you could calculate a solution up to 16 digits and then there's another small line of small windows here where you enter the digits you're going to be calculating with right so Leibniz's machine wasn't entirely groundbreaking there were actually earlier mechanical calculators much like this probably most famously the one that Blaise Pascal invented and designed a couple decades before Leibniz's work Leibniz worked on his but significantly Pascal's only did addition and subtraction right Leibniz found a way to build a machine that was capable of all four arithmetic operations right subtraction addition multiplication and division so you can imagine just the the complexity of engineering and also the skill of the people actually tasked with building this thing it's it's a monument of engineering and mechanics so you know this is I think I'm starting with this at these this pair of objects because it kind of encapsulates what I think we want to accomplish in special collections at CMU you know we're really interested in telling the history of computing over the long duree right sort of from the 17th century when you sort of get these early experiments in mechanical calculation all the way through you know mid-century pioneers like Alan Turing, Grace Hopper, Von Neumann, etc these are all names I'm sure this audience will recognize so but what is the genealogy right between through and between those centuries and how can we represent that genealogy and that history in the form of a collection so that's that's that's the main reason why I wanted to start with these two artifacts from Leibniz the other reason is you know in so far as a discipline can be said to originate in one individual I think the history of computer science arguably could be said to originate in Leibniz and his immediate circle because he was not only working on the design and and and sort of operation of mechanical computers calculators he also developed a system of symbolic logic and probably most famously he developed the system of binary code one and zero that influenced modern binary code that's still used today so you know he's working on these sort of three strands of thought right symbolic logic binary code and you know mechanically assisted computing that are still with us are still with us in the in the history of computer science you know just across campus too you know and I'll just kind of a brief aside something else another treasure in the collection that we have of Leibniz's is a first edition of his nova methodus which is his calculus right so Leibniz is probably most well-known for his invention independently of Newton of the calculus so this is this is the first appearance in print of the calculus 1684 and again this is in an annual volume of another scholarly journal called the acta eruditorum it was published in Leipzig so very very common for Leibniz to publish in this form he published hundreds of articles like this over the course of his career and this just gives you a sense of the the range of activity right that that he was up to over the course of his life with this next object we jump forward in time three and a half centuries to another very familiar name in the history of computer science and that's Alan Turing so this is a volume of the academic journal another scholarly publication mind which specialized in human psychology which might strike us as odd Alan Turing is obviously dealing with computer and machine intelligence not human intelligence but he saw a close connection there so this is volume 59 it was published in 1950 and I'll turn to Turing's article now so the name the title of the article is computing machinery and intelligence and this was an incredibly groundbreaking event in sort of the history of the theory of artificial intelligence it's where Alan Turing first described what became known as the Turing test but what he called the imitation game and that's just a sort of thought experiment in which you know you test a computer for a human standard of intelligence of course famously we have still not built a machine that reaches that standard so Turing in this article of course introduces that concept that standard and it was kind of groundbreaking because before this the idea of a computer intelligence machine intelligence was a bit vague right no one knew how exactly to determine whether a machine whether a computer was in fact intelligent so he sort of wandered into this debate and came up with a fairly brilliant way of determining this kind of controversial question but I think what makes this article really significant is actually the second idea that you find in its pages and I'll turn to a subsection of the article which Turing gives the title learning machines right so you can see there section seven learning machines so again before this article it was assumed that you know machines could only do what they were told to do right and that was it machines Turing says were often thought to never be able to surprise human beings of course he responds by saying actually machines surprise me all the time and he has this really enduring view of the potential to actually program a machine a computer to behave in a way that was similar to the child's mind right so this was hugely groundbreaking the assumption had long been been you know let's program a computer to resemble an adult mind but what Turing said was no let's let's consider a heuristic model for programming a machine that might attain artificial intelligence so that we'd actually teach the machine through sort of you know question and response interactions rather than trying to sort of anticipate every possible scenario and program for every possible scenario so it's like I use the word endearing because I really like Turing's later papers his later scholarship for this sort of boundless optimism that they show you know after his academic work first at Cambridge and then at Princeton and then his wartime service of course at Bletchley Park in the crypt analytic unit which I'll actually be seeing at Enigma machine in just a moment so after that sort of official work and academic work he joins the University of Manchester and a lot of his work in this period in the early 50s right before he dies it's much more speculative and imaginative and I think this article is a perfect example of that imaginative range that Turing was capable of and literary style he writes in a really engaging way one thing that I really like about this article is the way that it ends so I'm going to turn to that line and he writes we can only see a short distance ahead but we can see plenty there that needs to be done right so again this sort of sense of boundless optimism and you know applying our imagination to these fairly technical problems you know if we do that if we accomplish that sort of imaginative element in this work really we can sort of find out amazing things so you know there's also a tragic element to this story because of course four years after the publication of this article Alan Turing would die by suicide after being convicted by the British government for a gross indecency basically being prosecuted for his homosexuality and he was forced to undergo a regime of chemical castration essentially so just terrible end to a brilliant life and the loss is somewhat incalculable you know what what would he have contributed had he been given another 50 years of life we just can't know another reason why I'm sharing this book is I often get the question you know what makes a rare book rare and I think this is an example of a book that might not look the part but is rare for a variety of reasons you know we tend to assume you know a rare book is not only valuable and significant in history but it looks good and as as you can see with this one it's not a very good-looking copy of this book you know there are library stamps on the edge this is in sort of a basic library what's called a buckram binding and there are actually pieces of tape holding pages of Turing's article together so you know and there there are copies out there in the anti-crane book market that come up at auction that are in pristine condition that have the original binding but you know I kind of like that this copy is a bit beat up because this is CMU's copy as recently as two months ago this copy was in circulation CMU students could come in and check it out so and that's sort of an ongoing project in special collections in the libraries here it's something that we've been able to do during quarantine but we're sort of methodically going through the catalogue and looking for important articles in the history of computer science much much like this one Turing's here you know articles by people like John von Neumann, Claude Shannon, Grace Hopper right sort of pioneers that are contemporaries of Turing and we're identifying those as culturally significant particularly in this story that we're trying to tell about the history of computing and removing them to special collections right so that we have you know an original copy of you know the the material artifact of the the origination of that idea contained in them but you know and there's also the there's also the aura of provenance in this particular copy which I can wax poetic about you know this was the copy of Turing's computing machinery and intelligence that was in circulation when Herb Simon was working right here so you know it has it has the history of place right in it and for that reason you know I don't particularly care that it's beat up it's CMU's copy so that's that's just an example of how you know books despite condition despite appearance can become rare right particularly as the ideas they contain kind of transcend to cultural relevance and importance and that's certainly true of Turing's book here so next as promised these are probably the most well known objects in the collection if not the most immediately recognizable at least when their cases are closed these are two enigma machines they're they're cipher devices used most famously or infamously by the Nazi military in World War II and we have a three rotor model and that'll make sense when I open them up and a four rotor model and I want to start with them closed because you can see the original serial numbers at the back of the boxes here and here so what I'll do is I'll turn them around carefully you can see that they're encased in these sort of oak veneer boxes that have handles on them so that they would have been you know portable in field service and then on the backs there are these metal latches which open and then very gently there we are so I'll move that one there that's the three rotor model you can see the three rotors and this one is the four rotor this one carefully set that down there so these are of course famous objects they have a very dark history naturally being used by the Nazis but they're also relevant to the collections we're building not least because Alan Turing and we just met worked on deciphering the enigma code at Bletchley Park during World War II so the enigmas were or the enigma machine was originally invented by a man named Arthur Scarabius who was a German inventor in the 1920s and the original object of the enigma machine was actually to disguise commercial secrets right business secrets so that's who they were sold to initially but the German military of course very quickly recognized the security that they provided and secret communications and co-opted the technology refined it developed it further and both of these are examples of the later sort of 1930s 1940s era enigmas that have these sort of plug boards at the front gently fold this down right so I mentioned that this is the three rotor model it's the earlier of the two and primarily used by the German army and also the air force so you know again this is kind of like Leibniz's calculator earlier in that you really can't see the mechanical complexity of the device without removing the case but it's electromechanical it was powered by a battery and you know very very basically when you press one of its typewriter like keys you initiate an electrical current which passes through a secret you know electrical route to one of these windows each one showing a letter and that window would light up right and also every time you pressed one of its keys the leftmost rotor would advance one step right and then once you got through all positions of the leftmost rotor this one would then move forward once kind of like an odometer on an old car right but what that meant is that once you set up you know the three rotors in the initial position right with every press of a key the settings would change right so that if I were to example for example press E three times each of those three times it would be assigned another cipher character to hide it right so that's what made it almost unbreakable right because basically the settings of the cipher would change with every single character of of what the plain text entered in to become the cipher text but that wasn't enough right for the german navy and so the four rotor model was developed which basically added again another rotor another level of complexity to the cipher system and the reason why the navy the german navy the nazi navy developed the four rotor model was to protect communications regarding its very important u-boat fleet and you know if you if you look at these one thing I really like about the two enigmas we have this one is in impeccable condition you almost wonder if it ever saw a service in the field whereas this one is a little bit beat up it's the the metal components are more rusty the sort of finish on the casing is kind of worn away it has this patina and I like to think that that's because it was exposed to sort of the salt air of you know its naval commission you know I don't know but but I like to think that so these are naturally incredibly rare objects you know the Nazi people soldiers in the Nazi military were required to destroy the enigma machine if they were ever captured or at risk of imminent capture so the fact that these survive you know is kind of an accident of history I think there are about 350 surviving examples of the three rotor model and about half that number of the four rotor model which is far rarer so these are in the trauma cordic collection were incredibly lucky to have them and already they've kind of made possible really interesting educational programs in fact students graduate students came in to special collections about a year ago year and a half ago and took these apart right actually look at how they're put together and that was an initiative of a program sponsored by the host program the history of science and technology group at CMU so there's there's huge potential here right that you can learn people studying you know block ciphers or you know advanced encryption methods it kind of starts here right the history of cryptology and these join a number of books in the history of cryptography and cryptology and so we're sort of this is sort of a side a side branch of this this history of computing collection that we're starting to develop we have several things in the history of cryptography in addition to things in the history of computing and of course those two fields are kind of inextricably linked but you know I'll just end by saying that it's there's again that aura having these shelved nearby works by Alan Turing you kind of have that history that relationship that he had with with these very dark machines kind of at the height of his career encapsulated in the form of a collection on shelves so this last thing that I'll share is actually a new acquisition it's something that we purchased for the collection just last month and it arrived just last week so this is actually the first printed book on the subject of robotics it was published in Venice in 1589 so it's quite an old thing and I love the it's an Italian translation of a classical work by Hero of Alexandria on mechanics and automata which are sort of early robots but the Italian translation I think is of the title is really wonderful it's on automatons or machines that move themselves right which is I think a brilliant definition if you will of robots so you know there was there was a 17th century and late 16th century fascination with robots and automata there's sort of you know advanced puppets in a way but again you know as I was talking earlier about the sort of genealogy of the history of computing you know there's a there's a comparable genealogy in the history of robotics and this gives me an opportunity to call out my really brilliant colleagues in the university archives who are currently working on putting together a robot archive right a sort of archive of CMU's work and innovation in innovations in robotics and the story of robotics sort of from a more contemporary lens so you know I view it as special collections role to tell the prehistory of that story and I think I would argue that that prehistory really starts here with this book on the sort of reception of classical mechanics and the sort of effort by Renaissance scientists to build machines that do things like robots might do today so I'll give you one example with an illustration here which I just I really like it's an image of a machine that cuts wood automatically right so you have this sort of awkward apparatus the gear with these sort of sharp teeth and these two weights which actually power the machine the idea being you know I'm not sure this actually worked but you know the idea being can we construct an automata that would be capable of doing something that might be viewed as difficult by humans right cutting wood or something comparable