Advent of Computing - Episode 108 - The Mundaneum, Part II

Episode Date: May 21, 2023

This episode we pick back up where we left off. We are looking at the roots of the Mundaneum, the applications of the Universal Decimal Code, and how it call connects to hypertext. Selected Sources: https://web.archive.org/web/20051227184732/http://people.lis.uiuc.edu/~wrayward/otlet/xanadu.htm - Visions of Xanadu https://www.ideals.illinois.edu/items/4184 -- Selected Essays of Paul Otlet              

I kind of keep hitting this recurring theme lately. It turns out that when discussing tech history, we don't really get clean firsts to hinge stories off of. It's a bit of an obsession for us humans to put things in, you know, nice orderly boxes. Events have to follow a certain sequence. Therefore, there must be some first event in that sequence. But that's not always the case, and that can be a little confounding and a little confusing. Take, well, the advent of computing as a great example. The first electronic digital automatic computers appear at the tail end of the Second World War.

automatic computers appear at the tail end of the Second World War. But there were actually a smattering of earlier machines that also fit that bill. Computers like the Atanasoff-Berry computer were electronic, they were digital, and they did fancy math. That machine was built and operated back in the 1930s. Going even further back, we have analog computers that existed decades prior to the ABC, in between our strange transitionary machines with relays, gears, cogs, and motors. The point is, we don't get a nice cutoff. We have this gradient of solutions to the same problem. Folk don't like doing math, it turns out, so they work up ways to make it more

easy. That, perhaps inevitably, leads to machines that can crunch numbers automatically. The technology changes, but the intent stays the same, and it has for perhaps hundreds of years. Computers aren't the only example of this blurry starting point. Another example that's vexed me for years at this point is hypertext. It's kind of the quintessential futuristic technology. Even early visionaries in the field describe hypertext as this utopian thing, as something that will make the world better. It's the technology that was going to reshape the whole world for good. People were saying that as early as 1945. And the best part is, they were right.

Now, I should back up here because I did make a little slip up. What exactly is early when it comes to hypertext? Should we think of that period as the 1950s, the 40s, or is it sometime even before that? Welcome back to Advent of Computing. I'm your host, Sean Haas, and this is episode 108, The Mundanium, part 2. As the name suggests, this is the second part of a series. In fact, this is going to be the conclusion to my shows on Paul Outlet's Mundanium. Normally when I do these series, each episode can kind of stand alone, but that's not going to be the case here. Last episode we discussed some of Outlet's earliest papers and the early development of the Universal Decimal Code, the U here. Last episode, we discussed some of Outlet's earliest papers and the early development

of the Universal Decimal Code, the UDC. Today's episode is going to be heavily based off the groundwork laid in part one. As such, I have to encourage listening in order. I usually try to avoid that, but this time it's going to be a must. That said, I do want to drop a quick recap so we're all on the same page. Remind listeners of what happened two weeks ago. In 1892, Outlay publishes this article called Something About Bibliography. This was the start of his path, at least his paper trail, that leads to the Mundanium. The key takeaway from this article is that Oatley doesn't think books are a reasonable way to store human knowledge. Sequential access, well, that just

doesn't cut it, especially as we accumulate more and more information. He further argues that as more information accumulates, we will eventually hit a point where scientific progress is impossible. hit a point where scientific progress is impossible. Thus, we have to have a better way to organize and access data. Put another way, Outlay identified the information problem back in the 1890s. His solution is non-sequential data storage. This, to me at least, is a shockingly modern analysis. Outlay is on the same page as Vannevar Bush in a lot of ways just about 50 years earlier. That forms the basis of the Mundanium. The idea is to liberate data from books and put it onto organized cards. Each card has a fact that has a place in a larger organization scheme. The tool used to manage this data was the UDC, a system that

was initially inspired by the Dewey Decimal System. However, the UDC isn't just a knockoff of Dewey's work. It's a full syntax for describing knowledge. Numbers are built from tables of context-free aspects, which allow for an infinitely flexible indexing scheme. It may look like the Dewey Decimal System on the surface, but it really blows the closed-minded Melville right out of the water. This idea culminates in the Mundanium, Outlay's attempt to chronicle the sum total of all human knowledge. It's often called an early precursor to the internet, and that's true in a sense, but I think that phrases things in the wrong way. The modern internet isn't a descendant of the mundanium any more than the mundanium

itself was an attempt to create the internet. Both of these things are attempts to, well, house and organize all human knowledge. The comparison is made to the internet because that's the biggest and best pass at that lofty goal. I prefer to think of all those attempts, the Mundanium, the Mimics, NLS, Xanadu, the internet, and everything in between, as different approaches to this shared problem. Each project shares parts of others, but it's not really a family tree. It's more like a web of connected hopes, dreams, and technologies. I'd wager this is a useful viewpoint because it lets us take a more open approach to these older projects. We don't have to spend our

time looking for direct lines between 1890 and 1990. We can disregard lineage, at least to a point, and focus on how each project tackled the information problem. We aren't looking for who read what papers or who attended which lectures which led to which patents. Instead, let's look at ideological similarities, even if they may have been developed in a vacuum. I'm giving this long-winded introduction partly as a reminder to myself. I used to be one of those people that didn't get why an abacus would show up in a computer museum. At one point, I just assumed the history of the computer started in 1945. How can something that doesn't even have a logic gate have anything to do with

computing? That's a really easy viewpoint to have. It reduces the story to a pretty short and pretty straight line. It also leaves you with a much less interesting story. So while the mundanium might not look like the hypertext we know, we can start to see bits and pieces of the familiar systems. We can look at this earlier technology as a different approach to the same problem of managing data. What are the finer details of the mundanium? Well, it's time we get into that. Today, we'll be looking at the actual opening of the mundanium, and its closure, and its grand reopening. There are going to be some ups and downs along the way. Throughout all of this, I want to keep the high-level view in mind. How is Outlay tackling the information problem? How does the technology of the time impact his

solution? And, most of all, what familiar ideas does he land on? Before we get started, I want to give my usual plug for Notes on Computer History, my attempt to build a non-academic journal for the history of computing. Right now, I'm sitting on, I think, enough articles for the next issue. I have one outstanding draft that should be in very shortly. So hopefully, not setting any firm dates yet, but I think in the next month or so, the first issue should come out. That said, I still want submissions. If you have an article or a draft article, or if you just want to write about the history of computing, then please get in touch with me. If it doesn't make it into this issue, it will definitely make it into the next. You can find all the details for that at

history.computer. Now, let's get back into the actual episode. Would it surprise you to know that hypertext exists in science fiction? Well, it kind of does. Let me explain. Isaac Asimov is one of my favorite sci-fi authors. When I was in middle school, I read a lot of his short stories, and in college, I somehow made time to read the Foundation series. It's one of those generation-spanning space operas. I can highly recommend the core trilogy of series, especially since many of the old pulp printings have these nice small pages. It's really manageable for me. Anyway, there's this plot device throughout

the series called the Encyclopedia Galactica. The foundation of which the series is named is started to create this grand encyclopedia that contains all the knowledge of the galaxy. At least, that's the cover story. Intrigue ensues, and the full encyclopedia isn't completed on schedule. By the later books, the encyclopedia is actually complete. At least, it's finished in the background. In Foundation's Edge, one of the sequels to the main series that's published in 1982, Edge, one of the sequels to the main series that's published in 1982, the encyclopedia is described as something like a computer database. Its articles are stored digitally, they're updated regularly, and they're instantly retrievable. We don't get a lot of detail on the encyclopedia, but it's clear that it's used as a knowledge tool. And a fun literary device in some cases.

a knowledge tool. And a fun literary device in some cases. An encyclopedia doesn't necessarily mean a shelf full of dusty old books. That's just the older use of the term. I bring this up because the Mondanium is sometimes called an encyclopedia. That can make it sound like the stale tomes of old, but that's not the case. Asimov describes the Encyclopedia Galactica as a living, breathing thing. It's a collection of human knowledge that's constantly being updated and amended. To readers in 1982, that may have sounded like a dream that was just around the corner. Computers were only starting to enter the home. Networking existed in institutions and small-scale communications were possible with microcomputers and modems, but nothing on the scale of the Galactic Encyclopedia was possible.

At least, not yet. More modern projects, like Wikipedia, have shown that this dream was actually right around the corner, just a number of decades away. A universal living encyclopedia does exist, we just don't really call it by the exact same name. Alright, so here's the twist. Here's what I like so much about the history of hypertext, and what keeps me coming back. A universal encyclopedia already existed before Asimov ever started his career. It's very likely he didn't even know about it. The Mundanium is an encyclopedia from the turn of the 20th century.

Unlike its contemporaries, it actually lived. So banish the thought of tomes on shelves, because while we aren't going to be talking about encyclopedias, we aren't going to be talking about any leather bindings. Now, it's a little hard to quantify when the Mundanium actually begins. The ideas are all in place by the middle of the 1890s. Outlay describes large chunks of the project as early as 1892, so ideologically speaking, the thing's old. As far as nailing down a date, well, that gets really slippery. The problem is twofold, or onefold, I think it's mainly the

language barrier striking yet again. I've made no secret out of the fact that I've been working from translations and secondary sources for these episodes. Some articles that I've made no secret out of the fact that I've been working from translations and secondary sources for these episodes. Some articles that I've read claim the Mendanium started in earnest in 1919. Others claim that there were proposals written as early as 1910. That's a bit of a gap. If it was a matter of a few years, I wouldn't care, but this one irks me. There were some big events between 1910 and 1919. It's probably useless to dwell on this, but I just have to illustrate the confusion here.

One of the first things you find if you're looking into the Mundanium is this article on JSTOR Daily called The Internet Before the Internet. on JSTOR Daily called The Internet Before the Internet. It claims that the Mundanium was proposed in 1910, and they have a citation to Wikipedia. That citation just leads to the wiki article on the Mundanium, which also claims a date of 1910. That is cited to a New York Times article that's behind a paywall. Once you deal with that, however you choose to deal with that, the article doesn't even mention the year 1910. This issue is made worse by the fact that Autlay and Henry LaFontaine,

his longtime partner in crime, had a whole lot of simultaneous projects going on. The website for the current incarnation of the Mendanium gives us a nice little timeline of events in this period. In 1900, the UDC is presented at the Paris World's Fair, which is the system's first big public outing. In 1905, a codified version of the UDC is published, allowing for its adoption by libraries. a codified version of the UDC is published, allowing for its adoption by libraries. In the years between 05 and 10, Oatley and LaFontaine found and reshape a number of three-letter groups. You get this maze of three-letter names and three-letter-dash-three-letter names and changes of names that eventually take us up to the Mundanium. Perhaps the most consequential development in this period

was the creation of the Offices of Documentation. These were semi-independent groups that operated under the IIB, the International Institute of Bibliography. That's just one of the handful of organizations that Otley and LaFontaine founded in this period. The IIB was to serve as the larger organization that would oversee the universal repertory, Otley's first pass at organizing all knowledge. But a key point in the repertory was that it wasn't some monolithic thing. It was broken up into smaller collections that were managed by experts. The offices of documentation were the mechanism used for this division of data. This is yet another point where I have to stop and discuss some of the terminology used here. Each of these offices managed their own encyclopedia.

In this sense of the word, an encyclopedia is just a collection of organized data. I think the closest analogy today would be something like a database. It's a set of information that's structured for easy and quick retrieval. The documentation part here is also a little weird. Coming from a programming background, I'm used to documentation meaning, well, written words that describe a program. Ault-Laye's conception of documentation was pretty different. Around the turn of the century, he adopted the word as a way to describe a new idea. To Ault-Laye, a document was anything that could convey meaning or store information. This meant things like books, cards, and loose

paper. It also meant images, engravings, archival objects, and this weird new thing called microform photography. That means that, really, anything can count as human knowledge. That expands the horizons of what we should consider the, quote, sum total of human knowledge. It also complicates matters for a bibliographer. Now, as a bit of an aside to the aside, Otley also seemed to be a stickler for words. In The Science of Bibliography and Documentation,

he opens with this passage that I just have to share. I think this will give you a good idea of the sources we're working with. Quote, Inappropriate terms lead to erroneous or imperfect concepts. When one consults any dictionary taken up by chance, one reads bibliography, science, knowledge of bibliography, bibliographer, one who is acquainted with books, editions, etc. One reads, The vagueness of such notion is striking. Not having a precise definition, one must first of all find out what knowledge of the book has involved up until now. End quote. Issues with translation aside, I think the vibe here comes through pretty

clearly. I've seen a few researchers refer to Outlay's body of work as scattered and repeating. So, back to the main point. What are the implications of this expanded view of the document, and why does Outlay seem to care about this idea so much? In this early period, Outlay is trying to establish bibliography as its own scientific field. He's trying to apply the same type of scientific rigor used in fields like physics to the study of information. I may be making a leap here, but this is how I see it. Outlay is attempting to generalize the new field. I'd wager that he saw that as a step towards carving out a real field of study, an actual science instead of just knowledge of bibliography.

We can look at physics as a good analog here. The field covers a very wide range of topics. a good analog here. The field covers a very wide range of topics. I'm trained in astrophysics, which is concerned with stuff happening out in space. Within that subfield, there are further delineations. Back in the day, I worked on the theory and computational side, which means I'm actually pretty useless with a telescope. I can't go out and identify constellations worth a darn. Each of these subfields and sub-subfields is specialized. I have friends that are material physicists who might as well be speaking a different language when they tell me about their job. However, there's a lot of crossover here. We all use the same tools and we all adhere to the

same basic tenets. We all use calculus, the scientific method, and we follow a set of physical laws. So we can think of physics as this big umbrella with a bunch of related specializations underneath it. Outlay hits the scene in an era where bibliography just meant the study of books. That's already a specialized field. Alt-Lei is trying to expand bibliography to be more like an umbrella. That's how I see his shift towards documentation. He's making this broad, generalized field that houses specialized subfields. Each of those subfields may be very different, but they all share the same scientific tools. These subfields are, in reality, the offices of documentation. Each has a special area of study,

and they all use the UDC as a common means of organizing and handling data. There's just a lot of advantages to this approach. It allows experts to manage what they know best, which helps improve overall quality. It also pushes bibliography more in line with other sciences. It takes it from a library science to a full-on field of study. That may just sound like an honor type of thing, but I think it's a lot more important than just lip service. Alt-Lei is holding bibliography up to the same standards of physics or chemistry, which makes it look a lot more professional. That appearance, real or otherwise, openups more opportunities for funding and acceptance. A librarian might not be able to easily publish in a scientific journal. One of those new-style bibliographers, though, might have an easier time.

Let me bring us back to the topic at hand. How does this have to do with the nebulous founding date of the Mundanium? The IIB comes into being in 1895. The first offices of documentation form in 1908. These offices each held and managed their own part of the universal repertory. Call it a galactic repertory, since, you know, it holds part of the total universe. Anyway, these repertories indexed quite a bit of information. Here I'm working off W. Boyd Rayward's paper, Visions of Xanadu. I'm working off W. Boyd Rayward's paper, Visions of Xanadu. I'm not fully at the primary source level because, you know, the whole French thing and the repetition thing. In part of this paper, Rayward discusses the International Office of Hunting as an example of one of these early

repertories. This office contained books about hunting, that's to be expected It also contained charts, engravings, photos, articles, newspaper clippings, and any other documentation that could pertain to hunting It was all organized by UDC cards, which meant it could be queried Now, this wasn't the type of database that a random person off the street would walk in and operate. The UDC, although a very cool system, took some training. There were employees at the office of hunting that would step in and help. These were folk that worked for the office and knew the UDC system backwards and forwards.

An end user would simply ask a question of one of these clerical workers. That worker would then make the conversion from the question to a UDC number. They'd go hit the shelves and come back with an answer. So to use the repertory directly, you had to be literate in the UDC. But with the help of a savvy operator, you could get around that requirement. but with the help of a savvy operator, you could get around that requirement. Personally, I find something fun here. For this weird train of thought to make any sense, we need to look at other visions of hypertext. Rayward's paper is titled after Xanadu, so let's use that as an example.

Xanadu is this theoretical hypertext system that was designed by Ted Nelson, the very person who actually coined the term hypertext. A huge part of hypertext has always been the interface, that is, how a person actually uses the thing. Nelson saw Xanadu as a system that anyone could use. The point was to make something powerful and user-friendly. As such, the underlying technology was always left a little vague. It would use computers and networks and hard drives, but that was all secondary to some type of intuitive user interface. Ideally, this would be an interface that was simple enough that would require no training, no need for hypertext literacy to reap its benefits.

The offices of documentation, viewed from the outside, offer a very similar benefit. The interface here is just talking to someone that works at the office. This clerk is the one who knows how to actually work with the repertory. who knows how to actually work with the repertory. The end user is able to get data that they normally wouldn't have access to without needing to know exactly how everything works. They don't know how they're getting the data, just that they are. There are also more technical similarities here. Hypertext systems are all about decomposing larger swaths of data into smaller chunks than adding context into those chunks. Today, that context is usually just links, while older hypertext systems offered more features. The UDC, as used by the offices of documentation, works in exactly this way. Each card in a catalog

is simply a single idea. Some cards contain facts, while some describe where an object or book can be found. Everything is organized for retrieval. The references to outside objects are, clearly, a type of link. That's some really big context right there. The organization scheme, the UDC itself, provides another type of context. It's still a little rough, but this idea is pretty close to hypertext. You might notice that we're still mired in the past a bit here. The offices of documentation start forming around 1908.

That's before the 1910 date that gets thrown around online. That's before the 1910 date that gets thrown around online. I think that, by way of a ramble, this should be illustrating why a starting date for the Mundanium is hard to decide on. It doesn't really make a lot of sense as a question. The offices of documentation are already pretty close to pen and paper hypertext. But really, they're just implementing ideas that Atle wrote down in the 1890s. The big M name isn't in circulation, not yet, but we have a lot of the core features already coming together. So then, why does 1910 get thrown around? Let's continue the repetitive

ramble towards the Medanium. The next stop is the City of the Future. This is getting us to another one of those weird twists in the story, and it's one that I didn't expect to see again. That said, I really should have seen this coming. We already had some signs to scry from. You see, when I started doing my sporadic hypertext series, I called it hypertext and utopianism. I kind of dropped the tagline once I started hitting systems in the 70s and 80s, figuring that the connection wasn't really as strong as I initially thought. But, well, that's not really the case. Outlet wasn't just a peace activist and an ardent pacifist. He was also a utopian. He believed that a perfect society could be built thanks to the wonders of science and

strategic planning. So, what's the connection? Why does his political philosophy matter here? Early hypertext has this somewhat strange utopian slant to it. Vannevar Bush, Ted Nelson, and Doug Engelbart all write about hypertext as a tool to aid in the betterment of human society. Hypertext will solve the information problem, it will augment human intellect, and help create a brighter, more connected future. This is another one of those odd ideological planks in the overall platform. And it's a plank that also shows up in Alt-Leh's work. This is venturing into a bit of a source gap.

At least, a translation gap. Sometime around 1910, Alt-Leh starts to write about and discuss this thing called the World City. The core idea was, simply put, to construct a grand city that would serve as the center of a worldwide federation. Altley and La Fontaine were both big peace activists. They figured that connecting the world would be a surefire way to prevent future wars. The world city would serve as a location for a number of international organizations. From what I've read in secondary sources and some translations, it sounds less like a seat of government and more like a grand meeting place. Office spaces for organizations that

fell outside the purvey of any one government. There were a number of proposed sites over the years, but the forerunner was always Brussels. And to be clear, this wasn't just moving into downtown, this was building a whole new city, purpose-made for international collaboration. Once again, this is a spot where Rayward comes in clutch. In the article From the Index Card to the World City, Rayward writes an ominous line about this choice of locale. He had argued that Brussels might be considered to be such a place both because of the location there of increasing number of international associations and its central geographic location as the quote, crossroads of Europe, end quote. Foreshadowing aside, the world city smacks

of utopianism. We can think of it as a type of proto-United Nations. At its heart would be a massive collection of knowledge that Otlet called the Palais Mondial, that roughly translates to World Palace. This is often pointed to as the direct predecessor of the Mondanium. The Palais was actually established in 1910, so we even get the first date here. However, as we've discussed, the offices of documentation were already acting like smaller mondaniums, so the date is a little loose. Initially, the Palais Mondial was going to be a place to centralize the various repertories that had been developed

over the last decade or so. I don't think this was so much about centralization of management, more physical centralization. All this data needed a permanent home, and it would, of course, be in the world city. However, the actual world city didn't really materialize, in Otlea's lifetime at least. He had to settle for just the Palais Mondial. The repertory of repertories was first housed in the Palais du Cinquantenaire in Brussels, and I am so sorry that I don't speak French. In this new home, it started to expand. By 1913, the Palais Mondial's catalogue broke 11 million unique cards. That's at least 11 million pieces of information spread around actual cards, texts, images, and objects. Outlay and his colleagues were well on the way to creating a universal book of human knowledge. But then a setback occurred.

Belgium really was the crossroads of Europe. Outley said as much, as did one marshal Alfred von Schlieffen. This is where our story runs headlong into normal history, and in this case, that's the First World War. On June 28th, 1914, Archduke Franz Ferdinand is assassinated in Sarajevo. The assassination was a bumbling affair with many false starts and failures. The whole affair comes to a head when the Archduke's car stalls in front of a café. An assassin who failed earlier in the day was sitting at that very café. The result of multiple failures and this chance encounter is a war that would consume Europe. The assassination is really the first domino in what's a horrifying chain reaction.

Europe's patchwork of countries were tied together with innumerable public and secret treaties and agreements. The death of Franz Ferdinand caused a chain reaction as all of these agreements go into effect. By the time troops are moving, the war was inevitable. If you want the whole story, I'd recommend Hardcore History's series on the First World War. As always, Dan Conlon doesn't like no one else can. That said, I want to take one chapter out of the story. That's the Schlieffen Plan. This was the German military's plan to invade France. The general idea of the plan was to catch France off guard with a sweeping assault from the north.

A force would come through the northern part of France before turning down towards Paris in this broad hammer motion. This was backed up by smaller columns crossing the border further south. Now, France and Germany share a border, but they have other neighbors. The Netherlands and Belgium are sandwiched between the two larger countries. For the Netherlands, this is a small strip that's in between Germany and France. Belgium, however, is fully surrounded by the larger countries. This is one of the reasons it was viewed as a crossroads. From Belgium, you can go to France, Germany, or the Netherlands. Part of the Schlieffen plan was to make use of this crossroads. The French-German border was heavily fortified and defended, mainly due to earlier wars between

the two countries. But Belgium was a neutral country. As such, its borders were much softer. The plan was to march the first column of German troops through Belgium. That neutral country should just let the troops through, at least that was the plan on paper, but things wouldn't work out that way. On August 4th, 1914, German troops arrived at the Belgian border. This shouldn't come as a surprise, but Belgium refused to let Germany use them as a simple crossroads. As such, Belgium became the first country to be invaded in this war. They'd put up an admiral defense, but there wasn't much that could be done. The German army was simply too large and too well equipped. The smaller country was laid waste and occupied, thus becoming an actual crossroads. Now, disasters like this never

strike at a good time. Henri Lafontaine, Atlet's constant collaborator, had been awarded the Nobel Peace Prize in 1913. He earned the prize in part thanks to his work with Outlet. The two had been champions of international collaboration. The whole point of the World City was to bring nations together and ensure peace. La Fontaine was a tireless organizer, working for years to help establish and promote international organizations. The Palais Mondial was, of course, one of the cogs in this larger international machine. This peace prize came with just over $200,000 in prize money, that's adjusted for inflation. That was put back into the Palais Mondial project, which, at the time, was kind of strapped for cash. It seems like there was never enough money to keep the museum

operating at the pace it needed to. Paul Otlet was actually abroad when the invasion of Belgium began. This made things all the more complicated. He was touring the United States looking for more funding. He returned to a Europe that was in the first throes of a devastating war, where he would spend the next four years drifting from country to country. The Palais Mondial was kind of put on hold. At least, there weren't any major developments in this period. There was the beginning of more funding, but in an occupied Belgium, it just wasn't safe to keep going. World War I would come to a grinding halt in 1819.

La Fontaine himself was present for the negotiations of the Treaty of Versailles. The next year, 1919, is when we reach an actual development in our main story. The Palais Mondial would reopen, and it would now be known as the Mondanium. So, if we want to be exacting, then the Mondanium is established just after the First World War, in 1919. What was this ultimate form of the Mondanium like? It may not come as much of a shock, but we've already seen it. The core of this new library was really just an implementation of what Outlay wrote about all the way back in 1892.

It's a refinement of earlier attempts. That is, as far as I'm concerned, the real reason we can't date the founding of the Mundanium. Really, this is why we can't date the origins of hypertext. Outlay identified the information problem in the 1890s and described a workable solution that was to collect data and place it all in some type of context. The actual solution would go by a few names, the Mundanium just being the best remembered.

What I'm getting at here is that hypertext isn't a unique creation. From my research, it seems to be a common solution to the problem of handling large amounts of information. But let me back up a little bit. This series is about the Mundanium, so let's talk about the Mundanium as it stood in this interwar period. In the Mundanium, there were two types of cards, bibliographic and data. Bibliographic cards are what we've been discussing up to this point. These are cards that contain information about some document and where to find it.

When organized by the UDC, this gives us a great way to search physical collections of documents. That already has the sheen of hypertext, but we can go a step further. I've only referred to data cards so far. These are cards that, instead of referencing some document, actually contain a fact or some other piece of information. Some may have had references, but those would have been more like citations. Data cards represent Outlet's ultimate triumph over books. He believed that books were an unnatural construct. A text might contain hundreds of facts, but they were organized in an arbitrary way chosen by the author. There wasn't any standardization to it. It's all actually pretty illogical. Add in the simple

sequential nature of books, and that's kind of how you get the information problem. Data cards were a way to get around this whole book thing. They were simply facts extracted from the morass of a bound text. Then each fact was given a UDC number that reflected that fact alone. This UDC didn't have anything to do with where the data came from. It was coded to the single fact, the single piece of data written on the physical card. This is really taking the Dewey Decimal System to its logical conclusion. That system had been developed to make it easier to search libraries for answers. It gave a standardized way to look for books that covered certain categories. There were limits to that system. You only got a thousand big categories, which, honestly, were pretty poorly allocated. The UDC gave a more flexible framework for doing this simple task.

You got more complex categories that could be made up on the fly using a standardized syntax. Both of these systems address the information problem. At least, they help. Both the DDC and UDC make it quicker and easier to find a book. In the case of the UDC, it makes it easier to find anything. However, there's still the issue of the book itself. Let's say you're trying to find the artist that painted a specific piece of religious art that depicts a lake in Cornwall? That's a really specific question. You can pretty easily construct a UDC number that will lead you to religious art in southwestern England. You can even add in the code for lakes. That will probably

get you one book. I can't imagine there'd be a whole lot on that specific discipline. But then you're left with some giant tome covering religious art history. You still need to look through the book to find your artist. In some cases, books have nice indices and tables of contents, but some don't. It's a luck of the draw kind of thing. You still have to blindly go through the unknown to find a single fact. By ripping up books and filing away each fact separately, well, you circumvent the need to ever leave the UDC. You can do all your research from a nice standardized system. With a little skill, you can use the UDC to carry out complex searches that lead to an exact card. Now, there's some weird results of this setup that

I haven't seen discussed in depth, but I think bear mentioning. First is that the UDC number, while not quite a unique identifier, can be treated as almost unique. You can create a code that's so specific it can only refer to one possible card. There's just no enforcement mechanism for the uniqueness. In other words, it makes for kind of a bad unique identifier, but stick with me here. Let's just assume that each UDC code in your collection is specific enough as to be basically unique. Now, where do we often see unique identifiers? Why, the internet, of course. Those long strings of words, dots, and backslashes that you use to navigate to websites are called URIs, Universal Resource Identifiers. Each URI is a totally unique address to some resource that's accessible on the net.

URIs also follow a syntax. It's just that it's not very standardized. Sure, you have something dot something followed by path slash path slash path, plus some parameters if you're feeling spicy. The structure follows a standard, but what you put into that structure can be anything. The actual path is just a reference to some path on a hard drive somewhere, or some weird back-end logic in PHP. In other words, while conveying some meaning, a URI isn't a very reliable source of information. It's meaningful, but only a little bit. A UDC, on the other hand, is totally standardized. This means that, assuming the earlier assumption, you can have an address that's not only unique, but also holds standardized information about what it addresses.

Another implication of the Mundanium's filing system is that similar facts are grouped together. One of Atle's main complaints about books was the whole random organization thing. You're left to the whims of the author, which, in my case as an author, would be awful. But with the Medanium, ideas are organized by UDC numbers. All facts and ideas about English devotional art in the 18th century, well, they're stored together, regardless of their source. For one, this gets us away from the whole book thing. This is kind of an aside, but I think it's funny that Oatley and Ted Nelson both have this hatred of books. Anyway, by grouping similar ideas, you can do something really cool.

You can start to aggregate ideas. If you aren't used to that word, well, let me explain that a little. An aggregate is basically a combination of smaller aspects that create a larger whole. In data sciences, basically what I do for work, you generate aggregates from raw data. Those aggregates are usually more useful than the data alone. For my work, that often means taking something like raw download numbers for a file and turning it into, say, a report that shows trends over time. What does this mean for the mundanium? Power. Power is what this means. Let's stick with the perennial example in this series. You could use the Mundanium to figure out how English devotional

art has changed over time. This would be a pretty simple survey. You work up the UDC code for the history of art in England as it relates to religion. You find the section in the shelves for those cards. They will already be organized by smaller divisions. The one we're looking for is time, so you add the time code to your UDC number and you're done. You have your source of data spread out over a number of cards. That data may have come from dozens of separate books. Instead of reading all of those texts, you can now just consult the cards. You get your data more directly. This also allows for meta-analysis, that is, analysis of the data itself. The UDC number can contain information on geographic locations, so you could further group your temporal data by region. Maybe you discovered that devotional

art was more important in the south of England in the 17th century, but centers of production had shifted more north over time. That kind of analysis would take a long time and a lot of work if you had to surf volumes of books. With the Mundanium's data cards, well, this could probably be handled in an afternoon. It's a single coffee pot kind of problem. This type of analysis is very useful. This is how I handle most of my initial research for the podcast. I'll use the Internet Archive or some similar service to search through ebooks. That website lets you search by title, author, text contents, medium, and publication date. With some smart application, you can start to build up a

timeline of sources. I'll then often use this to identify which sources are relevant to an episode and which are actually useful to go into. Of course, not all my sources are stored in the archive, but the ones that are, I have a lot easier time working with. The point here is that the mundanium represents a sea change in how information could be handled. We saw similar things in 1945 with the widespread adoption, or at least somewhat widespread adoption, of digital computers. The mere fact that computers existed meant that new mathematical methods could be used. If you want a wild story, then go read up on the development of the Monte Carlo method. This is a way to solve equations that is only possible if you have an automatic digital

computer like ENIAC. That means that computers didn't just make math easier, they led to whole new types of mathematics. They expanded the horizon of research. The mundanium could do similar things for information. At least, the theory is there. What was the mundanium actually like in practice, though? How much can we actually say about the implementation of these great ideas? Once again, this turns into something of a difficult question. To figure out what the Mundanium was like in practice, we need something like a Yelp review. Of course, Yelp didn't exist back in the interwar period. I know, strange. This is confounded by,

of course, the language barrier and the fact that the Mundanium was more than just an archive of information. It also housed a part-time school, conference spaces, and it facilitated international projects. Add in the fact that the Mundanium's goals and scope shifted over time, and this leads to a bit of a headache on my part. Surfing newspapers, I can start to put some things together. And let me be clear here, I'm intentionally trying to find an outsider's perspective, or at least how Alt-Leh and his colleagues were describing the Mundanium to the outside world. I want to give us an idea of how you might feel going to the Mundanium as a layperson. By 1922, the Mundanium housed over 12 million cards. It's not clear if these were data or bibliographic cards or a combination. This was in addition to physical museums that were

set up under the same roof. Operationally, the Mundanium was a truly international affair. Not only was the space used to hold international conferences, but collections of documents and data were brought in from collaborators around the world. So you could expect to find documents in many different languages. Once again, we're dealing with a truly international organization, the remnants of Otley and LaFontaine's attempts at a world city. remnants of Otley and LaFontaine's attempts at a world city. As far as direct experience, here is one thing of note I was able to find. This comes from an article in a 1927 issue of The New Era. The author is... someone? It looks like a by-the-editor type of article, and it's unclear who the editor is. Anyway, the author gives a description of making

a trip to the Palais Mondial to give a lecture and meeting one Dr. DeCrolly. As yet another aside, in some texts and some articles, the Mondanium is referred to as the Palais Mondial even after the name change. Now, back to the source at hand. De Croli was an educator and one of Haute-Lez colleagues. It seems that de Croli was using the Mondanium as a teaching tool. A bit of a ramble to set this up, but check out this quote. A school without textbooks. The children use the library and museums of the town and make their own individual books. Boys and girls are retained until the age of 16, when they can pass, without examination, into the higher secondary schools to complete their studies. The author continues, I also had the pleasure of visiting that remarkable history museum organized by Mr. Paul O'Lea,

which presents a synthetic view of the whole scheme of evolution. History teachers would do well to plan a definite project for their senior pupils of a week's visit to the Palais Mondial. They would gain more real understanding of history in that week than in years of study from books. Synthetic here doesn't mean fake, but instead a synthesis, as in the combination and analysis of data used to create a larger whole. What we're seeing here, indirectly, is that the Mundanium was at least somewhat user-friendly. Outlé wasn't just slamming down a pile of cards for experts to sort through. The facilities at the Mundanium could be used by everyone, from professional researchers to children. And, once again, it wasn't just about data. The evolution display is interesting here

because, at least ostensibly, that same data on display would also have been held in cards filed away by their UDC numbers. It's all very physical, but we're looking at something really cool. That's alternative representation of the same data. This is something that's common and automatic today. HTML, for instance, is displayed in multiple forms. Most indie users view HTML as nice graphical pages on your screen. It can also be viewed as source code, the raw data that's later interpreted as nice graphics. This dual representation also appears in other hypertext systems like HES and

NLS. Now, the mundanium isn't doing the exact same thing here. We don't have this automatic way of switching representations. That said, I think we can see a glimmer of the future. This leads me to a final question. How does the history of the Mondanium link up with the larger hypertext story. How is Outlay connected to later works? From what I can tell, he isn't. I haven't found so much as a single reference to Outlay in the works of Vannevar Bush, Ted Nelson, or Doug Engelbart. That's kind of a small cross-section, but it's essentially the core canon of later hypertext. The only time any of those names appear together is in later works, works done outside of the development of hypertext itself. This may sound strange at first, at least to our modern ears.

How can there be isolated origins of hypertext? How can there not be some sort of link? To me, this makes the story a lot more powerful. Like I said at the top, hypertext becomes more interesting if you look at its history as a web. We can look at it as a story of lineages, but that prevents us from examining lines that don't lead directly to the modern internet. One of the things that makes Alt-Le so interesting is that his work is so disconnected from that main lineage. He grappled with very similar problems and came to shockingly similar solutions decades before Bush, Nelson, or Engelbart.

Alright, that brings us to the end, for now. Otle's body of work is really one of those research sinks. In these last two episodes, I focused primarily on the technical details of what was actually implemented. Even then, I've only really made it up to the 1930s. There's a lot more to say about Oatley, the Mundanium, and his work on hypertext. So, look forward to a return in the future, but honestly, I kinda need a break to process some of this. I'd also like to produce episodes not relating to Paul Oatley sometime in

the next year, so we're gonna put this on the back burner for at least a little bit. As always, production is a bit of a balancing act. The biggest mission here is Atle's more speculative works. Let's just say that Atle produced some sketches that are shockingly similar to the mimics, and he had articles to back that up. That's something that needs its own room to breathe, so I decided to just skirt around it in this series. Look forward to that in the future, sometime. Until then, I think I'm going to slip into something a little more digital. Thanks for listening to Advent of Computing. I'll be back in two weeks' time with another piece of computing's past. And hey, if you like the show, there are a few ways you can support it. If you know anyone else who'd be interested in the history of computing, then please take a minute to share the podcast

with them. You can also rate and review the show on Apple Podcasts and I think on Spotify now, too. If you want to be a super fan, you can support the show directly through Advent of Computing merch or signing up as a patron on Patreon. Patrons get early access to episodes, polls for the direction of the show, and bonus episodes. You can find links to everything on my website, adventofcomputing.com. If you have any comments or suggestions for a future episode, then go ahead and shoot me a tweet. I'm at Advent of Comp on Twitter. And as always, have a great rest of your day.