A New Medievalism

image link-topic-sf0.jpg

1. The Small Picture

I'll start with what is technically just a gripe about a computer interface - but which is actually something symptomatic of a much larger problem.

Scenario: I'm in my cubicle in a building in some vast corporate suburb. I want to scribble a drawing in a notebook. So I pick up my notebook from my desk where, lazily, I'd left it yesterday. I draw something in it, and take some notes. Then, overcome by an unexpected fit of organization and neatness, I go to put my notebook back on the shelf where it belongs and ... an alien transporter-cube materializes around me and, whoosh, suddenly I find myself transported back home to my living room. I look around in stunned amazement, my notebook still in my hand. What now? WTF?

There is a small shelf of books in my living room at home; it contains some well-read favorites, but it isn't where my notebook belongs. The transporter-cube is still here, so I look at it. It has an array of big buttons on it, each labelled with a place that I've been recently (the post office, an ATM across town, the dentist's office). But the shelf at my office isn't there.

Fortunately, my wife left her car home today. So I walk out to the driveway, get in her car, and drive back to the parking lot at work. I get out of the car, walk through the lot and the lobby, go up the elevator, down the corridor, into the Grand Hall of Cubicles, navigate my way to my own cubicle, sit down at my desk, and put my notebook back on the shelf.

That wasn't any fun at all, so when I need to scribble my next drawing I leave my notebook where it is and instead pull a legal pad out of a drawer. But when I go to put it back in the drawer, whoosh ... I'm back in my living room. This time I'm going to have to take the bus back to work.

Obviously this is a bizarre scenario, and unless powerful aliens who watched too much Monty Python on their way to Earth take control it will never happen.

Except that it has. This is exactly what happened to the FileChooser in GTK in 2011.

The concept of the "current working directory" has been an integral part of computer science since at least 1969 (the introduction of UNIX). It's been integral because it makes sense in a very deep way. You are someplace. You take something and change it. Your first choice for putting the results back is the very place it came from. If this doesn't seem simple, then the aliens have taken over your brain.

The aliens have, regrettably, taken over the brains of the GTK (and Gnome, and GIMP) developers. They're basically smart people. I've read the postings of the FileChooser developer on this very subject. He's a smart, reasonable person. But in 2011, as a result of deep study and consultation with his peers, he removed the concept of the "current working directory" from the FileChooser and replaced it with a teleporter which slams you back home every time you try to save a document/spreadsheet/drawing/whatever. At best it lets you go to places you've "recently been" (which of course are never where you actually want to go, because you've just been to them already).

The amazing thing is that even when informed in careful, well-reasoned, well-documented (and even polite) terms, neither he nor his fellow GTK developers could even see that this was a problem.

In itself, this is just a small daily annoyance - and the world is full of small daily annoyances; no big deal. But think back to all of the Cold War paranoia science fiction novels and films. When you wake up one morning to find that everyone around you is doing obviously nonsensical things that they never did before, and can do nothing but smile vacantly at you, then you know that things are going to get really scary really fast.

The problem with the GTK FileChooser is just an example of an alien pod-person smiling vacantly at you. Beneath it, something is going very deeply wrong with our civilization.

2. The Big Picture

The Gnome/GTK/GIMP developers are intelligent, thoughtful people who are trying very hard to do the right thing. The problem is that they have seen something so captivating, so sexy, so dangerous, that it has taken over their way of thinking and compelled them to turn a simple user interface into a Monty Python sketch from hell. They've seen Google.

More precisely, Google has seen the idea/character of "the librarian" from Neal Stephenson's 1992 novel "Snow Crash." This, in turn, is a much older idea: the oracle, or universal answerer. A friend once told me that the world of computing was changing fundamentally because we now had a "search paradigm" in place of older ways of thinking. This is not true. Nobody searches for things on Google. Really - a search is an adventure. Indiana Jones searches for things. Michael Faraday and Richard Feynman searched for things. Searching is hard, and in the process you learn more than you bargained for. Searching is good.

Google doesn't let you search. Google gives you answers. And its very, very good at it. I'm not slamming it; I like Google, and use it constantly. Google, per se, isn't the problem. The enemy is not Google - it is us. We've seen how very good an oracle can be, and have taken a lazy route and substituted it for the structure of our society.

So now there are active projects within software development communities such as GTK to develop frameworks which will be like Google and which will always know what you want before you want it (by comprehensively analyzing everything on your computer and everything you do - much as Google and the NSA do). And of course once you have something which always knows what you want to do, there's no point at all in offering anything else. This is the nightmare of both Stalinism and fascism, which is why it was a staple of science fiction in the Cold War.

But those were just paranoid novels and movies with bad special effects. They weren't quite convincing then, and they look more than a bit dated today. You could always tell when the Pod People came on-stage. Google is different. It's slick, and very well done. It works. It has demonstrated conclusively that it is possible to build a universal answer machine - an oracle - which is so useful you'd never wish to be without it.

The question is this: If you have such a society, with a universal oracle, what kind of society is it? If you look at the history of civilization for examples, what kind of society is it that we are building today? The answer is simple and unsettling: a medieval society.

I commend to you Umberto Eco's novel The Name of the Rose because it gives a compelling, accessible account of a medieval library. Yes, there were medieval libraries - and large ones by the standards of the time. Though learning was anything but general in the middle ages, where it existed, the standards were high.

So let's say you found a time machine and went back to a medieval library. No computers, so you go to look up your favorite book in the catalog (say, the second volume of Aristotle's Poetics). Right away, you'd make a startling discovery: that the library catalog hadn't been invented yet. Neither had the concept of the index. Instead, the books in the library might be listed one after the other, in the order in which they were acquired. The structure of this list would, at least, be familiar to you: its a that of a blog.

The second thing you'd find, just after you got thrown out of the library, was that access to the blog was carefully controlled. You can't just walk into a medieval monastic library and start poking around. Books are very dangerous. Even ordinary books can give an unprepared mind the wrong ideas, and certain very special books have killed millions. (I'm not exaggerating here; really, they have.) Access to books and knowledge in a medieval library is very carefully controlled by the librarian. Read The Name of the Rose if it isn't clear to you why this is a problem.

It is no accident that the model for Google might well have been the Librarian in Stephenson's "Snow Crash" (a novel which shaped our ideas of networked life as much as Gibson's Neuromancer, and with greater precision). They probably weren't thinking of Umberto Eco. But the difference between Google as omniscient "librarian" and Jorge of Burgos, the librarian in Eco's The Name of the Rose is that Google has a much better public relations department. Everything which has to do with the changes in the way we access the world - changes which Google has either brought about directly or inspired in other vendors and developers - is at present strongly promoted. It's cool. It's cutting edge. It' the obvious way of the future. By way of contrast, everything which has to do with a medieval library is, well, medieval. It's old and dumb (and, really, it was). It is obviously wrong. The trouble is, they're the same.

The medieval library was a system of information technology. It worked, but it had limits. It slowed research. In many areas it stopped research deliberately - it is the perfect technology for political control. It limited scholarship to particular modes (the citation of approved authorities). It slowed other technologies dramatically. There was some technological development in the middle ages (at least from the 12th century on), but major developments took decades and centuries.

The Renaissance changed this. There were any number of reasons, each of which would probably be sufficient. The development of movable type and the printing press was one. A rich and literate person in medieval Europe might have a dozen or more books. A really big medieval library might have one or two thousand books. From the start of modern printing in about 1450 to the end of that century in 1500 - a brief period of 50 years when printing was just getting underway - somewhere between eight and twenty million books were printed. This one factoid sticks in my mind as one of the most remarkable occurances in all of history. No librarian could regulate access to them.

What came out of this, over the next few centuries, was the technology of modern scholarship. People invented indexes and cataloging systems. Books started to have tables of contents (eventually, these even got moved to the front of the book). Forms of citation were developed. No longer was it sufficient simply to claim that Aristotle said something and quote a snippet; you had to point back to the source of that snippet in some way so that anybody else could track it down. This is the equivalent in the humanities to the reproducibility of experiments in the sciences. That you can do something means nothing - you may be lying. What matters is that you can show someone else how to do the same thing. On the basis of these and related information technologies from the 15th through the 20th centuries all modern civilization rests. They didn't exist in the middle ages. They were revolutionary in their time. They made our world. Now they're disappearing. (Or to use a phrase from certain political upheavals, they've been disappeared. The generation growing up today won't even know that they once existed.)

There are three aspects to the problems you get when you wipe out the information technologies developed from the Renaissance through the late 20th century and replace them with the information technologies of the medieval library: one technical, one scholarly, and one political.

3. Consequences

3.1. Practical Consequences

On the practical side, ceding our ability to organize information ourselves to an outside oracle means that we won't organize our own information. Already, this is making good information hard to find.

In the early 2000s, people stopped writing/building website. Real websites, with their information presented in a structured way, were too hard. Blogging sites were easy. So people started using blogging sites to create their websites.

The problem with this is that a blog has no real structure. Many of these blog-sites are written by very knowledgeable people who are embedding a lot of good information in them. But after a year or two of posts, you simply cannot find anything. Google only helps so much; clouds of keywords and tags are useless. The structure of good information is detailed and deep. Google and tag clouds are shallow. Indeed, a significant part of the information itself is its structured relationship to other information. Relying on Google and cloud tags denies you the ability to encode the larger-scale structure of your thoughts. Your blog-site can never actually say what you need to say. The only person who can meaningfully structure the information in your site is you, its author.

The pace of this is increasing, because even blogs are now too hard. They're old fashioned and boring. But how much deep, structured information can you really present in 140 characters? (Yet as I write this one of the best writers for a major financial newsmagazine just stopped his column and will now attempt to communicate his insights about the Asian economic world (which are good insights, let it be said) in tweets.)

There's also a problem of ownership. If you built a real website, then at least within the increasingly coercive law of intellectual property it was yours. When using a blogging platform to build a website turned over the control of its structure to the platform. When using a social media feed to try to structure your online presence, you turn over the ownership of your thoughts entirely. (If you believe that you actually own your own thoughts or their presentation in social media sites, think again. The whole point of such sites is that they present what you feed to them to others in one way, so that we are all individuals [1] together.)

This has practical, economic consequences. An example: my wife sells jewelry at various specialized shows. When you do this, there is a certain minimum amount of information that the show's organizers must convey (when does vendor registration start, etc.) Without this, the show will fail because nobody will know what you're doing. It is not hard to convey this on a very simple website. Yet for one show (which will remain nameless here), you can no longer find this information unless you "follow" its Fearless Leader on Facebook. This show was once one of the best in the country; now it's hurting for lack of vendors.

3.2. Consequences for Scholarship

Scholarship depends on citation, just as science depends on repeatable labwork. If nobody can repeat your scientific discovery in the lab, you haven't made it. If nobody can find the exact source you used for some particular point, you haven't made it.

Google's business is to deliver to you the "best" answer, not the same answer you got the last time. Indeed, if Google delivers the same answer to you twice in a row, then its services have not improved and it's doing something wrong. By design the Googleverse is flux. This is fantastic for searching, and Google does a great job of it. But it is not scholarhip.

Yet a generation is now coming into being which doesn't actually understand the difference between a scholarly citation and a search engine.

3.3. Political Consequences

If you're conditioned to believe an oracle rather than actually to understand the world around you, you have no political understanding.

The upshot of this is that, to take a particularly egregious example, the Fox network can actually call what it broadcasts "news" and a large part of the population will believe that it is. People who cannot structure the information in their computer systems, or their websites, or their scholarship cannot structure the information they discover about their political world. They'll just believe as they're told.

Of course, as we head toward a new medieval society with a tiny and increasingly hereditary aristocracy controlling all of the wealth, no doubt this is as it should be.

4. Coda: Fixing GTK

There is one slight reason for optimism, at least in the case of the technical example which prompted this essay. In 2012, this bug in the GTK file chooser was at least addressed . An option was added to restore the concept of the current working directory. This option was hidden, however, in a poorly documented and normally invisible configuration file two levels down.

To survive in our new medieval world, you have to learn secret incantations - just as you did 800 years ago. How different this is from that open-source movement called the Renaissance.

-----

[1] Yes, that's a Monty Python reference.