Weinberger was also a co-author of the notorious boom-era best seller, The Cluetrain Manifesto. A fellow at Harvard Law School's Berkman Center for the Internet and Society, Weinberger is now doing a regular podcast for Wired News called The Berkman/Wired Miscellaneous Podcast.
The interview was originally conducted via Skype for NeoFiles.
To listen the full interview in MP3, click here.
RU SIRIUS: When I first saw the title of your book — Everything Is Miscellaneous — I immediately thought of my old friend Ted Nelson. He had a saying: "Everything is deeply intertwingled." Sure enough, as I got into the book, you beat me to it. You actually deal with this quote in the book. How does Nelson's idea relate to your idea?
DAVID WEINBERGER: Nelson's idea is that the world is intertwingled. That's just a great, made-up word that says that things don't come in neat categories. Sometimes we need to put things into very strict categories, and we manage to do that. If you're working at the Department of Motor Vehicles and somebody comes in with a boat trailer, you've got to decide: Does it or does it not belong in the category of licensed vehicles. We have to make these sorts of decisions. But that's not the normal case. The meaning of most things is linked, loose and ambiguous. The category systems that we've had in the past, the taxonomies – each with its experts — have not generally reflected that intertwingularity. But the web, with its link structure, and with its messy, ungoverned, permission-free link structure, perfectly represents the intertwingularity.
RU: In the world of atoms as opposed to the world of material stuff, it's easier to make all that intertwingling available. It seems almost like we're in a virtual "Six degrees of Kevin Bacon" world. Everything is six clicks away.
DW: Yes. The internet works that way. And there are so many different links and ways to get to things because the significance of our world works that way. That's why things on the web have accumulated so many messy, unpredictable links. Lots of people have seen lots of ways in which things are related, and we can express that on the web. We don't have to minimize it.
You know, in a library, a physical book has to go on only one shelf under one category. That's not a natural restriction; a single book is about many different things. But even when you try to make up for that restriction with the catalog card, which is a very reduced form of meta-data for the book, the size of the card is dictated by the inconvenience of atoms. The size of the card means that you can't put in very many of those references. But on the web, everybody can put in his or her own references. We can have hundreds or millions of references and links and connections of meaning linked to a single resource. There's no limit. So, in some ways, the web reflects better the complexity of the linked nature of the world.
RU: The massive hyperlinked web of correspondences and information that Nelson talked about with his Project Xanadu in the 1960s is happening, but it's sort of self-assembling. There's a sentence in your book that's unobtrusive — or you might say it's miscellaneously in the middle of a paragraph somewhere — but I picked it out because it seems to go right to heart of what you're saying. This is the quote: "A big part of miscellaneous information contains relationships beyond reckoning." I think it's the "contains relationships" part that's important – because although everything is miscellaneous, we're not just talking about noisy chaos.
DW: No, we're not. I'm admittedly using the word miscellaneous in a slightly extended sense. The value is not that it stays miscellaneous, and we can never find or make sense of anything. Quite the opposite. It's all there as potential. We can mine knowledge and information from it. But I don't think that's all that interesting. What's interesting is that we can also discover meaning and its significance — stuff that actually matters to us. So every time we sort through the stuff, we cut through it and see the connections that are interesting to us. And depending on what we're trying to do, we see the world in a new way. We can now do this quite fluidly, and we'll get even better at it over.
RU: The order is found by the end user. A friend of mine has a business and his slogan is "living à la carte". That seems to be kind of what we're doing with information, and so many other things.
DW: Yes, but when you order à la carte, everybody orders individually, based upon their tastes. I wouldn't want to leave it there! The most exciting and important advances in how we're making sense of this miscellaneous soup is that we're doing it socially. We're doing it through social networks; through recommendations from our friends, from sites that do that more formally; and from what shows up in our inbox. So this is not the Daily Me constituting the world based on our own individual interests. It's the "Daily Bunch-of-Us." It's loosely defined groups of people making this happen.
RU: So this is not the wisdom of the individual or the wisdom of the crowds, but the wisdom of small social networks?
DW: Yeah. It's the wisdom of the group. The crowd actually turns out to be quite lumpy. We know some people better... I know that this person over here is really useful and knowledgeable about FCC rulings, but I wouldn't ask about cars! But this other person loves talking about cars.
RU: In Ethan Zuckerman's blog post about your book, he asks: "If knowledge is a pile of leaves instead of a tree, how does the shape of knowledge change?" Do you have an answer for that?
DW: Yeah! First, there's the tree-like structure of knowledge, in which categories are carefully arranged. So there's a root and then there are branches, and every thing has to be neatly on one branch and only one branch. Each thing has a special spot and only one spot. And that shape is very useful for some types of thinking. It's certainly the shape that you use to divide up your laundry. You divide it by person, and then by body part and so forth. So you are constructing a tree. It represents how we sort and order physical objects and it's very useful.
But when we make things miscellaneous, we get to shape it the way that we want. And frequently, the shape is going to be a tree. And sometimes the shape is going to be a cluster in which there is overlap. It's every type of human relationship. It's every possible shape and so there isn't a shape. It's this potential we have before us that we can shape in ways that make sense to us at the time. And the "us" is — in fact — a social group.
RU: You refer to this miscellany as a "third order." Can you explain a little bit about this idea?
DW: In the First Order, you organize the things themselves. An example would be the physical books on the shelves.
In the Second Order, we do something that we've gotten very good at over the past couple thousand years — we separate meta-data about the stuff in the First Order. So we're still dealing with physical objects. In terms of books, it's the card catalogue. We're separating the meta-data. We've reduced the amount of information we're dealing with to what fits on the 3 x 5 card. It's much less than all the information about the book. But with the Second Order, you now have a few different ways of sorting (or categorizing). For instance, you can sort by author, title, and subject.
In the Third Order, everything becomes miscellaneous — both the data and the meta-data; the content and the information about it. The principles that guided the organization of the first two orders no longer hold.
RU: So are you saying that the first order is basically pre-taxonomy? And the second order brings that into being. And then, the third order changes how taxonomy operates – or are we leaving taxonomy behind?
DW: We're not leaving taxonomy behind. Rather, we are embracing every possible way of organizing — every shape of organization that works. And sometimes taxonomies are exactly what we need. So we have taxonomies, we also have playlists. Playlists are not really taxonomies. They're just lists. (I guess you could say they're the edge case of a taxonomy.) Playlists are really useful for some things. They're really useful for music, for example, or for syllabi. But they're not a very good way of organizing a complex library, because the list gets too long. We will use every type of organization we need, including taxonomies, when they make sense.
RU: I guess if you label your iPod music lists — say, "anti-War songs" or something like that — then that becomes a sort of taxonomy. A little mini-taxonomy. In some ways, it seems like we're really obsessed with classification these days. You have things like the human genome project. There are various projects to catalogue biological life forms. And apparently Edward O. Wilson is now doing some kind of an encyclopedia of all life. Where do those types of projects fit into your schemata, if they do?
DW: They do, because we are developing knowledge out of a pre-existing taxonomy. We make links!
Let's just limit the discussion to tags — we are not doing that because we have an existing taxonomy, but we may be able to generate a taxonomy based upon the set of tags. In fact, the most important thing is that you can generate lots of taxonomies based on a single set of tags. So it is useful to have an order of species. And scientists have been arguing about the nature of species and how you cluster them since Darwin. The argument over what constitutes a species continues among biologists still.
Sometimes you'll define a species, and thus a set of categories; and then the relationship of those categories, because you're interested in the history of their actual descent. But sometimes you'll be interested in how populations — within isolated areas separated from each other — work. At that point, their common ancestry may not be as important to you in your categorization scheme. So sometimes it will be more useful to cluster things in ways that look at their functionality as opposed to their DNA. So there are all these potential ways of organizing species. The great thing is that now we can have them all. It's all miscellaneous. If we're doing epidemiology work, one sense of species may be more important to us than another. We can have it all!
RU: Can you conjecture about the personality changes that might happen with people whose ability to organize the chaos of information is being democratized? Is there any danger of a sort of virtual narcissism?
DW: I'm a little less concerned about that than some other are because I think this activity is as social as language is. In fact, it's very closely related to language. Actually, the old idea that you could sit down and organize the universe by creating a taxonomy seems to me far more narcissistic than the bottom-up stuff that we're doing now, which is more democratizing.
RU: Do you think people are empowered by it? Do you think it might be a sort of evolutionary step for human beings?
DW: I do think it's an evolutionary thing and I do think that people are being empowered by it. But I sort of think those two things separately.
Taxonomies are power. With a centralized top-down taxonomy, one problem is that somebody gets to say what you are. And lots of people will inevitably disagree with the categorization. A really bad example of this is what happened to a popular musician living under apartheid in South Africa, By the time he was fifty, he had become a different race five times because the law had changed. And once, he had to leave his wife and family because of it.
So taxonomy is power. It's not always that gross. But let's say you're trying to decide where Scientology or Jews For Jesus or Baha'i goes in the category of religions. Are they at the same level as the big ones? Power resides in that decision. Now that we can create local clusters of meaning, local taxonomies, categorizations — a lot of that power dissipates. That's a good thing.
RU: Not only can we democratize the taxonomies that are created, we can locate new taxonomies, in a sense. We can have lots of them, as per Wikipedia.
DW: Yes. And furthermore, what works about Wikipedia is the fact that just about every other word is linked. That's more important than the categories. The categories of Wikipedia are really more like tags – and that system totally sucks! It's broken! It's barely usable. (Maybe they'll fix it.) But the fact that every article is penetrated by link after link after link going everywhere says that this messy web of meaning is more important than coming up with a nice set of categories.
RU: How would you compare what you're saying with Everything Is Miscellaneous to the two big tech business memes of the times — Web 2.0 and The Long Tail? Do you feel you're extending that? Are you taking it a little further out? How would you relate to all that?
DW: Well, from a point of view of authors' narcissism, I started working on this before either of those things came along. So I've watched them, and I do think there are relationships among all three of them.
Clearly, the long tail is about how content and ideas and stuff is spread out rather than centralized. If you're doing long tail economics or long tail business, you've got to wonder how you are ever going to provide a single categorization scheme for your products that is going to work for the entire long tail. Because, in terms of big topics, the long tail isn't actually interested in anything. The consumers on the long tail are interested in their own quirky individual things. That's the power of the long tail. So I think you want to move towards a miscellaneous way of thinking through how your customers are going to find what you want. So it's a good match.
Web 2.0 is, of course, a notoriously free term. In some ways, it's a set of examples, things that you point too. You say, "Blogs are happening, and Wikis are happening, and APIs are opening up, and there are greater mashups of information." Many of those things enable the representation and use of miscellaneous data so it all pretty much fits the miscellaneous model.
SF Writer Rudy Rucker: Everything Is Computation
How The iPod Changes Culture
Jimmy Wales Will Destroy Google
Counterculture and the Tech Revolution