On the Nature of Ideas

I did an interview recently where the author, clearly having done some homework, called out an old quote of mine arguing that ideas aren't like widgets or screws, that they're not industrial objets.

I'd said that a long time ago, inspired by John Perry Barlow's Declaration of Independence of Cyberspace. Here's the money quote: "Your increasingly obsolete information industries would perpetuate themselves by proposing laws, in America and elsewhere, that claim to own speech itself throughout the world. These laws would declare ideas to be another industrial product, no more noble than pig iron. In our world, whatever the human mind may create can be reproduced and distributed infinitely at no cost. The global conveyance of thought no longer requires your factories to accomplish."

The world that John Perry was talking about has not come to pass, completely. The governments have certainly moved to impose more and greater controls. But as Lessig noted just a few years later in Code: and Other Laws of Cyberspace, the aspects of cyberspace that promised liberation, a nation of the Mind...those aspects were the output of human-controlled systems, and humans could and would change the rules if they didn't like the outcomes.

I was there for parts of these conversations. Gave JPB a ride around town, harassing him about the declaration and about Cassidy. I put together Lessig's book party for Code when it came out. But the thing about ideas stuck with me more than the rest.

I'd studied epistemology, the theory of knowledge. You get a lot of examples of attempting to codify ideas (and brains, the storage tanks of ideas) into the machinery of the time (See the masterful book, "Memory Practices in the Sciences" for more). But in the end, ideas resist complete capture.

They're ethereal. We've spent thousands of years trying to codify them, into Plato's forms, into machines, now into code. The dominant industrial paradigm tends to be the stuff we use to try and understand them and their human substrates - pumps and machines to explain the brain in the industrial age, circuits and pathways in the digital. This ethereal nature makes it hard to get the ideas into the powerful information systems of the day, which are based on bits and bytes. It's one of the reasons that the most powerful idea transmissions systems we have are humanist - text, sound, video. It's why something as lousy as powerpoint can take over, because it's a way for people to talk to people.

It's hard to make ideas into widgets or screws because of this. It's also hard because we all see the world differently, even those of us who agree. We use common words as proxies to help convey that this red ball is an apple and this green ball is also an apple. Making the word apple into an abstracted computation tool is hard, because you have to decide what it means, and convince others to use your meaning rather than their own. Cyc's been pushing on this for 25 years and we still don't have the Star Trek computer recognizing our voices.

But we're starting to have to try to make ideas at least representable as widgets. The problem is that the information space is overwhelming us as people. We can, using robots in the lab, sensor networks in the ocean, miniature microphones in public spaces, genotype smears on red light signals, generate data at such a level that we simply cannot use our own brains to proces the data into an information state that lets us extract, test, and generate ideas.

There's two things we can do, one easy and one hard. First, we can make the existing technologies for idea transmission (writing it down onto paper and publishing it) more democratic and network friendly. That starts with good formats: putting ideas into PDFs is a terrible idea. The format blocks the ability to take the text out, remix it, translate it, reformat it, text mine it, have it read to a blind person via text-to-speech, and on and on. It continues with open access (so we don't create a digital divide first, and so we enable the entrepreneurs of the world, wherever they are).

I'm at a conference in Hyderabad called Tech4Society that is packed to the gills with inventors and social entrepreneurs, who for the most part have no access to the scientific and technical literature. It's all in English - which many, but not all, speak here. It's very expensive - nuclear physics journals can cost more - per year - than a new car. And this is a tax on the entrepreneurs of the world.

Inventors have to invent. It's in their blood. And they have the capacity to rapidly combine information from multiple sources to assemble new projects. I heard today of systems that leverage sugar palms in Indonesia to power villages, of local decentralized power panels for wind and solar to give each house its own power, and more and more and more. But this is being done without the newest knowledge, knowledge that is on the web somewhere...but locked up by paywalls.

We as Americans send a lot of money. We'd be a damned sight better if we sent a lot of knowledge.

The Open Access movement is being driven mainly inside the developed world. US and EU librarians feel the pinch of the serials pricing crisis, and funders like the US National Institute of Health and the Wellcome Trust take policy directions that lead towards the availability of the biomedical research. And it's wonderful that the solutions to these problems all lift the developing world along the way. It seems that the scholarly literature will, in fits and starts, and in some disciplines faster or slower, find its proper place on the net, free of commercial restrictions, one of these days.

But it's not just ideas, it's what to do with the ideas. Richard Jefferson today made the lovely point that the patent literature is a giant database of recipes to make inventions. And that if you can find the inventions that were patented in the US, but not in India, you've got a lot of good stuff to work on in India. This is true. And deeply important.

But I got a little melancholy thinking of the stuff that comes before an inventor becomes a social entrepreneur, ready to apply for funding or speak in front of 200 people at a conference. Maybe they can't read the patents and understand the information. Maybe they just need to build some furniture for their house, or fix the stove. I had a sense-memory of long shelves of the books in Home Depot, the how-to guides, the recipes for doing simple stuff, unpatented stuff, but essential stuff, and I look at the amazing user-driven innovative spirit that rules the day in India, and I want to cry at the amount of knowledge that is deprived. Give these folks the books and get out of the way!

I wish we could come together as a culture and create an open source set of how-to books to parallel the scholarly literature. Those book are how I learned to rewire sockets, to fix plumbing. Where I learned what was dangerous and what was safe. They're a place where those ideas, laid out in the papers that are becoming free, became methods that I could use. Where the ideas became actionable for me. Imagine if those books were movable from my server where I wrote them, to a server in Africa who translated them into Kiswahili, or Chichewa. If they could be formatted to be read on the mobile phones ubiquitous across the world. If they could lead to one more hour of light per night through the creation of lightweight photovoltaics.

Has anyone out there done this yet? Anyone interested in doing it? Anyone immediately get a rash and freak out? All of those reactions are interesting to me.

The second part of why ideas are hard will have to wait for the next post. Suffice to say the word "semantic" will feature prominently.

I'll post more from day 2 tomorrow. Jetlag over and out.