Archive for September, 2010

[2b2k] Decisions, decisions

About two minutes ago I discovered that I was at the end of an EXTREMELY rough first draft of the chapter on decisions. If forced to lay odds, I’d say it’s about 12:1 that I will be doing a major rewrite of it, since I went through it with only a provisional idea of what I was going to say and how I was going to structure it. For example, I believe I may have the structure exactly backwards, and that the long first sections should be dropped or turned into a paragraph or maybe into a cute line drawing of a kitten.

This is the last chapter before the Grand Summation, of which we shall not speak, mainly because it causes formication over all areas of my exposed skin. In the current chapter I am writing about decision making because it is one of two proof points. The previous chapter is about science. Both that one and this one are intended to see if all the jibber jabber about networked knowledge that the reader has slogged through so far actually holds up in areas where we really really have to know what’s right and wrong. So, when we make a decision, does networked knowledge help? What happens when the rubber hits the node, so to speak?

The chapter as it stands begins by spending way too much time on the nature of distributed leadership. I spend page after page talking about Jack Welch as a counterexample (this will almost surely be cut drastically) to make the argument that modern business leaders take integrity as the chief attribute of leaders because organizations are Too Big to Be Led. Since you can’t be sufficiently competent in everything you would need to be, you claim that simply being a truthful, authentic person is enough. Yeah, sure. The fact that the memoirs of successful business leaders are often among the most inauthentic, squirmtastic writings around is just icing on the cake.

Anyway, I then argue that leadership, too, is becoming a property of networks, albeit it unevenly and certainly not in every case. I have a brief case study of the Army’s leadership center at West Point, based mainly on an interview with Lt. Col. Anthony Burgess. (The link is to a piece he wrote up after the interview.) I just don’t know if I’ve successfully sold the reader that an extended discussion of leadership is directly relevant to the topic of networked decision making.

I then make the point that I think I should begin the section with: If you look at decision-making as the isolated moment in which the bit is flipped, then you miss the networking of decision-making that goes on before and after that, even if the organization has no networked decision-making structures in place. Even when the decisions are made by the person at the top, they are made within a network that takes on many of the tasks and properties decision makers shouldered alone. So, the decision may still be a flicking of an leader’s thumb up or down, but that gesture may now occur within a network that has helped inform it, will carry it out, and will support it.

The final section takes a surprising turn for the practical. I was not expecting to end up there, but, when I checked my original outline, sure enough, that was exactly where I thought I’d be. I suppose that’s a good sign. Anyway, this final section’s premise is that to make smart decisions, we need smart networks (not in David Isenberg’s sense!). So, I quickly look at a bunch of properties of networks and loosely tie them to practices that will help make the network smarter than the smartest individuals in them. Nothing you haven’t heard before, which is, of course, a problem.

So, a very very very rough first draft that I may throw out tomorrow. Yay?


[2b2k] Science chapter

I’ve been over and over the draft of the chapter on science. I believe I’ve gotten the organization better, but it’s still 14,000 words, which is twice as long as it should be. I can see how to cut out about 1,000 words, but that’s not enough.

Here’s a rough outline of the chapter. Note that I’m paraphrasing myself as briefly as possible, so much of this will sound more over-stated than it is. That is, I’m over-stating my over-statements. Also, I proceed mainly through examples and interviews, which I’m not mentioning in this summary.

Intro: The traditional processes of science are turning out not to scale. We have so much more data, so many more connections. The Net does scale, however. The organizing hypothesis of the chapter is that science is starting to take on properties of the Net. Each of the six sub-sections looks at one such property.

1. Hugeness. The famous 1963 letter to Science, “Toiling in the Brickyard,” worried that science was being overwhelmed by facts. Yet, we now have exponentially more facts and data, and can scientifically address some phenomena that were too complex for prior techniques. In some instances, we get understanding without theories. This is a big change given our traditional Western idea of knowledge.

2. Flat, or at least flatter. Science has a long tradition of honoring amateurs. (The professionalization of science is relatively new.) This is a great time for amateur science. But, most of the contributions of amateurs are crowd-sourced and don’t require much scientific training; you still generally have to be a specialist/professional to make a big contribution. Even so, the fact that the work of professionals is available to anyone with a browser is changing the ecology of science. So, the distinction between amateurs and professionals will remain, but the Net is vining the gap.

3. Open and continuous. Rather than science following the publishing cycle of work being done in private before it is launched into the public as done, projects like “open notebook science” are making the process of science more continuous.

4. Open filters. Peer review continues, but is changing. E.g., PLoS. Publish-then-filter is becoming far more common.

5. Difference. The Net lets us see disagreements. Some important sorts of differences among scientists now are often (not always) left unresolved through mapping of schema rather than trying to come up with a single, right, true order of nature. (Many things are miscellaneous, I hear.) Differences among non-scientists are becoming of increasing concern to scientists because there isn’t a single set of authorities who can dole out the truth, and to whom the public listens.

6. Hyperlinked. Science had been governed and shaped by the requirements of paper-based publishing. Ownership and authority were established via getting published. But, science’s idea of knowledge itself at some level was also modeled on the publishing system: You work on an idea until it is ready, it passes through expert filters, and then it exists in the world in an almost thing-like way. Now that we can hyperlink all the way back to the source data, and now that what we make public (at any stage) gets linked to by those who discuss it, science is much more like a network than like a system for publishing results.

Finally, I talk about how this is not only a great time for science, it is also a great time to be stupid. If you want to ignore the inconvenient truths of science, you can surround yourself with a web of ignoramuses who provide a sham system of misconstructions that make falsehoods seem as profound as truths. But, the new stupidity in the Age of the Net is also due in part to the old stupid idea of science we got in the Age of Broadcast. Seeing networked science may — may — teach us more about how science works than did the announcements by the broadcast media of the latest, poorly-contextualized scientific study about Alzheimer’s and coffee, chocolate and heart attacks, and wifi and brain cancer. And it may lead us to view knowledge not as a set of self-standing truths, established and independent of us, so much as the constant processing of ideas through well-established, methods, by a complex network of humans — which long has been science’s view of truth anyway.


Government APIs rock

The FCC has launched a site for developers that provides APIsso that anyone can create apps that draw on FCC data. Heres the first one they list: “Over 1 million user speed tests were generated from FCC Consumer Broadband Test. This API delivers data on the number of tests, average user download/upload speeds, and more.”

The White House also launched, an Innocentive-like site where government agencies can pose challenges, offering prizes for the best solutions. There are almost 50 challenges posted so far.


[2b2k] Too much science

I finished a first pass through the science chapter of Too Big to Know. It’s supposed to be a practical application of what the book has argued so far — how does networked knowledge work in a discipline that is devoted to truth and reality?

In this book, a chapter should be about 7,500 words. This one is 15,000 words. But it’s that long for a good reason: I don’t know what it’s about. So, I’ve been reading through it, trying to figure out what to cut and how to organize it. For example, I have a 5,000 word subsection that has its own three subsections about science moving from a publishing paradigm to a network model. This afternoon I thought that maybe I could remove the publishing framing, treat the subsections as being part of the stream of subsections, and surprise the reader at the end by pointing out that what we’re actually seeing is science moving from being a type of publishing to becoming a network. That might work — I’m not sure yet — but it still leaves the chapter twice as long as the reader is expecting.

Or wants. The chapter doesn’t frame itself with a question that will catch the reader’s interest. Ulp.

Once I’ve read through the beast and outlined it, I’ll post about its content. Right now, I can’t even remember what’s in it. Which is not a good sign.