Dross, Gloss and Brilliance…

There was a recent posting on the Information Systems Manager’s forum that has me dredging up the past. There, the question was posed as to the value of blogs – more specifically do they lead or follow, or are they relevant at all. Should one read ‘em or ignore ‘em?

The author of the note postulated that, aside from politics and technology, they tended to be reactions to either traditional media, to other web sites, or just so much tripe about relatively inconsequential things like the babies of hyphenated or concatenated movie stars.

At first blush, I kind of agreed – after all most are pretty much regurgitated thoughts about stuff and junk found elsewhere on the web, a few notable examples aside. This got me thinking about two things: first, the birth of so-called “citizen journalism” and second, how the media universe has changed over the last few years.

While some pundits have postulated a new era of “citizen journalism,” it really seems to me that we’re seeing the birth of “citizen wonkism” or maybe the “citizen editorialist.” The idea of the citizen journalist, I think, is wishful thinking, something brought on by the rapid descent of traditional journalism into vapid sensationalism and outright propaganda. What passes for journalism today is rapidly losing the long and fine traditions of something called “reporting” and objectivism.

Despite the demise of decent journalism and despite our wishful thinking, most blogs don’t report either. Many blogs just regurgitate, or nitpick. Some are just incomprehensible blather. Others are entertaining, but insipid. A few are worth their salt. So, while the blogosphere and journalism seem to share some commonalities, they are not attributes to be proud of.

I temper these thoughts by remembering Sturgeon’s Law: “90 percent of everything is crud.” So it stands to reason that 90 percent of blog content would also be cruddy stuff. But the remaining 10 percent provide reason for hope.

[I should note that there is one exception to the "90 percent of everything is crud" rule — namely crud itself. 100 percent of crud is usually crud. There are always exceptions.]

Nevertheless, as I mentioned, the question about blogs also got me thinking: Where do the 10 percent that are useful fall in the general scheme of things? Are they useful sources of information about what’s going on in the world, what’s important, and what’s relevant? In general, do they lead, follow, or meander around in between? .

My question stems from my past. I used to pay a lot of attention to the media, lots of media; something on the order of 200 local papers a day, and over 150 trade journals a month. As I said, lots of media. Then, I read, daily, everything from the Wichita Eagle Beacon to the San Jose Mercury News; Chemical Week to Ad Week, and a little Industrial Distribution just for variety. Whew, was I fun at parties!

“Why?” you might ask. Well, simply put, it was believed that you could use the MEDIA — in the aggregate — to tell you about what might be, predict the future so to speak, track it, or at least consider the big “what if’s” and “suppose’s.”

In fact, my never-finished dissertation was all about how to spot and track trends, important political or social issues, and the like, by closely watching ideas and issues as they moved through various media – a quasi-science called “Early Issue Analysis” or sometimes “Emerging Issue Analysis” or sometimes just called “bull.”

It’s a fairly simple model — commonsensical even. It’s an expansion of some early OSS work during WWII and a refinement of something called the “Molitar Model.” The names used to describe it are many, depending on the marketing hipster involved. You may even see it bandied about today: sometimes called “Precursor Monitoring Model” or “Environmental Scanning;” or sometimes it’s just emerging issue analysis or, my favorite oxymoron, “Issues Management.”

I spent several years of my life applying, expanding, and testing that “emerging issues” model. I authored an extremely boring study [at least that dissertation got some use] examining how to use some fancy new things called “online databases” to automate the tracking and analysis of social issues through media. Somewhere in the archives of the Electric Power Research Institute (EPRI) you’ll find it, no doubt dusty, unread and unloved. Que lastima.

All-in-all, it’s a couple of hundred pages of testing and validating that “Emerging Issues” model; first historically against a fairly well documented set of issues (acid rain specifically), and then looking forward at some so-called emerging issues for the time (global warming, thermal pollution, deforestation, etc).

It was a variation of the same techniques we used on Megatrends – but using online databases. The truth is I was tired of reading all those newspapers and magazines and figured we could automate some of the process using these new fangled things called “databases.” And, given the power of technology, I finally had a way to see if the model held true looking backward. If it did, I wanted to try to apply it looking forward.

All in all, my colleagues and I started experimenting with some interesting tricks that one might use to spot and track emerging issues using high-speed, structured, scans through online collections of media — trade press, legislative press, local and national newspapers, etc. etc. High-speed content analysis of online media — fun stuff really. There were no words for it then. Today, I think, we’d call it “data mining,” but that doesn’t do it justice. Nevertheless, it was all based on the same “emerging issues” model.

Here’s a picture of the basic concept. I’ve dropped a few familiar names into the various quadrants to give you an idea of what falls where.

Da Media is Da Message?

Basically, the model said that if you watch the left side of the equation, and if you’re good, you can spot things as they “emerge” and before they hit the mainstream consciousness. The very classic example is Sir Arthur C. Clarke’s “invention” of the satellite in science fiction way before it was a reality.

Did the model work? Yep, it did. And, proving it with historic data, where you know the outcome, is a breeze. The difficult part is to use it looking forward. By the way, the question you should be asking right now is: “How long is the X-axis?” In other words, how long does it take for an issue to, first emerge, and then make it to the general public consciousness on the far-right side of the graph? The answer is: it varies by issue and by sector.

But I can also tell you that the X-axis is shrinking. What used to take decades no longer does. The time from the left to the right of the chart is now measured in just a few years, if not a few months.

To look forward, the trick is to separate “signal” from “noise.” There is a lot of stuff out on the fringes — a lot of noise — spotting what’s important is not easy. While surprisingly accurate on known issues, looking forward remains a problem — there is just too much noise to filter.

Filtering requires genius, and that is the essence of what is known affectionately and officially as “genius forecasting.” Most of the great futurists were, first and foremost, great genius forecasters. They just had a knack for spotting the important issue of the day, the next great technology, or the next bit controversy. I include in the list folks like John Naisbitt (who I then worked with), as well as Alvin Toffler (Future Shock, The Third Wave), and Herman Kahn (Thinking the Unthinkable). M. Kahn, just so you know, was supposedly the inspiration for Peter Seller’s rendition of “Dr. Strangelove.” Although M. Kahn was a wee bit larger. Strangely, in recent years, we’ve seen very few great futurists, or even lousy ones. I think it’s because the X-axis is so short now, and the future so seemingly unpredictable in a predictable way.

Just between you and me, I think half the trick to good “genius forecasting” is to hire a bevy of bright, young, starving innocents; fill their heads, five days a week, with a select collection of stuff from “precursor” media, and then force them to tell you what’s “important” every month or so. Hell, make’em read science fiction, fan-zines, and a half-dozen blogs, and you’d probably do better than most of today’s economists and other soothsayers.

Mind you, it’s the “innocence” that’s important. The last thing you want is “experts” — they’re way too involved in whatever topic they’re an expert in, to see reality. After five years of so, rent out your now-not-so-innocents as consultants, and get a new flock.

When “trend spotting” there are several key phases to look for, by the way:

First is the “naming.” It’s when the issue enters the lexicon. It is, for example, when the issue of coal fired power plants and air pollution came together in a single term called “Acid Rain.” Acid Rain first entered the lexicon — made its first appearance as a single descriptive term, in around 1854 according to my dusty research. Naming is very important. Controlling the naming gives you power, and can make or break an issue in the public lexicon, and in the public consciousness.

Second is the emergence of the “champion.” That’s the individual or group that takes the issue and makes it heard. Acid Rain’s champion, for example, was a fellow named Gene Likens studying the forests in the Finger Lakes region of up-state New York. In more recent history, I’d say Granny D is/was the champion for the campaign finance reform movement, and Cindy Sheehan for the current anti-war movement; perhaps — a slightly ironic choice — Michael Moore for health-care reform. Perhaps too, we can throw in Al Gore for global warming, Lou Dobbs for immigration reform, and Thomas Friedman for globalization.

I am sure this model would still work— with appropriate modifications for the Web2.0 world — if you could figure out what falls where. The ‘net has messed things up. Sorting out what goes where — what media leads, what media follows — would be the trick. Nevertheless, I had put the whole concept aside as no longer applicable in a wired world.

In fact, when I really thought about it, I figured a much more interesting approach would be to try and aggregate, quantify and track the questions people were asking on Google. The idea was to group generic questions into categories and then track the ebb and flow of the nation’s attention — as represented by those categories. It seems you can do that to some extent with Google Trends. Not quite, but close. If there were a way to close the universe, to track the questions as they ebb and flow as a percentage of a total zeitgeist, then you might have something. What I am not sure. Perhaps this is a million dollar idea, perhaps not. Either way, I digress.

When the “Precursor Monitoring” model was developed, there was no Internet to speak of. On and off, over the last ten years, I’ve noodled a bit about just how and where such a model could apply to all the “raw” stuff one now finds on the ‘net. And, that gets me to the point of this discussion: Where would blogs fall?

The answer is, obviously — and consequently difficult — everywhere. Blogs are a generic thing, just like a print magazine — some lead, some follow and some are way, way out in space. Many are crud.

That, by the way, is the big change between now and then. No, not the crud — 90-percent of everything has always been crud. It still is. The big change, the difference between the pre-Internet media universe and today is that media no longer fall into nice neat categories.

Pre-Internet, one could reasonably lump leading and trailing publications by type: fringe media first (including science fiction), followed by low-circulation newsletters and the alternative press, followed by local papers (for local events), followed by trade journals and specialty journals, followed by the national newspapers, followed by the national magazines, and, lastly, followed by various government records and congressional publications.

Science fiction probably still leads, somewhat, but the rest of that categorization is shot to hell. Everything is all over the block. “Big” media is bigger and more monolithic than ever, yet audiences are fracturing into smaller and smaller demographics. Simultaneously, independent and “personal” publishing — of all variety — are experiencing a renaissance the likes of which we’ve never seen.

Given that, to use this model today, one would have to individually categorize various media, including blogs and web sites and e-newsletters, and assign them, uniquely, to a particular quadrant. That would be the trick: to find and track the sites, online publications, e-newsletters, and blogs that cover first quadrant things, namely, what I would call leading ideas, events, authorities, advocates, and literature.

I would argue that we could use something similar to Molitor’s Model in today’s world, but, but we need to, first, expand it to include many more types and genres of media, and second, refine it by actually cataloging various sites, bytes, blogs, newsletters, magazines, ‘zines, and the like, and place them in the appropriate quadrant.

Finally, since we’re wishing, you could combine that with a good “first use” system, a kind of “reverse Google,” that tracked the appearance of new terms in the global lexicon and you’d have a decent environmental scanning model. Wouldn’t that be cool.