In my last post, I mentioned that two of NebuAd’s ISP partners had backed out of the deal—the potential privacy backlash, both from consumers and from Congress, was just too great. Following that news, NebuAd CEO Bob Dykes jumped ship (although he remains as Chairman).

Then, with no customers to deploy to, NebuAd made the announcement that they were holding off on deploying until Congress can get some satisfaction about the privacy aspects of their techniques. Evidently the combination of unfettered access to data doesn’t marry well with a desire to use that data for commercial gain. From washingtonpost.com:

Critics have likened deep-packet inspection to the phone company tapping a call. The technology allows a window into potentially all of a consumer’s online activity, from Web surfing and search terms to any unencrypted Web communication.

These are obviously tough times for NebuAd, but there’s one really positive thing for the market in general: the privacy issue is finally becoming one that people are reacting to with their pocketbooks in addition to their dramatic words.

For a long time, privacy issues have been a political hot button rarely backed up by action. As much as people rant and rave about it, we continue to use Google and leave cookies enabled and give every random developer of a Facebook app the keys to our social graphs. The ultimate irony was the anti-Facebook Beacon group that was formed on—you guessed it—Facebook. Obviously we were happy to complain, but unwilling to put in the necessary effort to change habits.

NebuAd, by tracking activity via ISPs, finally took things too far.

“The sense I get is the air is out of the tires as it relates to targeted advertising through deep-packet inspection,” said Robb Topolski, a technology consultant. “The users have made it very clear that they don’t want any part of ISP monitoring regimes that watch everything they do and say on the Internet.”

I’m really glad to see people’s privacy concerns having a direct result on the way these technologies are deployed, and it’s clear what it takes to get rid of technologies that don’t cut the privacy mustard: public outcry, PLUS government objections, PLUS commercial customer pushback.

It remains to be seen whether NebuAd will be able to successfully reinvent itself in the light of current market concerns. For now, though, the consumer’s demands have won.

What are your thoughts on deep-packet inspection?

Just recently, we’ve been squarely in that first phase, starting with

Anecdote 1: Elyse Tager’s piece Privacy and Behavioral Targeting Heat Up at ClickZ.

Elyse talks about the challenges faced by cookie-based behavioral targeters who use historical information to infer future behavior:

NebuAd launched with all best intentions, attempting to address the issue of scale with its now huge network — a major disadvantage for behavioral targeting in most cases. Plus, NebuAd has a robust privacy policy addressing consumer concerns directly.

But last week, two of those ISP partners backed out of the relationship. Charter Communications announced it was withdrawing due to subscriber concerns. CenturyTel is pulling out after the warnings from Reps. Edward Markey, a Massachusetts Democrat, and Joe Barton, a Texas Republican, who said the technology “raises several red flags.”

The red flags in question have to do with the amount and detail of data being collected about individuals. It’s a problem endemic to targeting solutions that rely on knowing as much detail as possible in the hopes that the behaviors will be repeated.

Tager points out two solutions being proffered: data portability and predictive modeling.

Another solution, which I covered in earlier columns, is predictive modeling to better target behavior. Companies such as aCerno and Epic Advertising use advanced algorithms and technologies that don’t rely on cookies to establish inferred behavior, which is less intrusive and far more predictive of future behavior, according to these suppliers.

As you’ll know if you read this blog with any regularity, VortexDNA relies on a predictive modeling approach.

Personally, I believe data portability is only a solution for the technological elite. It is simply not feasible to ask my mom to manage her data.

The FTC is pushing for self-regulation. If the issue were between the Market forces, on the other hand, will make a difference. NebuAd’s ISP partners backing out will make a difference. Will it be enough, though? Congress might not think so, which leads us to

Anecdote 2: Heather Green’s piece Congress to Push Web Privacy at BusinessWeek.

On the second page of the article, Green mentions The Center for Democracy and Technology’s desire for a Do Not Target list (along the lines of the Do Not Call list). What she doesn’t mention is the obvious problem with such a list: in order to not target you, they have to know who you are. So you have to register in some way, giving them the very information you’re hoping to keep from them.

She closes with an apt comment on the benefits of federal privacy legislation:

Some in the industry think that legislation might be the way to set a common standard and avoid inconsistent, piecemeal legislation on the state level. Microsoft came out in 2005 in favor of federal privacy legislation and thinks others are beginning to agree. “Companies are coming around to the notion that it’s not only compatible with their business practices but [that it] can help them by enhancing consumer trust and making compliance more streamlined,” Microsoft’s Hintze says. Microsoft advocates privacy baselines that cover not just the online collection of data, but offline collection as well.

David Hallerman, analyst at researcher eMarketer, says legislation would go a long way toward assuaging fears of advertisers who fret consumers don’t want their privacy compromised. He says that if an online privacy law were passed, “the benefit would be there for advertisers, publishers, and the public.”

Given public concerns about privacy, I tend to agree with Hallerman. Allow companies whose practices are aboveboard the opportunity to be recognized as such. I like that Microsoft is getting behind it. Google seems to be going in a different direction, as evidenced by

Anecdote 3:Wendy Davis’ piece Polish On Google’s New Chrome Tarnished By Privacy Questions at the Daily Online Examiner.

So Google is finally trying to take the battle to the Microsoft-controlled browser terrain, instead of just hanging out comfortably on the high ground of search and letting Microsoft lose battalion after battalion in a series of poorly-planned attacks. According to Davis, though, the new browser is a long way from offering any privacy benefits:

…the browser raises significant privacy questions. Google states in the Chrome privacy policy that it will log the IP addresses of people who download the browser. It also says that all URLs or other queries typed into Chrome’s address bar will be sent to Google, which will use that information to make suggestions to users.

The browser’s privacy policy says it will “process” information received from Chrome users but — in a crucial omission — doesn’t say whether it will retain the data or for how long: “Information that Google receives when you use Google Chrome is processed in order to operate and improve Google Chrome and other Google services,” the policy states.

I don’t think consumers are going to go for it. It’s too intimate to gather these different services together. It’s like your bank buying your DVD store, and the guy who approves your loans also gets to know about your perverted taste in movies.

The bottom line is that the privacy landscape is shifting. So how should you handle your own privacy policies?

Do the right thing.

Some years ago, a friend of mine urged me to become an SEO, touting tactics that were legal and fine at the time but that would be considered black hat today. “It’s so easy! All you have to do is use this automated program that will create hundreds of sites at once, all linking back to your client’s site.” Thankfully, I didn’t go for it; I might have made money in the short term, but it would have been bad news in the long run.

The same holds true for privacy. Forget about what you can technically get away with, or what you can assume your customers won’t notice. Just do the right thing. Be fair. Consider the customer. Consider the cost to them as well as the benefit. And behave in a way that lets you hold your head high.

In a mixed up, muddled up, shook up world, your integrity is a real asset.

Your thoughts on this topic are welcome.

“The Googling I: Google Maps”

“The Googling II: Google Moon”

“The Googling III: Google My Maps”

“The Googling IV: Google SMS”

Enjoy…

As you well know by now from reading this blog, the VortexDNA algorithm creates numeric profiles for websites or objects by aggregating the profiles of everyone who has visited that site or object. So if a site is visited by someone with number 100 and someone with number 200, the site’s profile becomes 150. That’s dramatically simplified, obviously, but the essence of it is accurate.

That mechanism leads to an obvious chicken-and-egg variable: in order to be able to indicate relevance, the objects or sites have to have enough data to generate an accurate profile. If only the person with the number 100 visits the site, the site’s profile will not be complete. You cannot access the wisdom of a system by looking at only one of its component parts.

Right now, we’re working at further developing the wisdom of the VortexDNA system; to do this, I’d like to ask your help. We’ve created the VortexDNA intention game. If you spend a few minutes playing around with it, you can help us generate more accurate data that isn’t specific to any one individual.

It only takes a few moments, so go on, take a break from work or the Democratic National Convention and have a play.

Thanks in advance!

To recap, the boundaries that define a company as a complex system are as follows:

• staff
• shareholders
• customers
• community
• society

The changing nature of the world has changed the way that these boundaries work. In The World Is Flat, Thomas Friedman makes the case that there is no longer any difference between Baltimore and Bangalore, between Seoul and Silicon Valley. Ever-increasing connectivity has dissolved distances and removed barriers.

So if barriers are gone, does that mean there are no more boundaries? Far from it. What it actually means is that our alignment across all focus areas is more important than ever — particularly our alignment with community and society.

In the old, non-flat, protectionist world, domestic manufacturers didn’t experience the friction of foreign products. Domestic unions were free from the friction of cheap overseas labor. Large publishing houses didn’t suffer from the friction of upstart bloggers.

In a flat world, all of these things are in contact with each other.

Imagine a ball rolling along a carpet. In the old model, the ball hovered above the carpet — the buffers of disconnectedness, isolation and protectionism meant that our actions didn’t provoke the friction with the outside world that they otherwise might.

Over the past few decades, all of those buffers have been ripped away. Thanks to the Internet, bandwidth, teleconferencing, and inexpensive travel, our behavior — which used to be compared only to our immediate neighbors — is now interacting with (or rubbing up against) the behavior of people halfway around the world.

So it’s more important than ever to have an inclusive focus across all stakeholders and boundary areas. The flat world means that we can’t ‘get away with it’ anymore. The highest global standard is the standard we are held accountable to. And, like the law, ignorance is no excuse for noncompliance.

How have the changing boundaries of the world affected your business?

The book Built to Last, by Jim Collins and Jerry Porras, describes many fascinating characteristics of visionary companies, with the overall gist of the book being that companies that align themselves around a shared purpose and values outperform those that don’t. One of the more interesting of the authors’ observations is that it doesn’t seem to matter what that purpose and those values might be.

In particular, Built to Last uses as one of its reference companies Philip Morris. Regardless of your feelings about smoking, few people think the words ‘values-driven’ and ‘tobacco’ go hand in hand. But Collins and Porras describe a strongly adhered to and clearly articulated culture, passionate about freedom of choice.

While those values are consistent with the outside world, the company will flourish. While they are inconsistent, the company will suffer. Good/bad, healthy/unhealthy… these subjective concepts make no difference in this equation. What matters to company performance is what matters to the stakeholders, all of them: staff, shareholders, customers, community and society.

If the staff is passionate about their values, the shareholders are getting a good return on investment, the customers love the product, and community and society are supportive or not opposed, then the company will do well.

It is only in the past few decades that cigarettes began to be considered a vice—and the tobacco companies have paid for it. Increased PR expenses, litigation costs, decreased revenues… you can see the challenges that can be created in a company when it falls out of alignment with the society around it.

So this is the concept for today: self-alignment is not enough. Customer service is not enough. Taking care of relationships is not enough. Our companies—our lives—are governed by the totality of our focus.

Have you observed this in your own experience? I’d love to hear about it.

Nobody on this planet is exactly like me. Nobody thinks my thoughts, dreams my dreams, or lives my life. In so many ways, I am a snowflake: an utterly unique, totally individual, mold-breaking creature.

And yet, I am vastly predictable. For example, I am guaranteed to cry at any and all opportunities — and I mean ANY. An episode of The Simpsons where Homer feels bad about not being a good father. Lethal Weapon 4 where Joe Pesci talked about Froggy. This ad for New Zealand Post (I have to change the channel when it comes on, or risk breaking down into an incoherent sobbing mess).

I am guaranteed to cheer, loudly, at sporting and theatrical events. I am guaranteed to want to see the latest big-budget action flick. I am guaranteed, in short, to behave in certain ways, ways that can be tapped into and understood and used to identify me as a potential customer.

So am I unique or aren’t I? In Sunday’s Guardian, a variety of thinkers sounded off on the Pros and Cons of a Google World. Adam Curtis, the director of The Power of Nightmares (a TV documentary on terrorism), had this to say:

The millions of searches that engines like Google record and store reveal the shifting desires and fears of individuals. They’re leading to a new fragmented sensibility among millions of people in the way they see and experience the world. Machines like Google know something about us as human beings that we really don’t want to know - that we are not individuals: ‘If you like this then you will like that…’. So Google is a paradox. It gives us the feeling we are wild and free individuals, powerfully reinforcing an idea of us as heroic figures in the consumer age. Yet at the same time it is powerfully proving the opposite - that we are completely predictable. Out of that is going to come some very interesting political ideas of how to organise society and also new artistic ideas. The really interesting question is whether it is really a cult….

Here is the thing: if we were truly unique and truly unpredictable; if our actions were in no way interconnected and in no way integrated; if we didn’t have some means of anticipating, to some degree, the behavior of others — our lives would be ruined.

You wouldn’t have any idea whether your co-workers would show up or whether your spouse would be waiting at home. Events that require critical mass, like rock concerts and political movements, would be impossible. And none of the products that tap into the short head of the marketplace would exist.

In order to function, the complexity of our lives requires repetition, patterns, and short cuts. If we had to make a truly individualized decision for every action we take, we would become paralyzed. If we tried to assimilate all of the data available to us at any given moment, we would go insane.

And so we become hybrids. We are predictable to one degree and individual to another, and the combination of the two produces a unique result that nonetheless overlaps with millions of others.

Thank goodness. If we didn’t overlap, there would be no airlines, no iPhone, no Internet.

So are you happy to be one of the crowd? Or do you see yourself as a unique individual?

P.S. I haven’t forgotten the last post, and my promise of a post on Big Tobacco and complex systems… coming soon! Also hat tip to Brian Hayes for the link to the Guardian story.

• staff
• shareholders
• customers
• community
• society

Each person’s complex system has five corresponding boundaries:

• happiness
• earning respect
• relationships
• community
• society

We’ve explored before the concept that the boundaries are not independent of each other. Those companies, for example, that have an inclusive focus on all five boundaries are the ones that significantly outperform the rest of the market. This makes a lot of sense if you think about it: imagine a company that took really good care of its shareholders, its customers, its community and its society, but that neglected its employees. This fictional company would not have a sustainable model, because ultimately the people who drive the corporate activity would not continue to perform.

In previous posts, I’ve shown the integrated nature of these boundaries by drawing a circle encompassing the five individual boundary circles.

But I’ve recently been visualizing it in a different way, and I wanted to share it with you and get your feedback.

Boundaries in Complex Systems

The reason these things are interconnected isn’t because there’s a circle around them, it’s because each one interacts with the others, and any friction in that interaction affects the effectiveness of the system.

So the boundaries are really more like gears with ratchet teeth, which work really well if they’re all traveling in the same direction. If one focus area starts to go the other way, though, the system faces real problems. So the boundaries might be better visualized this way:

The innermost circle is Happiness, followed by Respect, Relationships, Community and Society. You can easily see how one boundary might be bigger than the next but they all have to work together for the system to function.

In my next post, I’ll discuss Big Tobacco and complex systems. For now, I invite your feedback on this revised model. Does it make sense to you?

Anyhoo, I’m back now, and noticing an uptick in the number of articles mentioning the word ‘bubble’ or sneaking around its edges. Take for example Adam Lashinsky’s piece in Fortune, describing his sense of deja-vu during his visit to the Twitter offices:

Facebook and YouTube have yet to gush profits - a fact that is the talk of Silicon Valley. Yet here I am again, in July 2008, listening to yet another boyish entrepreneur discuss a quirky, compelling - and nearly revenue-less - startup… Only in the tech business are companies born with neither a clear reason for being nor a clue as to how they’ll produce profits.

When it comes to economic activity, be it real estate, stock markets, or venture capitalism, there’s a universal tension between the “fear of losing” and the “fear of not winning”. Nobody wants to be in the market when it crashes, but nobody wants to be out of the market if it keeps going up.

These two fears do battle as markets rise and fall, transitioning through what I’m going to call the Five Stages of Twitter. And, if you’ll indulge my flight of fancy, I’d like to take a look at them through the eyes of a potential shareholder, in a parallel universe where Twitter’s already had its IPO.

Stage 1: Denial
The first few indicators begin to amass that a market is ready for a slowdown, but you are having so much fun on Twitter, you’re absolutely certain they’ll be able to monetize it somehow. They can’t provide a service this important to this many people without finding a way to extract value. When they do, you know it’ll be huge—and you’ve got a big fear of not winning. You buy more stock.

Stage 2: Anger
You get the Fail Whale for the fifth day in a row. “They can’t get away with this!” you fume. You start the Twitter, Fix Your Servers! group on Facebook, and garner 42,376 supporters in three days. The passionate cries of the Twitterers show you how much grassroots support the company has—and you don’t want to miss out on that kind of movement. You buy more stock.

Stage 3: Bargaining
You read that Twitter not only has no revenue model, they actually think it would be a mistake to come up with one. The Twitter CEO says they’ll monetize “when the time is right.” Somewhere else, a man tells his girlfriend he’ll leave his wife “when the time is right.” You decide that you’ll only keep your shares if the company gets support from the venture community. They get $15 million. You buy more stock.

Stage 4: Depression
Your tweets get shorter and shorter as your tenuous grasp on the English language slips away. Where once 140 characters seemed like an impossible restriction, now you struggle to fill the space. You realize that the more time you spend Twittering, the less you have to tweet about. To feed your gaping spiritual void, you buy more stock.

Stage 5: Acceptance
Twitter has yet to realize any revenue. Ebay writes off $900 million of its purchase of Skype, a figure Google seems destined to outdo with YouTube. Facebook downgrades its valuation from $15 billion to $3.75 billion.

You sell your Twitter stock, and buy shares in Plurk instead.

The above commentary is intended to be a bit of a giggle, and does not constitute investment advice. Follow me on http://twitter.com/kcolbin.

You are also welcome to download the PowerPoint slide show here. (Note: It’s just over 5MBs.)

I hope you enjoy the presentation, and I look forward to your feedback in the comments!

Part I: What is a complex system?

Part II: Complex systems at work

Part III: Predicting human behavior