Cory Doctorow wrote a very interesting fiction piece entitled ‘Scroogled’ which provides an anecdotal account of the potential of such privacy infractions

http://www.boingboing.net/2007/09/17/scroogled-cclicensed.html

As you say, this is a clever move by YouTube to enhance the accuracy and efficiency of its advertising model. When Google acquired YouTube, it saw the immense potential of a popular user driven, and interactive website, where users voluntarily tell the website about their personal interests, and through doing so, can be delivered targeted advertising. Now that Google has snatched up its prize, for the hefty price tag of 1.65 billion, it needs to find a way to justify the purchase by converting video watching into advertising clicks (and revenue!).

Google’s motive is clear: it is trying to come up with a way to effectively monetize YouTube, and that means intimately knowing the interests of its viewers and relating that to advertising dollars. Increasingly people are drawn to the Internet as a way to pull interesting content in front of them, and are deterred by the old TV approach that says ‘here’s what we’re going to give you, take it or leave it.’ YouTube is content rich, but Google has discovered (as has Microsoft with its investment in Facebook), that it is challenging to deliver an *un-intrusive* and *effective* advertising system that is consistent with the Internet user’s ‘what you want, when you want’ approach.

This new YouTube Insight system might be meant to improve the tools that users have for creating custom content…but I can’t quite believe it. Yes that might be a byproduct of the system, but how many people are really going to customize their video content regardless of what demographic is watching? Don’t get me wrong, there are some people or organizations that will take advantage of this – users that have a business stake in knowing their audience can use these tools to plan around the information they gather. Perhaps ABC wants to know what demographic is watching Jimmy Kimmel’s recent skits and promote the show accordingly. Maybe a music executive will want to know who is watching their firm’s most popular music videos and use that knowledge to better design merchandise. But my guess is that the average Joe is not going to have either the inclination or the resources to use this YouTube Insight to change the way they interact with the website.

Is YouTube Insight a bad thing though? Not necessarily. The truth is that YouTube probably already had most of these statistics gathering tools in place before Insight became a public system. It’s just that now other people can use these tools at their discretion and additionally, there’s full disclosure that YouTube collects this information. It’s very Google-esque that the company would say “hey, we’ve got all these great tools” let’s put them in a package and let people play around with them.

I do agree with you Danny, that today’s information age now presents increasing concerns to the public, whose actions are unknowingly, or knowingly, being observed under a microscope (or MS excel!). That said, I don’t feel that what YouTube is doing is any different than what TV companies or any business has been doing for years … collecting information about its customers to enhance the efficiency of the business. What is the boundary that companies need not cross in relation to data mining? I don’t know, but what I do know is that we as consumers and Internet users should be aware of these increasingly common practices, and how they may or may not make us vulnerable.

Hi Danny,

I think you’re right on both counts. Any tools like these can be useful and privacy issues should always be questioned and explored.

Perhaps privacy options are in order?

It is their platform that is being utilized and they’re entitled to the right to expand it’s capabilities. However, since they are generating revenue based on the use of their platform, the users should have options.

A few separate but related points:

.It seems that most people only react strongly to a ‘clear and present danger.’ If there’s any ambiguity or time table beyond the immediate it’s usually ‘it doesn’t matter’ or ‘wait-and-see.’ Most seem unconcerned until something happens that negatively impacts their lives, and then they’re all over it.

.In a nutshell it’s up to a vocal minority to address these issues. Eternal vigilance and slippery slopes and all that.

.Web trend analysis is generally based on flow of numbers, not on individuals. This should make it easy to protect individual privacy. It’s OK with me if you know my general location and when I looked at your stuff, but I’m not so sure I want my IP address showing up all over YouTube.

.Any of you who haven’t used something like Site Meter or WhoIs might be surprised by what you can find out from your own home computer.

JudeF makes some good points, as well.

It seems like there has always been an underlying concern that ad-generated revenue models like Google might turn out to be not so potent, as evidenced the story that came out this past week. I’m pretty sure there’s always someone toiling away in the labs trying to find the killer app that will solidify their viability and satisfy shareholders.

JudeF is right, I think, in asserting that this tool will be far more attractive to corporate users than to the rest of us. They are the ones with target markets to identify and attract. What is interesting is that the vast majority of YouTube content that I (and I think most of my friends) watch is pirated TV shows, music videos, and movies. This is corporate-created content, but is being hosted on YouTube by users unaffiliated and at odds with that content creator.

YouTube and file-sharing programs are really the only places I go to consume media-company content, and I think that is true for an increasing number of people, especially members of the net generation. But media companies are still struggling to make money off people like me. I don’t watch TV because it has too many ads. I don’t buy content offline or online, because I don’t like paying for things I’m used to getting for free. I don’t use the online shows provided for free on some network websites, because they are slow-loading, full of software bugs, spliced with intrusive ads, and have horrible user interfaces. The media companies have been hoping that if they can crack down on content on YouTube and file sharing sites, they can force me to use one of their inferior delivery methods. But of course, with sites like AllUC.org and others flourishing, that approach isn’t going too well. The crazy thing is that my content is still ad-subsidized, both on YouTube and on the search engines I use for file sharing.

YouTube is still far from becoming a cash cow, and having media-company content is probably essential to that occurring. At the same time, media companies can’t survive in a world where everyone gets their content for free on YouTube. The obvious solution to both of these companies’ problems is for YouTube to share some of its ad revenues with content creators: both corporate and ordinary. Can YouTube ads, without becoming more intrusive, generate enough more revenue than TV ads and CD/DVD sales to satiate both YouTube and the media companies? I think the answer is yes, and the reason is YouTube’s potential ability to micro-target ads at minimal cost.

YouTube Insight has the potential to provide infinitely more timely and reliable viewership data than what media companies currently collect, at a fraction of the cost. Combine that with the fact that YouTube is quickly becoming one of the most popular ways to consume media, and for a media company it sounds like a pretty amazing way of distributing content (provided revenues are shared).

Is it possible that Insight is part of a larger YouTube strategy to move towards such a revenue-sharing model? Is such a model feasible and realistic?

First, thanks for the insightful posts from all three of you.

A couple of thoughts and a couple of questions.

I think the point that that Youtube is clearly targeting this towards companies is spot on. When I scan through the current favorites on the main page, I can’t imagine most of the videos being drastically re-written and re-crafted for a new target model somewhere around the world. But that’s the point. I think companies still need some serious convincing that they can effectively use Youtube to generate revenue.

Some time in the future, I suspect Youtube will come to a point where it needs to decide whether it continues to be a user defined community or whether it becomes a new business model to address media on the internet. Is this the fate of successful online communities in the future?
The privacy comments got me to thinking. Are we conditioned to think less of our privacy online? Do click boxes and pop-up fine print diminish our watchfulness online?

I think this is a large part of monetizing the internet. I can’t imagine someone giving a company the sort of information over the form or by mail that we’re prepared to give them online.

I think these are larger questions on the future of the internet. Openness, Privacy, Online Economics. How do you see these trends shaping the future of the internet?

I think the reason why this is such an interesting discussion, is because we’re talking about the future of the Internet, and although we use YouTube as an example, really we’re talking about issues that will affect businesses, big and small, across the globe. So far we’ve highlighted a few issues that I’d like to address: online communities, website content, and privacy concerns. Danny you posed the question of whether YouTube needs to decide to become a user defined community or develop as a new business model to address media on the Internet, and I think that the answer is that it needs to do both. To explain why, I need to give some backstory:

Wikinomics talks about the demographic change that has brought so many young and sophisticated technology users to ‘live’ the Internet, and I think that this Net Generation has seen an incredible change in how people perceive the Internet, compared to people brought up without this (and other) technology. Although the Internet as a technology goes back decades, I would argue that it has only been a mainstream fixture for the last 10 or so years. For the most part, the Internet is what it is now, because of many technical developments that have opened up a (virtual) world of possibility. Don talks about HTML vs XML in Wikinomics and even before that there was the transition from handwritten to typewritten, computers, DOS and Windows, Netscape, Google and much much more! If you’re interested, “The World Is Flat” is a fascinating book that discusses some of the factors that are leading to a more interconnected global society and it really complements some of the issues that Wikinomics addresses. I mention the Internet’s development because it has evolved a long way from where it started, and part of that evolution is the way in which people now interact with, and identify with, other people on the Internet. And that’s what brings us to the issue of online communities.

If you had asked me 10 years ago whether I thought online dating would succeed, I would have hesitated to say yes, because there was a stigma related to the Internet, and the pervasive idea that the real world was for real people and the virtual world was for business, functional tasks like e-mail/MSN/AIM, or ‘geeky things’ like gaming. But during that time, online dating really took off, and over the past 5 – 10 years people stopped viewing the Internet and other people on it (everyone!) as being weirdos or nuts … rather than a place for geeks and freaks, the Internet became hip and geek is now chic. Communities are essential for websites because they humanize the internet and create an environment where people can relate to other people – a dynamic that allows for an exchange of ideas and collaboration on a level that has never before been possible. Taking this a step further, businesses like YouTube, Digg, Flickr (and even this blog) have found the value of *enabling* people within their respective communities, who actually go out and build the content that a website ‘sells’ as a product. And that’s the world of Web 2.0, where your best resource as a business, are the interested and skilled ‘volunteers’ who become a part of the business and its growth. And this is what I was initially talking about with YouTube – it is a new model of addressing media on the Internet but an inherent characteristic of that model is that it is user driven and defined.

This brings us back to an interesting point that Will brought up. As a business, how do you engage and court an audience that has interests that are contrary to your own? Continuing our discussion of YouTube, Will mentioned that he likes the website because of its content (some of which is corporate/pirated), and because the content is abundant, convenient, and most importantly, free. As a fellow Internet enthusiast those are characteristics that I and many of the Net Generation also look for in a website, but let’s remember YouTube before it was purchased by Google, where you could watch almost any show, in full episode, and without ads. The irony is that now that there is a sizable company backing the website, it is now accountable to legal constraints and has (understandably) been forced to remove many of the videos that drew people to the website in the first place. And now that it has paid 1.65B for the website, Google needs to find a way to make money off of the community, which may involve measures that we as ‘customers’ and community members dislike, and which may actually drive us off.

Which finally brings us back to privacy. It’s apparent by now that these issues are all related and that privacy violation can result in a strong and public backlash that affects the size and strength of your online community. I certainly agree with you Danny, that we are conditioned to think less of our privacy online and there’s a few reasons for this. The first is that relatively few people understand the mechanics by which “click boxes and pop-ups,” Internet traffic, e-mail, and IP addresses relate to violation of privacy. They may care about privacy issues, but they don’t know enough to be concerned, which is why there’s been so much media coverage of Facebook in relation to privacy, and the public realization that “Oh my god, you mean other people can see what I’m doing?!” The second reason why privacy online is seen as less of an issue than in ‘real life’ is because, as John discussed, we as people react to things when we see it impacting us, (“clear and present danger”) and right now we simply don’t see Internet data collection as that life impacting. People are driven by convenience, and as Internet users, we don’t have to fill out the hand written forms that give away our personal information, which normally may require us to intervene in protest and opt-out. On the Internet, the information is collected anyways but even if we know it is, it’s convenient not to care, and it will stay that way for better or worse, until a situation arises where the risk is large enough that it’s convenient to care again.

When tracing the development of the Internet, it is interesting to note that someone’s view of the Internet’s current and future role in society will largely determine what they identify as its predecessor technologies. When the Internet was first conceived as a way for big databases and calculators to communicate with a telephone line instead of a leased line, it might have been thought of as a fax machine for computers. Then there was that phase when the Internet was the new Gutenberg. Everyone was going to have their own website that they could use to publish whatever they wanted. Newspapers and magazines were going to be replaced. Amazon and eBay brought with them the idea that the Internet was a new shopping mall or marketplace. The digital music revolution has painted the Internet as a descendant of radio and CDs. VOIP and email seem to make phones and snail mail the Internet’s predecessors.

The Web 2.0 craze has been redefining the Internet as a social networking and collaboration tool. Barack Obama has shown that it can replace door-to-door canvassing. Wikipedia has shown that it can replace a centralized editing process.

The range of ways in which the Internet is affecting our society leads me to think that it should be defined not as a particular technology, but as a trend in technological development much like the industrial revolution. The industrial revolution was a change in the way production was organized. Assembly-line factories allowed almost anything to be produced cheaply by unskilled workers by spreading out production across many workers with very specific tasks. This organizational model could be adopted for any product.

The Internet revolution has also provided us with new ways to organize production. Our computers are like blank factories: machines that can play music, print documents, display websites, etc.. Software and files are instructions to the computer on how to organize production of a given output.

During the industrial revolution, production using the assembly-line factory model required an initial investment. First, a production process needed to be designed for a particular factory. The same processes could be used for factories producing the same products. However, it was often hard to copy competitors because of secrecy and intellectual property rights. Also, many times parts of production processes were undocumented, making them hard to duplicate. After the design of the production process, infrastructure needed to be bought and the cash for initial operating costs secured. Finally, a factory’s success depended on being able to attract and train a large workforce that closely adhered to the centrally defined production process.

The corporation’s ability to centralize both the organization of production and the use of investment resources is a likely explanation for why it has become the most important economic institution in the wake of the industrial revolution.

While during the industrial revolution a large initial investment was required for every increase in production—i.e. every factory—increases in production on the Internet do not require any investment. Every user enjoys the ability to easily, cheaply, and completely publicize any production process—software, videos, etc.. These production processes are in a format that allows computers to perform them quickly and for an individual user. Because software is designed with user-centric interfaces, and users have free and easy access to a wealth of user-created documentation, it requires very little training to employ. The computers and data transfer technologies that are the infrastructure and factories of the Internet revolution cost virtually nothing. Once a production process is developed, it can produce an unlimited amount of output at near-zero cost.

The vaguely named “online communities” are the dominant economic institution of the Internet revolution. They will be different from corporations because the Internet revolution is different from the industrial revolution.

The value of any economic institution—whether a traditional industrial corporation, a government, or an online community—is its ability to produce consumption. For traditional corporations, the ability to produce consumption comes from the necessary investments for new production: production process, infrastructure, and training. The output of these companies is constrained. Factories have a limited output because they have a limited amount of infrastructure and trained workers. If firms are able to keep production processes from being duplicated by competitors, there will be additional output constraints. These constraints on production give both the firm and its output market value: people will be willing to pay for things that are scarce.

In the economics of the Internet revolution, users generate their own output in the form of user-specified computer behaviour such as playing movies or creating Word documents. This is done using production processes downloaded for free from the public domain, free infrastructure, and speedy self-training. What Internet communities provide in this equation, then, is production processes: ways of organizing, or automating, production. This includes the software tools we use, the media we consume, the information we apply to decisions, etc.

What distinguishes Internet communities from industrial corporations is that they provide for free production processes that can be duplicated quickly and cheaply by anyone, rather than selling scarce output. As many have said, they provide information. I am reminded of this Steward Brand quote:
“On the one hand information wants to be expensive, because it’s so valuable. The right information in the right place just changes your life. On the other hand, information wants to be free, because the cost of getting it out is getting lower and lower all the time. So you have these two fighting against each other.”

In the age of the Internet, information always manages to become free. Proprietary content posted on YouTube and blogs gets duplicated too quickly to prevent it from becoming public. There is plenty of software underlying sites like eBay and YouTube that is not open source. But in some cases it can be copied and hosted on another website. Many programmers are able to easily make modifications to programs they are not legally allowed to edit: uTunes comes to mind. Software design has also become increasingly focused on the user interface—which is obviously public—rather than the underlying code—which is becoming increasingly routine. Look at how many YouTube imitators there are.

The value of Internet communities comes only from the value of the information they generate in the future. Information that has already been created has already been made public. Even the technologies that enable communication and collaboration in the community are likely public enough that users who become dissatisfied with the community’s evolution can start instant carbon-copy competitors.

Because everything an Internet community creates becomes completely public, it does not own anything. Its value, therefore, comes from the value created by new information. In the collaborative, Web 2.0 model, most of this information is created by users. Since an Internet community’s value comes entirely from the future production of its users, those users will inevitably define the evolution of that community.

Where the industrial revolution was marked by labour strikes, the Internet revolution has had user revolts. Digg went through one. Facebook’s Beacon project is a popular example. These revolts have shown that users are incredibly powerful. Because their future output is what gives the community value, they effectively own the community.

As community owners, users will maximize their interests. Internet communities reward users who produce valuable content with community status. Producers are also rewarded on a personal level. However, there is no monetary reward in most communities. Again, this is because production has no market value because it is in infinite supply.

Monetary rewards to production were essential during the industrial revolution. They enabled workers in one industry to trade output with workers in another industry. Everyone in the economy can consume an unlimited amount of the output of Internet communities for free. However, producers in Internet communities must still pay for industrial and non-information goods. That is why it is so essential to find a way to give these users monetary rewards.

Notice that if a company owns an Internet community that generates a lot of revenue, that revenue will have to be shared with producing users. If it isn’t, producing users could easily and quickly setup a new site that operates in exactly the same way, but distributes revenue to themselves.

The big potential revenue source for online communities is advertising. There is no doubt that the ability to micro-target on the Internet could make ad revenue substantial for many Internet communities. But I see a few problems.

First, if Internet communities are user defined, and many users do not like ads, some communities may see users revolt against ads. We already have pop-up blockers and Tivo.

Second, it is tempting to suggest that the Internet will eliminate the need for ads. A lot of companies already complain that ads are ineffective. Starbucks has built one of the strongest brands in the world with no traditional advertising. Web 2.0 is all about marketers interacting with customers using Internet communities. These communities allow customers to communicate in two-way discussions with companies, and talk about different products and companies with other Internet community members. Successful marketing therefore involves developing a good reputation amongst Internet communities. Paid advertising may become completely ineffective as consumers learn not to trust these messages get all their information about companies from community websites.

Third, people are concerned about privacy. And the more concerned people are about privacy, the less revenue ads can generate.

One revenue source for Internet communities could be donations. Wikipedia is donation based. In Rainbows was pay what you can: it could be a model for all other media content. Industrial companies that want to build a reputation with an Internet community could donate to that community. But unlike ads, where the money goes to splash the company’s name everywhere, these donations would be subtle and not directly tied to clicks or demographics. A set percentage of donations could be paid to owners, while the rest would be distributed amongst users based on their relative levels of production.

Another possibility is giving revenues from non-information products to Internet communities. Apple has suggested making all music free, and paying for it through iPod sales. Companies like Amazon are already selling products. Maybe the highest-skilled members of Internet communities could sell their expertise to those who want private consultation. Owners could take a commission on the transaction, like eBay does.

Whatever the evolution of the Internet, I believe that it will be primarily determined by users who produce content. With no ownership over anything, and the inability to charge for past output, Internet communities’ value comes only from these users’ future output. Finding an appropriate way to give monetary incentives to these users will be important to helping these communities to prosper.

Just a quick comment, but I came across an interesting website that has a few stats about YouTube viewership. I can’t really attest to its accuracy but it demonstrates the type of information that YouTube Insight might provide in its coverage:

http://ksudigg.wetpaint.com/page/YouTube+Statistics?t=anon

[...] behavioral targeting good customer service? A few weeks ago, Danny wrote a post about the lack of concern� in divulging� personal information online. In it he worried [...]