The graph’s y-axis is the percentage decline in page views at El’s Extreme Anglin’ from the 7 days before each event, to the first 7 days of the event. Pageviews are a good proxy for overall angler interest. El generates hundreds of thousands of page views each week, so even small changes are significant. The x-axis orders events by date, from January 2008. The axis isn’t scaled correctly to show time, but holidays are fairly evenly distributed throughout the year. Events are shown by green dots, with a shortened date (month and year) and the name of the event.

The data is expressed as a percentage of the previous week, because while interest in fishing “waxes and wains” from year-to-year, changes week-to-week are normally minor.

All the events included last at least 7 days. Where one holiday runs concurrently with another event (for example, the “Lunar Festival” and “Love is in the Air” often clash), only the first event in the sequence is included. Interest in fishing also changes dramatically in the month new content is added, so events that clash with major fishing patches have been excluded (Noblegarden 2008 with patch 2.4, Hallow’s End 2008 with patch 3.0.2, and Noblegarden/Children’s Week 2009 with patch 3.1). Winter Veil is also excluded: The period leading to Christmas is particularly unusual – first students stop studying and have a lot of time to play, and then many players stop playing to spend time with family. This causes large changes in activity from week-to-week, which makes it hard to isolate Winter Veil in the data.

Only 12 separate sets of data can be compared. There is one out-lier – Midsummer 2008 – perhaps the early stages of Wrath of the Lich King testing may have caused a small traffic spike in the week before? The pattern shown on the graph is not certain. But I’m growing confident that events are increasingly impacting on fishing activity.

But why?

Explaining the Trend

Fishing is a solo activity that doesn’t change from week to week. So if a player becomes busy because they are involved in special holiday activities, fishing is likely to be one of the easiest thing not to do: It can “wait till next week”. It does not break any group commitments, such as raiding dungeons as a party of players.

Yet that doesn’t explain why holiday events appear to be having an ever-greater impact.

I expected to find a dramatic shift following the introduction of achievements, towards the end of 2008. These added new incentives for players to complete holiday activities. Yet the trend continues for 2 years, achievements or not.

Holiday events are sometimes improved, but the pace of improvements is very slow. They could not explain the trend because the majority of holidays have not been changed in the last 2 years. Similarly, fishing has undergone changes over the last 2 years, especially around the middle of the period shown on the graph. But none of those changes explain the event-decline trend seen.

A possibility is that players are simply becoming more focused on whatever “is new”. That pattern is apparent in “the Warcraft media” (primarily news websites), which increasingly obsess over the slightest change, often months before the change is playable. Part of a collective “addiction to previewing the future“.

Perhaps holidays are becoming more popular because a broader range of players are finding them appealing?

Appeal of Holidays

Holiday activities tend to be fairly easy to complete. There are some exceptions, but generally these events are accessible to a broad range of players, which is not true for all WoW’s activities. It would be easy to conclude that these events have become more popular over time, because of a general “dumbing down” in the average level of player skill.

Perhaps the average amount of time players play World of Warcraft is declining? Players would now be busier people than players in past, so to complete event activities they have to commit a higher proportion of their playtime, so are more likely to not go fishing.

Holidays tend to fairly social events. From sitting around a Romantic Picnic Basket, to sharing food, or placing “bunny-ears” on passing gnomes (to make them look even cuter). So perhaps the trend reflects greater social activity within the game?

The themes of holiday events mirror physical world festivals. From the Chinese New Year (“Lunar Festival”) through to Christmas (“Winter Veil”). So perhaps the trend reflects a greater desire for a connection back to the physical world? Diluting immersion in the virtual.

• A Tangled Web
• Eyeballing the Economy
• State of Information
• Communities of Serendipity
• Virtual Consumerism
• Different or the Same?
• Multiplicity of Self
• Postscript: A Chronocentric Tale
• But…

Axiological Interlude
Axiology, from Greek, is the study of value. Definitions of “value” vary between disciplines, for example: Economists tend to reflect Adam Smith-era notions of value – benefit to the buyer when used, labor to produce, gains from exchange. Sociologists tend to examine value through cultural perception – both personal and communal ethical differences – such as the World Values Survey. Philosophers initially consider value in terms of whether something “is good” – value theory. In practice, references to value often encompass (or even confuse) all these things. There’s a reason that I have not defined value…

A Tangled Web

Transaction costs are the key flaw in almost every attempt to build a payment system for the internet. Originally conceived by Ronald Coase, the logic is best explained as a mental accounting barrier:

Take a website. The average revenue earned from advertising is (in my case) about 1 cent (0.01 US Dollar) per visitor. Even with the most efficient payment system in the world, no rational human will waste time trying to evaluate whether to spend 1 cent or not. Assume someone can earn $12 an hour by working. Rationally, if they took more than 3 seconds to decide, they’d be “wasting money”. Trivial example, but it demonstrates why the information economy is increasingly either priced “free” or “expensive”: Handling tiny payments isn’t worth anyone’s effort. (People like Mike Masnick and Chris Anderson have written much about the resulting pricing models.)

A visitor that generates 1 cent of advertising revenue reads the website for an average of 4 minutes. Assume the time they spent reading they could have been working, 4 minutes equals 80 cents. Yes, value of time is a rather abstract concept, with many flaws. But there is a huge difference between the value of information received (expressed in time), and the payment for that information (expressed in indirect advertising revenue). There’s a tendency to under-value. Why do we under-value information? Because information is inherently:

• non-rivalrous – the amount one person can consume does not influence the ability of others to consume,
• non-excludable – once produced, there is no way to stop anyone consuming it.

Economically, it’s a public good. Individualistic, selfish humans simply take these goods. It’s (Western) human nature.

My writing is saving readers a lot more than the 4 minutes they spend reading, because (in this case) the information gained avoids time-consuming trial and error. Which points to another truth: Often information cannot be accurately valued at the point in time it is first consumed. This is probably true of everything, but is more extreme for information, which is inherently re-usable in different or unforeseen circumstances, often alongside other information.

Eyeballing the Economy

This is all part of a wider problem. Our economies are increasingly built around information, knowledge, skills, and other similarly intangible things, rather than physical production, or before that, land. The diagram below is based on the work of Derek Halden.

There are 3 circles representing different phases of economic development, chronologically, from left to right. The original diagram showed transport market failure, where the “Production and Consumption” market economy caused environmental costs that were resolved within the knowledge sphere. For example, the original transport-orientated diagram shows aspects like workers, vehicles, infrastructure and speed within the production sphere. Noise and emissions are outside in the Knowledge and Experience sphere. Economically, noise and emissions are “externalities” – a market failure, often requiring government intervention. Externalities tend to treated like public goods.

The stark implication is that within microeconomic theory, the competitive market is increasingly dysfunctional as the knowledge sphere grows in importance.

State of Information

Creating a competitive market for public goods tends to require government intervention. For example, copyright or patent law can make a public good excludable. Rights can be allocated. In theory, government can even act to reduce transaction costs.

Unfortunately typical state interventions tend to scale poorly:

• They can’t be used to effectively monetarise or protect $1’s worth of information.
• They are constructed around single right-holders, when so much “content generation” is collaborative.

Failing to scale down to the smallest exchange creates an inequitable structure, which only allows the owners of the most valuable rights to make any money from them. Make rights perpetual (give the ability to transfer ownership, especially after death, or as property), and an almost feudal society eventually emerges.

As the vast majority of wealth-generation within the economy becomes information-based, government will effectively control the right to make income. Ultimately such a high proportion of commercial activity becomes government-influenced, that it becomes rational to re-evaluate the reasons for maintaining a free market-based process.

Ironically, political government isn’t good at dealing with intangible concepts. Prestige, evidence, action – all are difficult to convey without a physical component. The underlying reason is that people aren’t good at dealing with intangible concepts.

Much contemporary thought (re-)establishes links with tangible aspects of the world: Home grown food that doesn’t “come from the supermarket”. Holes in the ice-cap as proxies for rising carbon dioxide levels in the atmosphere. It’s backward. It implies a finite limit on the information economy, since everything needs to be cross-referenced with a physical concept. It often still fails to convey value meaningfully – for example, financial instruments secured against physical bricks and mortar property, routinely exceed the value of the bricks and mortar, creating a dangerous accounting illusion.

Yet it is possible to simulate the emotions behind physical concepts using intangible forms. The solution lies in people: Specifically, the ability of people to manipulate other people.

Communities of Serendipity

I disagree with William McKeen, who declared that the internet made it more difficult to make serendipitous discoveries. Serendipity, “making an unsought finding“, only entered popular English language in the last 50 years. We didn’t need the word until recently. Yet randomised discovery is increasingly the only sensible way to proceed:

• During the Age of Enlightenment, one could know almost everything worth knowing by drinking a lot of coffee. Now it’s impossible to even understand one broad subject entirely. The days just aren’t long enough.
• James Austin defined 4 causes of chance in (medical) research, only one of which was pure luck. Active curiosity, unusual background, and distinct hobbies are also important. Chance, and therefore creativity and discovery, already contains a significant chunk of serendipity.
• In a highly intangible economy, with perfect communication, duplication and automation, theoretically only one person is needed to perform any activity: Once first done, the activity can be duplicated by machines, as often as the rest of society needs. Clone workers, with identical skill-sets, belong in the age of industrial mass-production.

In an optimised information economy, everyone is slightly different. Information discovery still needs to be bounded: Understanding most subjects requires a degree of specialism. And without good tools for constraining serendipity, we might never get anything done!

Serendipity can be bounded by freeform communities – “Communities of Serendipity”. The internet already does this, with limited success – from “blogrolls” and pages of interesting links, through discovery services like Boing Boing and StumbleUpon, to shared bookmarking, like Delicious. The idea is simple: Find someone with some common interests to you, and then find what else they are interested in. They’ll probably be similar enough to you, for you to understand anything they link to, but sufficiently different to sometimes expose you to something new.

Communities of serendipity have flaws:

• They assume reciprocity – that each member of the community will give as much as they gain from others. Human nature disagrees, something seen in the balance of activity in internet-based collaborative projects – Wikipedia, a good example.
• That people want to be different, when the whole notion of community implies they want to be similar.

These flaws become critical when trying to apply a value system – any form of economy – to communities of serendipity.

Virtual Consumerism

Hot-Or-Not’s virtual flowers encompass 3 components – the image of a flower, the gesture of giving a flower, and the “trophy effect” of other people seeing that someone received a flower. The second 2 components are most important to the service’s success. A lot like the primarily emotional values of giving a physical flower, except there is no flower. And the virtual flower is only valuable within that community.

In many online games and other social virtual environments, an advanced form of consumerism is emerging. Jean Baudrillard’s System of Objects characterises the value of the intangible component of goods with “symbolic” or “sign” values – the value in relation to another subject or group. In traditional consumerism, signs and symbols are in addition to concepts such as “use value” and exchange value.

The virtual consumerism found in many “virtual goods” consists only of signs and symbols. No significant production or utility values. If correctly implemented, no possibility of subsequent exchange.

Critically, there is no requirement for production using scarce physical resources. And so the internet transpires to be much more than a resource-efficient transport network: It’s a sustainable form of consumerism. Broadly the same consumerism that seems fundamental to giving (post-) modern society an illusion of purpose.

Virtual possessions can be used to convey far less tangible accomplishments, because in the appropriate virtual environment, virtual goods convey status just like physical goods.

Virtual consumerism is a far more attractive idea than “virtual currency” because virtual goods can directly link action to status. Value is more closely aligned to accomplishment: A contemporary failing of exchangeable currency is that the value of money is not precisely the same as the value of value.

The artificial scarcity required for such an economy is self-regulating: Rare (and hence desirable) items are only attained by those that accomplish something unusual. Common feats reward items that “everyone” has, and are thus less desirable.

Key to understanding why virtual goods “are valuable” is community: The community of people that recognize their meaning. Owning an Amani War Bear is worth nothing on the streets of New York, but everything when idling in Dalaran. Unfortunately this requirement for community highlights a conundrum:

Different or the Same?

It’s a curious quirk of evolutionary psychology that we claim to value inventiveness and difference, but actually tend to value imitation and sameness. The evolutionary success of humans is based on their ability to learn to overcome new challenges. Yet social acceptance is based on similarity to others. The genius that overcomes society’s biggest problems is invariably not a “dedicated follower of fashion”. Remember serendipity: Difference is core to discovery.

Fortunately, similarity is measured relative to community. And humanity is already organised into many different communities. Some would argue it is best organised into many small communities.

Within defined communities, concepts like virtual consumerism and even serendipity are possible. The problem occurs at the edges, where communities overlap – and communities of serendipity are all about overlap.

What problem? Money: The value systems of distinct communities are required to use a currency which is interchangeable between systems, even though each community actually has its own distinct values. For example, within their respective professional communities, a banker’s (infamous) million-dollar bonus conveys much the same value as a box of chocolates to a teacher. It’s only the exchangeablity of money which consequently appears to value bankers far more than teachers.

Virtual consumerism sidesteps currency as a means of exchange. Money is simply required to offset inequality. Inequality is a community-based concern: We primarily seek to even the wealth of people like ourselves. (One might even argue that morality itself is community-based.) In small, self-contained communities, inequality becomes irrelevant: Everyone is doing broadly the same thing (so has broadly the same wealth), direct assistance with actions replace indirect exchange, and who cares about anyone outside?

Yet serendipity requires people to be in multiple communities, hence experience the inequality between those communities, and seek to try and balance that inequality. Back to square 1.

Well, maybe not. It depends how we regard ourselves:

Multiplicity of Self

The idea of multiplicity of self grew out of my attempt to reconcile the fact that my Gnomish persona was not only better known than my physical persona, but that “she” had started to attain a degree of phenomenal consciousness – in the minds of some of “her” readers. This isn’t simply a case of my own mind effectively acting through an alternative, virtual body. It’s that the degree of “existence” of this self is primarily defined by other minds, not by my own. It is the other people that validate “her” – give her a meaning beyond pixels.

This is a more extreme version of celebrity, where the image of the celebrity in the minds of their fans invariably differs from the reality of the celebrity in that celebrity’s own minds: Visually, “she” is not physically me. It’s different from acting, since “she” is merely acting out me, not impersonating anyone else’s mind. And far more interactive than a character in a work of fiction.

It follows that the same rules should be applied to the physical me, which makes this very interesting indeed. I’ve struggled to find a philosophical concept that specifically describes my observation. Perhaps, a sort-of inverted-existentialism (where existence is primarily defined by everyone else) mixed with relativism (in which understanding is relative to others), all integrated into a re-configured mind-body dichotomy? I don’t understand it, and consequently can’t start to rationalise or debate it.

Conceptually, the multiplicity of self is very useful. Simply: One mind, multiple community-specific persona. Basic actions, rewards and values remain in each community, but the mind transfers information between. The whole system of communities and participation within those is constantly evolving.

It’s the basis of human “dandelionhood” (to twist Cory Doctorow’s analogy slightly) – instead of being focused on one individual self, we can scatter ourselves to the 4 winds. A semi-emergent reality that retains a sense of the mind’s individuality, while gradually weakening the absolute notion of self. In turn weakening most established law and governance, up to and including the notion of sovereignty.

Scary stuff. At least for anyone that understood anything I just wrote.

Postscript: A Chronocentric Tale

Back in 1995, when the internet was new and exciting, a group of academics gathered in Teeside (North East England) to discuss The Governance of Cyberspace. Freshly inspired by the works of people like William Gibson and Howard Rheingold, cyberspace was to become a place where all sorts of utopian and libertarian concepts would challenge 20th century methods of political governance.

That it didn’t happen comes as no surprise to whose that have studied the history of communications technology. In his 1997 history of the telegraph, The Victorian Internet, Tom Standage concluded:

“The hype, scepticism and bewilderment associated with the internet – concerns about new forms of crime, adjustments in social mores, and redefinition of business practices – mirror precisely the hopes, fears and misunderstandings inspired by the telegraph.”

Perhaps those chronocentric revolutionaries of 1995 confused our use of technology to do something, with the deep underlying social drivers behind what we actually do? At a raw emotional level, technology changes very little. The correlation between anglers in the virtual world and people that fish in the physical world, is a simple example. Technically, the processes are very different. Emotionally, they are very similar.

I suspect that any proposal not built around what people already do, is likely to become a utopia, because people evolve slower than we might like to think. Likewise, any proposal which embraces peoples’ emotional desires may become a reality without anyone realizing.

But…

Unfortunately, this article is probably wrong. Almost certainly wrong. So much so, you might ask why I wrote it?

My attempt to write one compact text means that much of the supporting evidence and logic behind statements is missing. But I already know that much is contentious, and within many contentious issues lay further uncertainties. My normal method of analysis and problem solving considers a range of possibilities, and then a range of possibilities that influence the first set of possibilities, and continues this for several iterations. Eventually some themes recur. These themes tend to lead to the solution. In this case, the range of topics is too broad, and the number of iterations of thought too dense. The result is a sequence of ideas that I certainly keep thinking about, and vaguely seem to hang together, but yet remain utterly inconclusive.

In an essay on a similar topic to this article, Nick Szabo concluded:

“The measurement of value is one of the most intractable problems of civilization. Brilliant and highly non-obvious solutions to this problem – from markets to money to the time-wage to cost accounting – have constituted some of the most important steps from animal to civilization.”

Perhaps it is inevitable that I’m still searching for an answer to the question I can’t even define? Maybe I need the solution to reach the answer? Or is this simply too philosophical to ever reach the answer?