Should we plunge headlong into the waves and hope to remain bouyant in the social media storm or head for maven haven on the mountain top?

The web is awash with water metaphors, from streaming video to bittorrent file-sharing protocols. And now we have Google Wave, Mozilla Raindrop and open access to the Twitter firehose.

Similar metaphors are often used to describe human emotional experience and the unconscious realm of memories, dreams and reflections. However, as always when venturing into Neptune’s watery domain, not everything is quite as it appears to be.

The internet is not a series of tubes directly interconnecting nodes to create communications channels but a packet-switching network in which content is divided into little data parcels sent via multiple momentarily-determined routes before being reassembled at its destination.

This engineering not only makes the internet very robust it also makes it neutral, as the data packets are carried without regard for their content. All data is thus equal online, notwithstanding the recent use of deep packet inspection by some Internet Service Providers to discriminate against certain kinds of traffic.

So data does not travel in hyperreality the same way as water flows through the real world. What does this have to do with the way we think about the web?

Being able to distinguish between fantasy and reality is always important, but never more so than when we are considering what we might be inclined to view as a techno-utopian fountain of knowledge.

Computer networks may operate according to the packet-switching protocols that govern them, but the humans that use them continue to behave in ways described by social psychology: in herds, influenced by status and impressed by cultural and political authority.

This is what results in an internet that resembles the image above, rather than the egalitarian interdependency that is often promoted as the brave new world wide web.

A research project plotting the topological structure of the internet in terms of the connections between nodes, while taking into account the roles the connections play, produced some arresting results.

It turns out that a dense core of a few critical highly-connected nodes are surrounded by an outer periphery of many sparsely-connected nodes which are heavily dependent on the core. Between the two lies a mantle of very many peer-connected and largely self-sufficient nodes. If the core is removed from the network, about 30 percent of the outer nodes become completely isolated.

This concentration of traffic in a few dense nodes supports the view that behemoths like Google, Facebook and Twitter excessively influence the web in the same way as key superpowers influence global politics and culture.

It also reflects the inequitable distribution of power, wealth and influence in our world.

An open web is an equal web. We must not mistakenly assume that the neutrality of computer networks is a metaphor, and guarantee, for equality in the human interactions they enable.

And there is no need for another metaphor to make the point that, like water, not all of us have equal access to the resources on which this new world order depends.

The transition between relative and a more or less complete lack of privacy is easy to grasp in this form, and much easier to understand than the labyrinthine multi-page Facebook terms and conditions, or even the manual privacy controls which the majority of its users never alter.

So is this cause for celebration of a more open web or a revealing representation of the cynical interests of an unaccountable corporation in action?

Postscript: some important points on this topic from Jeff Jarvis.

Hosted on whose servers for free?

This is essential reading for anyone involved with the online world. It takes a refreshingly honest look at the first generation of web applications and the way the web has evolved. Check out the talk below then buy the book and delve deeper into his arguments.

Even if you disagree with his perspective, your understanding will be enhanced by having to work out why. And you get treated to a unique musical introduction, too.

How insidious is the slide towards ease of use inside walled gardens such as Facebook and the iPhone? And how many users truly take the red pill by running everything from the command line rather than relying on stable releases of operating systems with convenient graphical user interfaces?

Chris is correct to say that ‘the internet has won as the transport medium for all data’, even though it will take years for this to roll out in practice, but I am not convinced that there must be a one-size fits all ‘universal interface for interacting with the web’.

Some will want hobbled devices that restrict their destinations to a few corporate data silos, but it is still hard to imagine open web search being excluded from that list.

Meanwhile many will continue to relish the relative freedom they currently enjoy to create and contribute on the web by making use of whatever technology they understand and are comfortable with.

I would prefer everyone to own their identity online by publishing and communicating autonomously, making use of open software and systems, instead of depending on proprietary web applications designed to harvest personal data for private profit.

However, there are compromises to be made here as there are in all human endeavour.

I would still rather take part in the Twitter experiment, however profound or trivial it turns out to be, than insist on its open distribution before signing up.

In other words, surely pragmatism is the best solution to such ideological problems?